Professional Documents
Culture Documents
Brunstein J, 2018 (Achievement Motivation)
Brunstein J, 2018 (Achievement Motivation)
6
Joachim C. Brunstein and Heinz Heckhausen
did emerge, reflecting what Max Weber (1904) constructing a tower of building blocks between
had termed the “Protestant work ethic.” This the ages of 2.5 and 3.5 (for illustrations of pride
semantic core covers the life spheres of work, and shame reactions, see the photographs in
learning, and knowledge. It is associated with an Chap. 15, Figs. 15.2 and 15.3). Their first
open societal system characterized by personal responses are facial expressions: smiling when
freedom and in which individual initiative is con- an activity is successful and turning down the
sidered a precondition for personal success in life. corners of the mouth when it is not. Assuming
Family values, tradition, and interpersonal rela- these two forms of expression to reflect the expe-
tions are all subordinate to this value orientation. rience of success and the experience of failure, it
The social recognition of an individual hinges seems that success is experienced earlier (from
primarily on his or her willingness to perform. the 30th month) than failure (from around the
Research on achievement motivation has gen- 36th month). This developmental sequence may
erated an extensive body of findings that can only protect younger children from being discouraged
be outlined in broad brushstrokes in this chapter. by failure before they develop the ability needed
More comprehensive and detailed accounts of the for success. The emotions of joy vs. sadness sig-
development of this research area are available nal that the child is concerned with attaining a
elsewhere (Heckhausen, 1980; Heckhausen, certain action outcome and has started to mea-
Schmalt, & Schneider, 1985; Schultheiss & sure his or her actions against a first, simple stan-
Brunstein, 2005; Weiner, 1985). dard of excellence. However, it is uncertain
whether children at this early stage establish a
link between the outcomes of their action and
6.1 ntogenetic and Evolutionary
O their own abilities. There is clear evidence of
Perspectives such a connection being made just a few months
later, at the (mental) age of about 3–3.5 years,
Achievement-oriented behavior implies commit- when facial expressions of joy and sadness are
ment to standards of excellence and the evalua- supplemented by postural elements that express
tion of performance outcomes. This requires pride and shame. In pride, the upper torso is
cognitive abilities individuals have to acquire stretched and the head thrown back in triumph.
during their development before they can behave Shame reactions are characterized by a lowered
in ways that are motivated by achievement. head and “crestfallen” torso. These expressions
But how can we determine whether standards clearly demonstrate that pride and shame are
of excellence are applied to behavior and self-evaluative emotions. A causal relationship
whether behavioral outcomes are subject to any has been established between the self and the
form of self-evaluation? Studies investigating success or failure of one’s actions. Children now
observable reactions to unambiguous successes see themselves as responsible for the outcomes
and failures provide crucial information here. of their actions. Thus, all of the requirements
From the ontogenetic perspective, such reac- stipulated in the previous definition of
tions can be observed from relatively early in achievement- motivated behavior are now met
life (see Chap. 16 for a detailed discussion of (Heckhausen, 1974):
the development of motivation); this has been
shown in studies about the emotional expressive
reactions of children (Geppert & Heckhausen, Definition
1990; Heckhausen, 1984, 1987; Heckhausen & In achievement-motivated behavior, a stan-
Roelofsen, 1962). dard of excellence is applied to evaluate
one’s actions, and the outcomes of those
Self-evaluative Emotions actions are associated with one’s own
Children begin to display self-evaluative reac- competence.
tions to success and failure on activities such as
6 Achievement Motivation 223
In evolutionary terms, joy and sadness are anchored in biological evolution and observable in
related to expressive behavior observable in pri- early phases of ontogenesis.
mates in the context of affiliation and bonding
behavior. Joy and sadness are expressed in response Summary
to the acquisition or loss of a desired object or upon In achievement-motivated behavior, people eval-
reunification with or separation from a close con- uate their actions and competence against a
specific (Darwin, 1872; Eibl-Eibesfeldt, 1984; standard of excellence. The first signs of achieve-
Frijda, 1986; Kaufmann & Rosenblum, 1969; ment-motivated behavior in human ontogenesis
Plutchic, 1980). Pride and shame, on the other can be observed in the expressive behavior of
hand, are much more closely related to the behavior children (mental age approx. 3.5 years) playing
systems of dominance and submission observable competitive games. The expression of self-eval-
in social primate groups, but also among humans uative emotions, such as pride and shame, indi-
(Eibl-Eibesfeldt, 1984; Lawick-Goodall, 1968; cates that these children evaluate not only the
Riskind, 1984, Weisfeld & Beresford, 1982). outcomes of their actions but also their own
In microgenetic terms, it is noteworthy that 3- competence against a standard of excellence.
through 4-year-old children who win or lose a com-
petitive game first show joy or grief and that these
expressions are then expanded to pride or shame, 6.2 Motive Measurement
respectively, as the child establishes eye contact
with the (adult) opponent (Geppert & Heckhausen, One way of finding out more about people’s
1990). Expressions of pride include spellbound motives is simply to ask. There is no shortage of
fixation on the opponent. Shame prompts an embar- questionnaire measures that present respondents
rassing smile, as though it were important to with statements describing characteristic features
appease the superior opponent and to reestablish of achievement-motivated behavior (e.g., “I often
harmony within the troubled social relationship. set myself challenging goals” or “I like situations
Drawing on these observations on the devel- that tell me how good I am at something”).
opment of children’s expressive behavior, it is Positive responses are taken to indicate that the
possible to speculate on the evolutionary origins respondent has a strong need to achieve.
of achievement motivation and to reason that Responses are structured, with participants indi-
evolution did not need to create a unique affective cating their agreement or disagreement with each
base for this motivation system. Instead, two statement on rating scales.
existing pairs of behavioral and expressive sys-
tems were combined: Direct Measurement
McClelland (1980) called this direct measure-
• Acquisition vs. loss of a treasured object, ment of motives “respondent,” by which he meant
linked to emotions of joy vs. grief that highly standardized stimulus material and
• Dominance vs. submission, linked to pride vs. structured response formats leave very little
shame and associated gestures of superiority scope for participants to provide spontaneous
and appeasement descriptions of their motives. Although this
approach has clear advantages, such as its high
This combination seems to suffice in providing psychometric quality and ease of analysis, it also
an independent affective base for achievement has its disadvantages. Responses may be biased
behavior. The achievement motive is not biologi- by the tendency to present oneself in a socially
cally anchored, but primarily socioculturally medi- desirable light. Moreover, statements such as
ated. It can be subjected to various evaluations and those cited above may assess respondents’ evalu-
take many forms, provided that it is concerned with ations of their own abilities rather than the
a binding standard of excellence. Nevertheless, the motives actually driving their actions. Indeed,
affective bases for these phenomena are deeply respondents may not always be in a position to
224 J.C. Brunstein and H. Heckhausen
reliably identify the motives governing their tests (PSE) that are traditionally also known as
behavior. Given his distrust of the validity of projective methods, in which the respondent
self-
report measures in general, McClelland describes the actions, thoughts, and feelings of
(1980) proposed that “operant” methods be used other people – those portrayed in the pictures.
to measure motives. The concept of “projection” has a checkered
history in psychology (Heckhausen, 1960). Freud
Indirect Measurement used the term to describe a defense mechanism
Operant methods offer a great deal more scope that enables paranoid individuals to attribute the
for differential responses. The test material is feelings and impulses they cannot accept as their
much more open and ambiguous than that used own to other people, thus alleviating the threat
in questionnaire measures. Participants do not posed by these feelings and impulses (e.g.,
react to structured statements, but generate their aggressive and sexual needs) by “projecting”
own responses. As a rule, they are not informed them to the outside. Although empirical evidence
that the assessment aims to investigate their for such processes has not been found (Murstein
motives. The advantages of this kind of indirect & Pryer, 1959), the TAT soon produced very
method of motive assessment are clear: the test interesting findings with respect to motive mea-
situation is more lifelike, specific, and vivid and surement. At a birthday party, Murray (1933)
offers more opportunity to tap an individual’s presented children with pictures of unfamiliar
characteristic ideas and experiences. However, persons both before and after a scary game of
the test situation has to be endowed with stimuli murder in the dark. The children were asked to
that activate the motive under investigation – evaluate the maliciousness of the persons por-
only then can this motive be expressed. trayed. They judged the strangers to be far more
Furthermore, researchers are faced with the task malicious after the scary game than before it.
of filtering out, from the myriad of different Subsequently, Sanford (1937) found that the fre-
responses, those components that provide quency of food-related interpretations of TAT
insights into the nature and strength of the motive pictures increased when respondents were food
aroused. The responses of different individuals deprived. These findings suggested that the TAT
can only be compared and contrasted with refer- could be used to measure the need states acti-
ence to an objective evaluation system. vated at the time of the assessment, such as fear
of strangers or need for food.
• The best-known method that has been devel- The next logical step was to use the TAT to
oped on this basis for the indirect measure- measure enduring motives. Rather than using
ment of motives is the TAT. self-report measures to tap people’s “latent” psy-
chological needs, these needs were to be inferred
from stories generated in response to picture
6.2.1 The Thematic Apperception cues. The pictorial material induces a particular
Test (TAT) motive theme, which then elicits thoughts and
fantasies that may differ markedly from person to
Inspired by the work of Freud (1952), Morgan person. Respondents are instructed to consider a
and Murray (1935; see also Murray, 1938, 1943) picture cue and to write a story explaining how
developed the TAT to identify a person’s needs, the situation has arisen, what the people in the
concerns, and worldviews from the stream of pictures are thinking and feeling, and how the
fantasy-like thoughts this person produces in story will end. The content of the stories obtained
response to ambiguous pictures, usually showing is then evaluated to identify the specific motive
one or more persons. The respondent is instructed activated, e.g., the achievement motive.
to write a short, spontaneous story about each Murray’s (1943) concept of motive (“need”)
picture, giving free rein to her or his imagination. and his taxonomy of motives were presented in
The TAT is one of the families of picture-story Chap. 3. Both played a crucial role in the con-
6 Achievement Motivation 225
struction of the TAT (see also the excursus example (other TAT pictures often used to mea-
below). However, McClelland and colleagues sure motives are reproduced in Smith, 1992).
took the decisive step of applying the method to The “relaxed” and “failure” conditions were
the measurement of motives. originally assumed to be the two poles of a
motive-arousal continuum. By analogy with food
deprivation and the need for sustenance,
6.2.2 TAT Measures McClelland, Clark, Roby, and Atkinson (1949)
of the Achievement Motive interpreted failure to be a form of thwarted satis-
faction (or deprivation) of the need for achieve-
In the late 1940s, McClelland and his associates ment. This somewhat questionable analogy
began investigating whether the TAT could be (“hunger for achievement”) was later abandoned.
used to measure current need states as well as Instead, McClelland et al. (1953) contrasted the
individual differences in the strength of more relaxed with the achievement-oriented condition
enduring motives. They based their work on an and sought to find ways of distinguishing between
experimental paradigm known in the literature as the two, i.e., imagery that occurred more fre-
motive-arousal study (see Schultheiss, 2001a). quently in the achievement-oriented than in the
First, the motive state under investigation is relaxed condition. On this basis, they developed a
induced through experimental manipulation. For coding system to measure the strength of
example, the physiologically regulated need of achievement-related motivational states in TAT
hunger can be activated by temporary food depri- stories.
vation. Atkinson and McClelland (1948) capital-
ized on this mechanism in a study with sailors
stationed at a submarine base. Depending on
their duty schedules, the sailors, who were not Excursus
informed that they were participating in a psy- The Route to the TAT: Controversy Between
chological experiment, had not eaten for 1, 4, or Murray and Allport
16 h prior to the test. Sailors were first shown As a historical aside, it is interesting to
TAT pictures containing food-related cues for note that the development of the TAT tech-
20 s and then given 4 min to write a story about nique sparked a controversy between two
each. As expected, an analysis of story content Harvard professors: Gordon W. Allport and
revealed that longer deprivation times were asso- Henry A. Murray. Whereas Allport (1953)
ciated with a higher frequency of food-related held that non-neurotic individuals experi-
imagery. Relative to participants who had eaten enced no difficulty in reporting their
more recently, sailors who had not eaten for 16 h motives, Murray maintained that motives
made more frequent references to such themes as are not readily accessible to introspection
food shortages and efforts to obtain food and and thus cannot be properly measured by
were more likely to have the figures in their sto- self-report methods. He did not attribute
ries express hunger. this phenomenon so much to repression, as
The questions remained of whether similar to the very early development of motives in
findings would be obtained for “higher” motives, human ontogeny. Whether people are or
such as the need to achieve, and whether the TAT are not conscious of the motives underlying
could be used to measure enduring personality their actions remains a subject of some-
motives as well as current motivation states. In times lively debate (Wilson, 2002). Indeed,
their influential work on the achievement motive, the distinction between “implicit” and
McClelland et al. (1953) addressed each of these “explicit” motives, addressed in more
two issues (see also the study presented below). detail in Chap. 9, has recently revived this
Participants were shown pictures that suggested discussion.
achievement-related themes. Figure 6.1 gives an
226 J.C. Brunstein and H. Heckhausen
Table 6.1 Impact of arousal conditions of various achievement (here, the photograph of a female
strengths on the frequency of achievement-related imageries
Olympic champion). Both arousal factors, pic-
in TAT stories
tures and situational context, increase nAchieve-
Condition N Mean Standard deviation ment scores to approximately the same extent.
Relaxed 39 1.95 4.30
The question of which combination of the two
Neutral 39 7.33 5.49
factors permits the most accurate measurement
Achievement- 39 8.77 5.31
of individual differences in the strength of the
oriented
Success 21 7.92 6.76
achievement motive was finally resolved in favor
Failure 39 10.10 6.17 of weak situational influences (neutral instruc-
Success-failure 39 10.36 5.67 tions making no reference to achievement-related
issues) and pictures fairly high in motive-arousing
Based on McClelland et al. (1953, p. 184)
content (Heckhausen, 1964).
• The context in which the test was embedded
(e.g., the demeanor of the experimenter)
• The instructions given 6.2.3 Success and Failure Motives
• The administration of the test (group vs. one-
to-one setting; written vs. oral responses; time McClelland and Atkinson were aware that their
limitations) thematic coding system for nAchievement con-
• The achievement-related content of the founded two very different achievement-related
pictures tendencies: approaching success and avoiding
• The coding system used to analyze story content failure (see the study on the above). In the coding
system described above, both types of imageries
Three of these features – instructions, admin- are reflected in a single score. Early attempts to
istration, and coding key – are fixed (for a separate success- and failure-related content
summary of the respective procedures, see Smith, categories were less than promising (Scott,
1992), leaving the level of arousal induced by the 1956). Researchers noticed that the behavior of
cover story and the achievement-related content some respondents with moderate to low
of the pictures to be determined. nAchievement scores was characterized by fear
Extensive studies were conducted to gauge the of failure rather than lack of motivation. It was
sensitivity of the nAchievement measure to these practically impossible to predict how these
two aspects (Haber & Alpert, 1958; Jacobs, 1958; respondents would behave in performance situa-
Klinger, 1967). Findings showed that the higher tions (Sorrentino & Short, 1977).
the achievement-related motivational content of
the picture cues, the higher the nAchievement
score. Nevertheless, pictures differing in motiva- Study
tional content were found to discriminate almost The Zeigarnik Effect
equally well between respondents high versus A study conducted by Atkinson (1953)
low in achievement motivation (McClelland on the Zeigarnik effect (the tendency to
et al., 1953, p. 198). Comparable findings were remember interrupted actions more easily
reported for the situational context: the TAT than actions that have been completed)
proved to be sensitive to even subtle differences illustrates early attempts to assess failure
in experimenter behavior (e.g., gestures and motives. Participants were given a test book-
facial expressions; cf. Klinger, 1967). Shantz and let containing 20 tasks to be completed
Latham (2009) took advantage of this observa- under relaxed, neutral, or achievement-
tion in order to demonstrate that the saturation of oriented conditions (in the latter condi-
TAT stories with achievement-oriented words tion, they were told that the items tested
increases substantially if the test instructions are important abilities). The test booklets
accompanied by a stimulus or cue pertaining to
6 Achievement Motivation 229
were constructed such that only half of the motivation decreased steadily from the
items could be completed in the time relaxed, to the neutral, to the achievement-
available. The participants then wrote TAT oriented condition. Atkinson interpreted
stories. At the end of the experiment, they these findings as indicating that individu-
were interviewed informally about the als low in nAchievement behaved as might
tasks, and the number of references to be expected of individuals high in fear of
completed vs. uncompleted tasks was failure, suppressing uncompleted tasks
noted. For the analyses, the sample was from memory, much like an experience of
split at the median of the nAchievement failure. On the same lines, McClelland
distribution, and participants assigned to and Liberman (1949) found that people
high vs. low achievement motivation low in nAchievement take longer to rec-
groups. No differences were found ognize words flashed on a screen when
between the two groups in terms of their these words are associated with failure.
ability to recall completed tasks. The They interpreted this phenomenon as
results for uncompleted tasks were quite indicative of “perceptual defense” against
different, however, as shown in Fig. 6.3. inimical stimuli.
Participants high in achievement motiva-
tion recalled more uncompleted tasks, as
Assessment of Failure Motives
Moulton (1958) also endeavored to identify fear
Mean Percentage of Incompletions Recalled
70 High nAch
of failure as a motive in its own right in TAT sto-
ries, but his efforts made little impact on research.
Low nAch
60 Instead, researchers in the USA employed anxi-
ety questionnaires such as the “Test Anxiety
50 Questionnaire” (TAQ; Mandler & Sarason, 1952)
to assess fear of failure. Atkinson (1964, 1987;
40 Atkinson & Litwin, 1960) assumed fear of failure
to be accessible to introspection and thus measur-
30 able by questionnaire techniques. In the risk-
taking model, he defined fear of failure as a
motive that counteracts the success motive.
Relaxed Neutral Achievement-
Oriented Achievement anxiety questionnaires do not
assess motives, however. Instead, they tap behav-
Fig. 6.3 Mean percentage recall of uncompleted ioral symptoms that may be experienced in overly
tasks by respondents high and low in achievement
motivation (nAchievement) under three arousal demanding achievement situations (e.g., a diffi-
conditions (Based on McClelland et al., 1953, cult exam). Responses to achievement anxiety
p. 266) questionnaire items thus correlate with self-
perceptions of insufficient ability (Nicholls,
predicted by Zeigarnik (1927), when they 1984a, 1984b). Findings soon showed that, apart
had been exposed to achievement-oriented from feeling more nervous when faced with per-
conditions. Participants with low achieve- formance demands, individuals high in achieve-
ment scores recalled far fewer uncom- ment anxiety often doubt their abilities to cope
pleted tasks under these conditions. In with these demands (Liebert & Morris, 1967;
fact, the percentage of uncompleted tasks Wine, 1971). In the same vein, they rate the sub-
recalled by participants low in achievement jective difficulty of tasks to be higher than do less
anxious individuals (Nicholls, 1984a, 1984b).
230 J.C. Brunstein and H. Heckhausen
6.2.4 T
AT Measures of Hope coding system was developed on the basis of
for Success and Fear of Failure TAT stories generated under conditions of neu-
tral instructions (no reference being made to
German researchers took a different approach. achievement) and picture cues high in motive-
Heckhausen (1963; see also Meyer et al., 1965) arousing content. Three pictures unmistakably
developed a TAT technique to measure both depicted hope for success (e.g., a student sitting
“hope for success” (HS) and “fear of failure” at a desk and smiling happily), three others for
(FF) using the same set of picture stories. The fear of failure (e.g., a student being watched by a
6 Achievement Motivation 231
teacher as he writes something on the board). Table 6.2 Correlations between nAchievement
Behavior in a level of aspiration experiment was (McClelland et al., 1953) and the motive variables of
Heckhausen’s TAT procedure
used as the criterion for identifying success-
related or failure-related statements, allowing Hope for Fear of Net Aggregate
success failure hope motivation
the coding system to be fine-tuned relative to a
Teacher 0.73** 0.15 0.32* 0.63**
validity criterion. Specifically, the TAT stories education
produced by respondents who set goals that were students
slightly higher than their previous performance (N = 71)
level (indicative of success motivation) were University 0.60** 0.21 0.27* 0.62**
students
compared with the stories generated by respon-
(N = 77)
dents who set excessively high or low goals
Based on Heckhausen (1963, p. 74)
(both indicative of failure motivation). Content *p < 0.01, **p < 0.001
categories that were found to distinguish between
these two groups were then used to construct a
coding key for HS and FF (Heckhausen, 1963). anxiety (Fisch & Schmalt, 1970). Table 6.2
The following overview documents the individual reports the correlations between nAchievement,
content categories (examples are given in as defined by McClelland et al. (1953), and the
parentheses): two variables of Heckhausen’s TAT instrument in
Only one point is allocated for each content two samples of college students. Whereas
category present in a story. Total HS and FF nAchievement shows strong correlations with
scores are computed by aggregating the points HS, it does not correlate with FF, confirming that
scored across the entire set of six stories. The dif- fear of failure is indeed a motive in its own right.
ference between the two scores is termed “net
hope” (NH = HS – FF ); their sum is termed
“aggregate motivation” (AM = HS + FF ). 6.2.5 Psychometric Properties
As mentioned above, the coding system was vali- of the TAT
dated using an external criterion, namely, level of
aspiration: Classical test theory (Cronbach, 1990) holds that
the quality of a test is a function of the objectivity
• Success-motivated participants (HS) favored of test administration and coding procedures and
goals that slightly exceeded their previous the reliability of the scores determined. Both
level of performance. objectivity and reliability are considered prereq-
• Failure-motivated participants (FF ), in con- uisites for the validity of test scores.
trast, fell into two subgroups: Objectivity. Because TAT instruments are sen-
–– Some opted for excessively low goals and sitive to situational influences (Lundy, 1988), the
others set themselves unrealistically high objectivity of test administration is critical. Strict
targets. adherence to standardized administration proce-
dures is thus imperative (Smith, 1992).
Correlational analyses show that the two
motive tendencies, HS and FF, are mutually • The objectivity of the TAT coding procedure,
independent, indicating that there must be people measured in terms of the agreement between
who both strive for success and seek to avoid fail- independent raters, has proved to be satisfac-
ure. Neither of the two TAT variables correlate tory to high.
significantly with questionnaire measures of
achievement motivation (Halisch & Heckhausen, Interrater agreement on content categories is
1988). There is only a slight overlap between FF at least 85% because only the data of raters who
and TAQ scores, indicating that fear as measured satisfy this criterion are included in empirical
by the TAT is conceptually different from test analyses. Interrater reliability coefficients range
232 J.C. Brunstein and H. Heckhausen
between 0.80 and 0.95. Coefficients of this mag- 1958; Heckhausen, 1963; Sader & Specht,
nitude can only be achieved when raters are prop- 1967), ranging between 0.40 and 0.60 over a
erly trained; training material and expert ratings retest interval of 3–5 weeks. Correlations in the
are available for this purpose (for nAchievement, same range are found after a 1-year interval
Smith & Feld, 1958; for HS and FF, Heckhausen, (Lundy, 1985). It should be noted, however, that
1963). From early on, computer programs to ana- it is impossible to reproduce the original condi-
lyze the content of TAT stories have also been tions in a TAT retest. Respondents are often able
developed (for nAchievement, Stone, Dumphy, to remember the pictures shown and the stories
Smith, & Ogilvie, 1966; for HS and FF, they wrote at the first administration and make a
Seidenstücker & Seidenstücker, 1974). Despite conscious decision to draft very different stories
their parsimony and objectivity, computer-based at retest. This phenomenon was illustrated for
measurements of motives have initially gained the power motive in a study conducted by Winter
little currency in research practice. There are now and Stewart (1977). At retest, which was taken
many indications that this situation might change 1 week after the first TAT administration, partici-
in the foreseeable future. It is becoming increas- pants were given one of the following
ingly common that TAT stories are typed with a instructions:
keyboard instead of being written by hand
(Blankenship & Zoota, 1998; Schultheiss et al., • To think back to the previous week and write
2008). At the same time, key words that clearly stories as similar as possible to their original
represent a particular motive (e.g., “successful” ones
or “win” for HE) are being collected in motive • Not to worry about whether or not their stories
dictionaries (Hogenraad, 2005; Schultheiss, were similar to their original ones
2013). Using suitable text analysis software, such • To write stories as different as possible from
as Pennebaker and Francis’s (1999) “Linguistic their original ones
Inquiry and Word Count Program” (LWC), The test-retest correlations for each instruc-
allows for the identification of how often “marker tion were 0.61, 0.58, and 0.27, respectively.
words” relevant to a motive (Schultheiss, 2013)
or phrases associated with such words
(Blankenship, 2010) appear in computer-based
texts. Moreover, such software can uncover if Excursus
formal or stylistic text features (e.g., word Fear of Success
lengths, tense) are in a regular relationship with The material used by McClelland et al.
motive-relevant thoughts (Pennebaker & King, (1953) and the picture cues employed in
1999). Studies on the convergence of motive Heckhausen’s TAT instrument (1963) were
scores based on marker words and comparable tailored exclusively to men. Women were
scores determined with traditional coding meth- not featured in the pictures. In fact, the
ods have yielded encouraging results. At the very results of early studies, particularly in the
least, they suggest that the time-consuming pro- USA, indicated that women’s achievement
cess of coding the content of TAT stories can motives were not in line with the tradi-
fairly soon be done by automatic and easily tional “feminine” role orientation, making
reproducible evaluation systems (for an example behavioral effects difficult to predict.
from the field of applied psychology, see Shantz Many studies did not even include women
& Latham, 2009). as part of their sample which seems par-
Reliability. Reliability is primarily concerned ticularly strange from today’s perspective.
with the stability of test scores over repeated Martina Horner (1974a, 1974b), a Harvard
administrations. When compared with question- professor who became at age 32 the young-
naire measures, the test-retest correlations of est president in the history of Radcliffe
TAT techniques are modest (Haber & Alpert,
6 Achievement Motivation 233
which the authors demonstrated how a modified facets. These include adaptations of the TAT
Rasch model can be used to find suitable, i.e., method as well as objective tests, most of them
particularly stimulating, pictures and to combine questionnaire measures. We do not seek to pro-
those pictures to a meaningful set of stimuli. vide a comprehensive overview of these instru-
Clearly, the use of models of stochastic test the- ments in the present chapter (cf. Fineman, 1977;
ory is not only relevant for measuring motives. It Heckhausen et al., 1985; Rheinberg, 2004;
is about much more than merely the psychomet- Stiensmeier-Pelster & Rheinberg, 2003), but out-
ric analysis of the properties of the TAT. Such line a selection of the most established.
models do in fact also test theoretical assump-
tions about which processes of “apperception” Adaptations of the TAT
express a particular motive in the instrument The French Test of Insight (FTI) developed by
(TAT). Proceeding on this idea, Tuerlinckx, and named for French (1955, 1958a) uses the
Boeck, and Lens (2002) tried to separate relevant beginning of stories, rather than pictures, to acti-
and nonrelevant achievement fantasies in the vate imagery relevant to the motive under investi-
diagnosis of the achievement motive (TAT). For gation (“Don is always trying something
this purpose, they specified a stochastic model new…”). The manual used to categorize the
that integrated aspects of dynamic action theory imagery generated is equivalent to the coding
by Atkinson et al. (1977), but their results were system for nAchievement.
underwhelming. Using the specified (“dropout”)
model, Atkinson’s idea that expressing an • The FTI is employed when the investigator
achievement-oriented fantasy creates a consum- deems it appropriate for pictorial cues to be
matory effect that in turn weakens the related replaced by verbal ones, e.g., when comparing
motivational tendency could not be confirmed. In individuals from different cultures. Hofer and
light of this model, a large part of the fantasies Chasiotis (2004; see also Hofer, 2010) were
had an erratic effect and did not allow for any the first to systematically and successfully
conclusions about the strength of the ostensibly design the TAT in a culture-fair way.
relevant achievement motive. Using a different
stochastic model (“Thurstonian item-response Birney, Burdick, and Teevan (1969) developed
theory”), Lang (2014) was the first to show that another TAT-like technique specifically to assess
the TAT provides a reliable and construct-valid fear of failure. In contrast to Atkinson, these
measurement of the motives for achievement, researchers proceeded on the assumption that
power, and affiliation, provided that the measure- fear of failure is not openly admitted, but becomes
ment instrument consists of at least six pictures. manifest indirectly, in perceptions of a hostile
Both for real and for simulated data, Lang found and self-threatening environment. The variable
the best fit for a model that included dynamic assessed by this technique is labeled hostile press
processes as they had been postulated by Atkinson (HP) and overlaps to some extent with high FF
et al. (1977; competition between various motives and low nAchievement scores (Birney et al.,
for being expressed in fantasies; consummatory 1969; Heckhausen, 1968). This projective mea-
effect of the expressed motive on the strength of sure of fear of failure is used as a counterpart to
the respective motivational tendency). nAchievement, particularly in studies conducted
with US samples (e.g., Thrash & Elliot, 2002).
Schultheiss (2001b) translated Heckhausen’s
6.2.7 Other Techniques scoring key into English and thus made it acces-
for Measuring Achievement- sible to a wider segment of the scientific commu-
Related Motives nity. Pang (2006, 2010) specifically stimulated
HS and FF with positive and negative perfor-
Various other techniques have now been devel- mance feedback in order to optimize
oped to measure the achievement motive and its Heckhausen’s coding system. In turn, this engen-
236 J.C. Brunstein and H. Heckhausen
dered a review of the Heckhausen key that Pang • HS: The conceptual equivalent of the TAT
validated with relevant external criteria (e.g., per- success motive
taining to risk-taking behavior). Numerous varia- • FF-1: Active failure avoidance; also includes
tions on the TAT picture cues and coding system items reflecting a low self-concept of ability
have been proposed. Winter (1991a, 1991b) • FF-2: Fear of failure and its potential social
developed a manual that allows achievement, consequences
power, and affiliation motives to be inferred from
speeches, school books, and other documents, as The two aspects of fear of failure (active vs.
well as from TAT stories. It does not permit hope- passive avoidance) are thus also clearly apparent
and fear-related content categories to be assessed in the AM Grid. Schmalt, Sokolowski, and
separately, however. Langens (2000; see also Langens & Schmalt,
Kuhl and Scheffer (1999; see also Scheffer, 2008; Sokolowski, Schmalt, Langens, & Puca,
2003; Baumann, Kazen, & Kuhl, 2010) modified 2000) have expanded the Grid technique to cover
the TAT technique for the assessment of basic the power and affiliation motives as well. This
motives (including achievement) in the Operant Multi-Motive Grid (MMG; Chap. 8) measures
Motive Test (OMT). First, they used highly hope and fear components separately for each of
ambiguous pictures. Second, participants do not the three motives.
have to write complete stories. Instead, they are
asked to concisely write down their thoughts on
four central questions. This reduces the time Excursus
required for running and coding the test. The key Using the Rasch Model to Test the TAT
for the content analysis is driven by theory and Measures
involves concepts (for the achievement motive: Kuhl (1977, 1978a) tested whether the
flow, internal standards of excellence, coping TAT measures HS and FF can be scaled
with failure, pressure, and failure) that are taken according to the Rasch model. He analyzed
from theories of motivational self-regulation 6,204 TAT protocols produced by 1,034
(Kuhl, 2000, 2001). More detailed information respondents of different ages, genders, and
on this method can be found in Chap. 13. educational levels. The consistency of both
The achievement motive grid. Schmalt (1973, measures – or, more precisely, their content
1976a, 1976b, 1999) took a new approach to categories – was tested with respect to the
measuring the achievement motive. His theoretical construct (i.e., the Rasch crite-
Achievement Motive Grid (AM Grid) is a semi- rion of “specific objectivity” was applied).
projective technique that combines the advan- The first question to be addressed was
tages of the TAT method (picture cues) with the whether the frequency of content categories
merits of questionnaire measures (objective and relating to a specific motive (HS or FF) var-
parsimonious analysis). Respondents are pre- ied proportionally across each pair of pic-
sented with 18 pictures from different areas of ture stimuli. Given this to be true, it should
activity (sports, school, etc.). The same 18 state- be possible to map all individual content
ments – borrowed from the content categories of categories to a regression line with a slope of
Heckhausen’s TAT method – are listed below one when two pictures are compared. As
each picture. Respondents are asked to check Fig. 6.4a shows for FF, the content catego-
those statements that, in their opinion, apply to ries F and R deviate markedly from the
the person shown in the picture (e.g., a student regression line. Relative to the other catego-
doing his homework: “He feels proud; doesn’t ries, F and R were scored disproportionately
think he’s capable; is afraid of doing something more often in stories about picture D than in
wrong”). Three different motive tendencies are stories about picture B. Assuming that a
distinguished: motive can be expressed in terms of differ-
6 Achievement Motivation 237
ent content categories depending on the pic- Af, whereas participants high in FF scored
ture, this kind of interaction between the disproportionately more often in the catego-
pictures and the response parameters does not ries F and R. Thus, the results did not substan-
necessarily preclude the specific objectivity of tiate the assumption that FF is a consistent
a person or an item parameter. For this reason, disposition across situations and reactions.
Kuhl did not view test items as pictures isolated Further analysis revealed that it was not the
from responses, but conceived of the two as pictures, but the content categories that caused
fixed picture-response combinations. this inconsistency. Two classes of fear-related
Kuhl subjected the parameters calculated imagery could be distinguished:
to internal and external model tests. For HS,
the parameters of picture-response combina- • A tendency toward expectancy and action-
tions proved consistent across various sub- related failure avoidance (Nf, If, Af)
groups of participants. This finding held • A tendency to become preoccupied with fail-
whether the groups were divided on the basis ure (F) and its affective consequences (R)
of high vs. low HS scores (internal model test)
or high vs. low FF scores (external model Fear of failure (FF), as defined by
test). In other words, the HS content catego- Heckhausen, thus seems to incorporate active
ries yield equivalent and – from the perspec- (or “action-oriented”) as well as passive (or
tive of construct theory – consistent indexes “state-oriented”) approaches to coping with
for one and the same personality trait. A dif- failure (Kuhl, 1983; Schultheiss & Brunstein,
ferent pattern of results emerged for FF, how- 2005). Factor analytic studies yielded very
ever. The internal model test showed that the similar results. Whereas HS proved to be uni-
FF content categories were not unidimen- dimensional, two independent factors emerged
sional. Figure 6.4b illustrates these findings for FF: the need to avoid failure, on the one
for picture D in the Heckhausen TAT. hand, and negative affective states occurring
Participants low in FF scored disproportion- in response to failure, on the other (Sader &
ately more often in the categories If, Nf, and Keil, 1968).
Fig. 6.4 Response parameters for the “fear of failure” The deviation of the response parameters from the
(FF) content categories for (a) two TAT pictures regression line does not challenge the specific objec-
(Pictures B and D from Heckhausen, 1963) and (b) tivity of FF in case (a), but it does in case (b) (Based
two groups of respondents with high vs. low FF scores. on Kuhl, 1978a, pp. 40, 44)
238 J.C. Brunstein and H. Heckhausen
It has long been acknowledged that motives who describe themselves as being afraid of
measured by questionnaire techniques barely failure.
correlate with motives assessed using the TAT Covington and Omelich (1979), Kukla (1972b),
method (deCharms, Morrison, Reitman, & Meyer (1984a, b, 1987), and Nicholls (1984a) had
McClelland, 1955). This finding has been cor- drawn attention to this point and concluded that
roborated by numerous researchers (Halisch, perceived competence (or ability) is a major com-
1986; Halisch & Heckhausen, 1988; Niitamo, ponent of achievement motivation. However,
1999; Spangler, 1992; Schultheiss & Brunstein, inspection of the correlations for the TAT measures
2001). Table 6.3 illustrates the typical pattern of of HS and FF shows that neither is related to the
results with a dataset that Brunstein and Schmitt self-concept of ability, challenging the assumption
(2003) collected from university students enrolled that achievement-related motives can be equated
in various majors (psychology students were with ability-related self-views. These findings went
excluded). The correlations between hope for unheeded for many years in empirical research.
success as measured by projective (TAT), semi- Instead, the same labels (hope for success, fear of
projective (Grid), and questionnaire (AMS) failure) were used for measures of achievement
methods, respectively, all approach zero. There motivation that have very little to do with one
are weak, but significant, correlations between another on the empirical level. McClelland and his
fear of failure as measured by the TAT and by associates (1989; see also Weinberger &
questionnaire measures. Only the correlations McClelland, 1990) finally spelled out the dangers
between the two questionnaire measures (AMS of using the same terms to describe different con-
and MARPS) are really substantial in size. cepts and proposed that a clear distinction be drawn
Notably, there is a marked negative correlation between motives measured using indirect (TAT)
between HS and FF in the self-report measure methods and motives assessed with direct (ques-
(AMS), but not in the TAT. The correlations tionnaire) methods. Their reasoning and findings
reported in Table 6.3 support the idea that there is are presented in Chap. 9.
a considerable overlap between respondents’ sub- It remains problematic, however, that even
jective assessments of their cognitive capacities across indirect tests the convergent validity for the
(measured with Meyer’s, 1972, self-concept of same motive tends to be fairly low. Table 6.3 shows
ability questionnaire) and self-attributed achieve- this for measurements of the achievement motive
ment orientation. People who describe them- with the TAT and the Motive Grid. The problem
selves as success-oriented achievers rate their persists even if the degree of similarity between
intellectual abilities more favorably than people indirect instruments is increased. Schüler,
Table 6.3 Correlations between different methods of measuring individual differences in achievement motivation
1. 2. 3. 4. 5. 6. 7. 8.
1. HS: TAT –
2. FF: TAT 0.07 –
3. HS: MMG 0.10 −0.03 –
4. FF: MMG −0.07 0.02 −0.15* –
5. HS: AMS −0.01 −0.19** 0.04 −0.07 –
6. FF: AMS −0.05 0.17** −0.01 0.08 −0.57** –
7. MARPS −0.09 −0.19** 0.00 −0.08 0.57** −0.46** –
8. Subjective capacity 0.05 −0.03 0.05 −0.12 0.41** −0.55** 0.35** –
Data from Brunstein and Schmitt (2003)
*p < 0.05; **p < 0.01
N = 220 students with different majors, HS hope for success, FF fear of failure, TAT thematic apperception test, MMG
Multi-Motive Grid, AMS Achievement Motives Scale, MARPS Mehrabian Achievement Risk Preference Scale, subjec-
tive capacity self-concept of ability
240 J.C. Brunstein and H. Heckhausen
Brandstätter, Wegner, and Baumann (2015) used Kukla (1972a, 1972b) and Nicholls (1984a), in
several instruments, including the TAT (coded with contrast, assumed the achievement motive to
Winter’s, 1991a, scoring key), OMT, and the have an impact on expectancies. Achievement-
Motive Grid, in a student sample and could not find motivated individuals are more confident in their
a significant correlation of the three instrument- abilities, expect to be able to cope with difficult
specific variables for “one” achievement motive. It tasks, and are thus more motivated to tackle this
seems doubtful that these instruments assess the kind of tasks.
same construct. At least, the methodological vari- Although the issue of affective (or incentive-
ance appears to be substantial. based) vs. cognitive (or expectancy-based) inter-
pretations of the success motive is at the very
core of achievement motivation theory (Sect.
6.2.8 Anatomy, Mechanisms, 6.4.2), the debate is still limited to a few insiders.
and Measurement In view of the disparities between motive vari-
of the Achievement Motive ables tapped by TAT vs. questionnaire methods,
it might be speculated that HS as measured by the
According to Atkinson’s (1957, 1964) formula TAT has an impact on the incentive of success,
(Chaps. 2 and 5), motivational tendencies result whereas HS as measured by questionnaires has
from the interplay of three variables: incentive an impact on the anticipation of success. This
(I), probability of success (P), and motive interpretation would converge with the finding
strength (M). For reasons of simplicity, we focus that the scores on achievement motive scales are
here on the tendency (T) to be successful (s), related to the self-concept of ability, whereas
which Atkinson defined as follows: TAT scores are not.
Ts = Ms × Ps × Is Ultimately, however, neither the TAT nor
questionnaire methods distinguish carefully
In this formula, the success motive functions between incentive-related and expectancy-related
as a weighting factor that is combined multiplica- information. HS as measured by the TAT – origi-
tively with incentive and expectancy. The ques- nally defined by Heckhausen (1963) as an
tion arises of which of the two situational factors, “expectancy-related attitude” – covers both
incentive or expectancy, is weighted by the incentives (e.g., positive affect after success) and
success motive (or whether Ms applies to the expectancies (e.g., certainty of success). Much
product of both factors). the same can be said for the questionnaires men-
On the basis of the formula itself, it is impos- tioned above. In most cases, the statements to be
sible to give a formal or mathematical answer to rated relate to both incentives and expectancies.
this question. The fact that Atkinson combined People who state that they “like working on dif-
the two situational variables in a subtractive rela- ficult tasks” indicate not only that they find diffi-
tionship (Is = 1– Ps) complicates the matter fur- cult tasks attractive but also that they are confident
ther. Approaching the problem on the conceptual of being able to master them.
level, different achievement motivation research- A more accurate examination of the mecha-
ers have provided very different responses. nisms of achievement motives would require the
McClelland, Atkinson, and Heckhausen advo- disentangling of incentive-related and
cated the view that a strong success motive expectancy- related components. Global mea-
increases the affective value of success. The sures of achievement motivation are unsuitable
product term Ms × Is can thus be interpreted as for this purpose. Heckhausen (1977a, 1977b,
the valence of a success. The amount of pride felt 1986) thus proposed that the summary concept of
by someone who has mastered a challenging task “the” achievement motive should be abandoned
can be expected to increase as a function of the altogether, and instead split into a number of
strength of her or his success motive (Sect. 6.4.1). constituent parts connected with situational vari-
6 Achievement Motivation 241
6.3.1 T
he Achievement Motive tion of task-specific skills. Lowell (1952) was the
and Individual Performance first to take this approach. He presented partici-
pants with simple addition problems (“Düker
The first studies conducted to validate the tasks”) and scrambled-word tasks (anagrams)
nAchievement measure investigated the relations and assessed performance at 2-min intervals.
between the strength of the achievement motive Right from the beginning and throughout the
and numerous behavioral criteria, without paying experiment, participants high in achievement
particular attention to situational incentives. motivation outperformed those low in achieve-
Behavior was seen as a direct function of the ment motivation on the addition problems
strength of the motive and interindividual varia- (Fig. 6.5a). The same was true for the anagram
tions thereof. Meta-analyses have since shown tasks, but motive-dependent performance differ-
that such correlations rarely exceed the level of ences were not substantial until the middle and
.30 (Collins, Hanges, & Locke, 2004; Spangler, the last third of the test phase (Fig. 6.5b). In con-
1992). Because these findings have been docu- trast to the (overlearned) addition problems, per-
mented elsewhere (Atkinson, 1964; Atkinson & formance on the anagrams required the
Feather, 1966; Heckhausen et al., 1985), we limit participants to identify a learning algorithm.
our account to a few examples. Relative to less-motivated participants, highly
One of the fundamental characteristics attrib- motivated participants needed less practice to
uted to every motive is that it energizes instru- learn this algorithm and thereby optimize their
mental behavior; a second assumption is that test results. Lowell’s findings for simple arithme-
behavior is more easily learned if it serves to sat- tic problems were replicated in further studies
isfy a motive (cf. McClelland, 1980). It thus (Biernat, 1989; Wendt, 1955), showing that indi-
seemed reasonable to examine the predictive viduals high in the achievement motive tend to
validity of the achievement motive with experi- perform better on tasks requiring high levels of
mental tasks requiring high levels of effort and mental concentration than do individuals with a
mental concentration. As Thurstone had noted relatively weak achievement motive. This differ-
early (see also Thomas, 1983), this is generally ence is even stronger when respondents are pro-
the case when in speeded tests respondents are vided with feedback informing them about how
required to execute large numbers of tasks as their achievement develops across task trials
quickly as possible. Other studies tested whether (Brunstein & Hoyer, 2002; Brunstein & Maier,
the achievement motive is related to the acquisi- 2005). Lowell’s findings on the acquisition of
32
14 High nAch
Low nAch
Addition Problems
13
30
High nAch 12
Anagrams
28 Low nAch 11
10
26
9
24
0
Two-Minute Periods Two-Minute Periods
(a) (b)
Fig. 6.5 Mean performance of individuals high and low in the achievement motive (nAchievement) on (a) simple addi-
tion problems and (b) scrambled-word tasks (anagrams) over 2-min periods (Based on Lowell, 1952, pp. 36, 38)
6 Achievement Motivation 243
The creativity of research inspired by David graphic groups have also been established. This
C. McClelland’s efforts to investigate the effects strand of research took the bold, but plausible,
of the achievement motive on performance out- approach of using sociological, historical, and
comes in real-life settings remains unparalleled. economic categories as indicators of achievement-
Studies carried out in India (Singh, 1979) and related valuations and behaviors. It was initiated
Columbia (Rogers & Svenning, 1969), for exam- by McClelland (1961), based on Max Weber’s
ple, showed that farmers high in the achievement (1904) hypothesis of an intrinsic relationship
motive implemented more innovative farming between the Protestant work ethic and the spirit
methods and produced better yields than their of capitalism. According to Weber, the industrial
less achievement-motivated counterparts. In a revolution was sparked by the activistic work
longitudinal study, McClelland and Franz (1992) ethic of post-reformation religious movements
found that the strength of the achievement (e.g., Calvinist teachings of predestination).
motive, measured at age 31, predicted income
and occupational success at age 41. There is no 6.3.2.1 The Achievement Motive
doubt that findings of this kind are impressive and Economic Growth
and attest to the criterion validity of the nAchieve- McClelland (1961) reasoned that children
ment measure. However, it remained unclear brought up in the context of the Protestant ethic
which mediating processes (more learning, more are raised to be independent and accountable.
time devoted to work, higher curiosity levels, This kind of upbringing fosters the development
higher levels of aspiration, etc.) accounted for the of a high achievement motive, which in turn
relationships observed. stimulates entrepreneurial activity, leading to
As mentioned earlier, most early studies seek- accelerated economic growth, consistent rein-
ing to validate nAchievement paid very little vestment of capital gains, and an open-minded
attention to situational conditions. A study by approach to technological progress. A compari-
French (1958b) in which the fit between behav- son of Protestant and Catholic countries around
ioral motives and situational incentives was var- 1950 revealed the former to be wielding greater
ied systematically is an exception to this rule (see economic power. McClelland used the per capita
excursus). consumption of electricity as an index of eco-
Findings similar to those reported by French nomic power, taking into account national differ-
have been documented by McKeachie (1961), in ences in natural resources.
an analysis of college students’ performance, and How, though, is it possible to test the effects
by Andrews (1967), in an analysis of career of national differences in collective motives
advancement in companies. Here again, corre- on economic growth? And how can collec-
spondence between incentives and motives tive motives be measured at all? McClelland
proved to be the decisive factor in educational obtained a national motive index by analyz-
and occupational success. ing the content of stories in third-grade read-
ers using the nAchievement coding system. He
felt that few sources would reflect the motivational
6.3.2 T
he Achievement Motive, “Zeitgeist” in countries with compulsory school-
Historical and Economic ing as well as these early readers. In a prelimi-
Change, and Regional nary analysis of a relatively small group of
Disparities countries, the national nAchievement indexes
Between Educational for the year 1925 were correlated with the per
Achievements capita consumption of electricity between 1925
and 1950. Yielding a coefficient of r = 0.53,
Not only have differences in motive strength this correlation turned out to be sensationally
been related to individual differences in behavior, high. In a second analysis of a larger group of
differences in the motives of various demo- countries (Table 6.5), McClelland correlated the
6 Achievement Motivation 245
national nAchievement index with the discrep- motive seems to be associated with dispropor-
ancy between observed and expected increases tionately high economic growth, while low
in electricity consumption between 1952 and motive strength predicts below average growth.
1958. Differences in the countries’ baseline lev- Follow-up studies generally corroborated this
els of economic growth caused by disparities in finding, but data have shown that the relation-
the availability of natural resources and the level ship between nAchievement and the level of
of industrialization were statistically controlled. electricity consumption is no longer as strong
The correlation between the motive index for as it once was (Beit-Hallahmi, 1980; Frey,
the year 1950 and the increase or decrease in 1984; McClelland, 1976, 1984a, 1984b; Orpen,
electricity consumption between 1952 and 1958 1983). It seems that the validity of electricity
was r = 0.43. Thus, a high national achievement
Table 6.5 National motive index (nAchievement) for the year 1950 and rate of increase in electricity consumption
(deviation from the expected growth rate in standard deviations) between 1952 and 1958
National motive Higher consumption National motive Lower consumption
index (1950) than expected index (1950) than expected
Countries high in Turkey 3.62 +1.38
nAchievement
India 2.71 +1.12
Australia 2.39 +0.42
Israel 2.33 +1.18
Spain 2.33 +0.01
Pakistan 2.29 +2.75
Greece 2.29 +1.18 Argentina 3.38 –0.56
Canada 2.29 +0.06 Lebanon 2.71 –0.67
Bulgaria 2.24 +1.37 France 2.38 –0.24
USA 2.24 +0.47 South Africa 2.33 –0.06
West Germany 2.14 +0.53 Ireland 2.29 –0.41
USSR 2.10 +1.62 Tunisia 2.14 –1.87
Portugal 2.10 +0.76 Syria 2.10 –0.25
Countries low in Iraq 1.95 +0.29 New Zealand 2.05 –0.29
nAchievement
Austria 1.86 +0.38 Uruguay 1.86 –0.75
England 1.67 +0.17 Hungary 1.81 –0.62
Mexico 1.57 +0.12 Norway 1.71 –0.77
Poland 0.86 +1.26 Sweden 1.62 –0.64
Finland 1.52 –0.08
Netherlands 1.48 –0.15
Italy 1.33 –0.57
Japan 1.29 –0.04
Switzerland 1.20 –1.92
Chile 1.19 –1.81
Denmark 1.05 –0.89
Algeria 0.57 –0.83
Belgium 0.43 –1.65
Based on McClelland (1961, p. 100)
246 J.C. Brunstein and H. Heckhausen
c onsumption as an indicator of economic devel- and Moeller (1962) compared the number of pat-
opment has decreased somewhat. ents granted in the USA between 1810 and 1950
Content analysis of written documents makes with the development of the national motive
it possible to establish motive indicators for ear- index (nAchievement as derived from reading
lier historical periods as well. Samples of datable books). Again, changes in nAchievement her-
literary texts were analyzed to examine the cur- alded corresponding changes in the patent index.
rency of achievement-related themes in earlier
cultures. These texts included Ancient Greek epi- 6.3.2.2 Achievement Motive
grams, poetry, and funeral orations dating from and Regional Disparities
900 to 100 BC; Spanish novels, poems, and leg- in Educational Achievements
ends from 1,200 to 1,730; and English dramas, In many cases, national motive indices were
travelogues, and ballads from 1,400 to 1,830. The assessed using children’s and youth literature.
respective economic indicators were the extent of Therefore, educational achievements – both of
Greek olive oil exports, as shown on archeological individuals and of certain populations – should
maps; the tonnage of ships per year departing from be a central criterion for the validation of such
Spain for the New World; and annual imports of indices. In times of comparative studies on edu-
coal to Greater London. In all cases, periods of cational achievements, it has become possible to
economic prosperity were preceded by increases empirically test such assumptions.
in the nAchievement index, and periods of eco- Using Winter’s (1991a) coding key, Engeser,
nomic decline by decreases. Figure 6.6 shows Rheinberg, and Möller (2009) assessed the
another example of this relationship: deCharms achievement-related content of textbooks for
German language and mathematics classes in
grades 2 and 9 in the German states of Baden-
Wurttemberg and Bremen. These two states were
chosen because they differ substantially with
70 14
regard to indicators for economic (e.g., employ-
Patent Index
ment rate) and educational performance (results
Mean Frequency of Achievement Imagery
the different regions, which were chosen using in order to validate the achievement motive
best-selling book lists and sales. Across states assessed with the TAT. In fact, many studies have
they found substantial positive correlations chosen this approach since the beginning of
between the level of educational achievement research on achievement motivation (see Hall,
(measured with comparative studies across Stanton, & Schultheiss, 2010). The achievement
states) and the frequency with which achievement- motive has been associated with measurements
oriented content was featured in regionally pre- of muscle activity (Mücher & Heckhausen, 1962)
ferred children’s literature. and the concentration of uric acid (Mueller &
Whether or not we are willing to accept Beimann, 1969). Pharmacological studies have
McClelland’s (1961) idea that collective shown that taking methylphenidate (Ritalin), a
achievement motivation fuels economic and stimulant with an activating effect, increases the
societal developments, his hypothesis does not production of achievement-related thoughts in
provide any concrete explanation for the origin the TAT (Bäumler, 1975). McClelland’s (1995)
of this relationship. Engeser, Euen, and Bos work was inspired by the observation that the
(2015) tried to find such an explanation. In strength of the need for achievement was nega-
essence, they argued that achievement-related tively related to the urine volume in male adults.
key words appearing in textbooks (“being suc- McClelland thus speculated that stimulating the
cessful,” “master something”) function as cues achievement motive leads to the release of argi-
that activate the achievement motive via seman- nine vasopressin, an antidiuretic peptide which is
tic behavioral priming and thus lead to more thought to have beneficial effects on memory per-
effort and eventually better performance. In a formance. All of these trials, however, were too
joint analysis of the achievement-oriented con- sporadic in order to justify safe assumptions
tent in textbooks and the educational achieve- about the physiological and neuroendocrine basis
ments of more than 3,000 students in Germany, of achievement-motivated behavior.
Engeser et al. (2015) found preliminary evi- This line of research has not received much
dence in support of this view. More experimen- attention until recently (see the more detailed dis-
tal research will be needed to scrutinize the cussion of the biological foundation of motivated
aforementioned explanation (semantic prim- behavior, Chap. 10). Schultheiss, Wiemers, and
ing). A group of researchers around Engeser are Wolf (2014) as well as Yang, Ramsay, Schultheiss,
currently working on such a study (Engeser, and Pang (2015) tested the role of the achieve-
Baumann, & Baum, 2016). ment motive in stress regulation during demand-
ing tasks. These studies were founded on the
assumption that people with a strong achieve-
6.3.3 Physiological ment motive (TAT) process the difficulty of a task
and Neuroendocrine and the associated uncertainty whether they can
Correlates of the Achievement succeed in a different way than their counterparts
Motive with a weak achievement motive. As reason for
this expected difference, the authors suggested
The very name of the construct seems to encour- the recollection of positive experiences in the for-
age researchers to validate measures of the mer group. People with a strong achievement
achievement motive with criteria of task perfor- motive perceive the difficulties that arise when
mance. Because performance is dependent on working on a task as challenges that indicate suc-
various factors, however, this validation process cessful mastering of the task at hand. By antici-
is difficult. One potentially interfering factor are pating this rewarding experience, they see
cognitive abilities that may differ substantially difficulties in a more positive light and feel much
across individuals and thus obfuscate motiva- less stress when working on a task.
tional influences. Alternatively, it should be pos- In order to test this hypothesis, Schultheiss
sible to use indicators of physiological activation et al. (2014) chose tasks (e.g., the Trier Social
248 J.C. Brunstein and H. Heckhausen
Stress Test by Kirschbaum, Pirke, & Hellhammer, shown that a strong achievement motive mea-
1993) that had been shown to induce stress and sured with the TAT lessens the stress reaction to
activate the HPA axis that releases the stress hor- challenging tasks.
mone cortisol. Saliva cortisol, which was
measured both before and after the tasks, indi-
cated that stress reactions had a lower intensity in 6.4 he Risk-Taking Model
T
participants with a strong achievement motive as the Dominant Research
(TAT) than in those with a weak achievement Paradigm
motive. Yang et al. (2015) replicated these find-
ings with tasks that were followed by negative Atkinson’s (1957) model of risk-taking behavior
feedback. Moreover, they found that participants has informed achievement motivation research
with a strong achievement motive not only had a since the 1960s and dominated it until the late
weaker stress reaction (cortisol) but also subjec- 1970s. Indeed, it is often referred to as the theory
tively reported a more positive mood. of achievement motivation. An introduction to
These findings show that a characteristic that the model can be found in Chap. 5. In this chap-
distinguishes people with a strong achievement ter, we examine the empirical data it has gener-
motive more than anything else is their keen- ated. The risk-taking model is characterized by
ness to deal with challenges in which it is the distinction it draws between a directional and
uncertain whether they can succeed. This trait an intensity component of motivation. The direc-
is stabilized through affective and physiologi- tional component (dominance of the success or
cal adaptation processes. Later in this chapter failure motive) determines the preferred level of
(Sect. 6.5.1), we will see that cognitive evalua- task difficulty; the intensity component influ-
tion also plays an important role in this kind of ences the efficiency of task performance.
adaptive achievement. Before we present the empirical findings, let
us briefly review the three basic assumptions of
Summary the model:
Subsequent to the development of the TAT
method of achievement motive measurement, 1. The success incentive increases with the sub-
relations between nAchievement scores and a jective difficulty of a task, while the failure
range of behavioral characteristics were investi- incentive decreases.
gated. Individuals high in achievement motiva- 2. The relationship between incentive and prob-
tion were found to outperform those low in ability of success is multiplicative. From these
achievement motivation on simple arithmetic two assumptions, it follows that the resultant
problems and learning tasks. High nAchievement motivational tendency (the difference between
scores predicted innovative and creative out- success and failure tendencies) is symmetrical
comes in real-life contexts. On the societal level, in form as a function of task difficulty:
nAchievement was found to correlate with indi- • Tasks of moderate difficulty maximize the
cators of economic development and productiv- tendency to achieve success or to avoid
ity. Recent studies suggest that failure, depending on which of the two
achievement-related content that appears in chil- motives is dominant.
dren’s literature and textbooks at school have an • For very easy or very difficult tasks, differ-
activating effect on the achievement motive. ences in the resultant tendency are rela-
Regional differences in educational performance tively small. Thus, the behavior of
are reflected in the achievement-related content success-motivated individuals can be
of the preferred schoolbooks. In addition to per- expected to differ from that of failure-
formance criteria, physiological measures of acti- motivated individuals on tasks of moderate
vation are associated with the strength of the difficulty, but not on extremely easy or dif-
achievement motive. More recent studies have ficult tasks.
6 Achievement Motivation 249
Atkinson expected this model to apply very simple tasks that no differences are to be
not only to task choice but also to persis- expected between the two groups. Here the incen-
tence and achievement outcomes. He thus tive is so low that a success is trivial. The same
explained both decision-making behavior pattern holds for the failure incentive, the only
(task choice) and execution of instrumental difference being that the failure motive now acts
activities (task performance) by reference as the weighting factor:
to the same model parameters. The prob- Vf = Mf × If
lems involved in equating these two aspects
are discussed in more detail in Chap. 5. In other words, individuals high in failure moti-
3. The valence (V) of a performance outcome is vation feel more shame at failing on a simple task
the product of motive strength (M) and incen- (If = –Ps) than do less failure-motivated individu-
tive (I): als. If the task is very difficult, however, the failure
incentive will be low because it is no disgrace for
V = M ×I anyone to be defeated by a very difficult task. In
this situation, differences between individuals
This latter assumption applies to both the scoring high versus low on the failure motive
valence of success and the valence of failure. The should be minimal. To summarize, as task diffi-
stronger the achievement motive, the stronger the culty increases, the valence of success can be
weighting of the respective incentive, producing expected to increase more steeply among individ-
marked differences in the tendency to approach uals high in success motivation than among their
success or to avoid failure. This assumption of less success-motivated counterparts. Conversely,
the risk-taking model has attracted far less as task difficulty decreases, the valence of failure
research attention, although it is critical to the can be expected to increase more steeply among
logic of the model. individuals high in failure motivation than among
their less failure-motivated counterparts.
Taken together, it can be assumed that (dis-
6.4.1 Motive-Dependent Valence tinct) successes are more attractive to success-
Gradients motivated individuals than to failure-motivated
individuals, whereas failure-motivated individu-
• One key assumption of the risk-taking model als feel more shame at (distinct) failures than do
is that valence gradients are motive success-motivated individuals. These effects are
dependent. not restricted to actual success or failure. Rather,
even before individuals have begun to tackle the
This assumption can be illustrated for the task at hand, the valences of success and failure
valence of success. The success incentive can take effect in anticipation of these outcomes.
increases with the difficulty of a task (Is = 1– Ps).
The more difficult a task, the more pride is to be 6.4.1.1 Early Findings
expected upon a successful outcome. According These assumptions have rarely been tested
to the logic of the risk-taking model, however, the directly (cf. Halisch & Heckhausen, 1988), and
success motive, which weighs the incentive asso- the few available studies yielded mixed results.
ciated with success as a multiplicator, must also The first study was conducted by Litwin (1966),
be taken into account in this prediction: who measured the valence of hits in a ring toss
Vs = Ms × Is game in terms of the prize money participants
judged to be appropriate for throws from various
Thus, success-motivated individuals experi- distances. After ten practice trials, participants
ence an even higher degree of satisfaction upon were asked to specify how much money (from 0
solving a difficult task than do less success- to 1 $) should be awarded for hits from each
motivated individuals. It is only in the context of distance. As shown in Fig. 6.7, the valence of
250 J.C. Brunstein and H. Heckhausen
a button activating a video camera at the moment tionnaire measures (e.g., MARPS, AMS, TAQ;
the spot filled a window in the beam. Task diffi- Sect. 6.2.7).
culty was manipulated by varying the speed of The achievement motive scores that were
the spot of light. derived from these two types of instruments were
A psychophysical scaling method was used to virtually unrelated. All questionnaires overlapped
measure valence in terms of respondents’ antici- with scores on Meyer’s (1972) questionnaire on
pated satisfaction or dissatisfaction with their per- the self-concept of ability (Halisch, 1986), while
formance. Respondents first identified standards the TAT scores did not.
for success and failure by specifying an upper and The results revealed a significant relationship
a lower boundary (or task difficulty level), beyond between the TAT measures and the slope of the
which they would experience success or failure, valence gradients for success and failure. The
respectively. These estimates served as anchors for same pattern of results did not emerge for any of
determining “minimal” success and failure levels. the questionnaires. Oddly, it was not the TAT net
Based on these anchor points, participants were hope score (HS – FF), but the aggregate motive
asked to specify the difficulty level at which they score (HS + FF), that interacted with task diffi-
would experience “twice” as much satisfaction culty. Individuals high in aggregate motivation
(success) or dissatisfaction (failure). The closer had a steeper valence gradient for success than
this estimate was to the respective anchor point, for failure; the reverse held for individuals low in
the steeper the valence gradient. In this method, aggregate achievement motivation. In line with
slight deviations from the anchor point thus indi- the predictions of the risk-taking model
cate a high level of emotional sensitivity to success (Fig. 6.8a), a more detailed analysis of subcom-
or failure. The two achievement motives were ponents of the success motive revealed that the
assessed with TAT (Heckhausen, 1963) and ques- content categories “positive affect,” “praise,” and
Fig. 6.8 Slope indexes of valence gradients for success and failure in (a) individuals high vs. low in success motivation
(positive affect, praise, and expectancy of success) and (b) individuals high vs. low in failure motivation (total FF score)
(Based on Halisch & Heckhausen, 1988, p. 60)
252 J.C. Brunstein and H. Heckhausen
7
valence gradient predicted the extent to which
6
a participant’s performance on the mental
5 concentration test improved throughout the
4 trials. If they had not received positive feed-
3 back in the previous trial, participants with a
× High n Ach strong achievement motive immediately
2
Low n Ach
reacted with an increase in mental concentra-
1
tion in the subsequent trial.
0
0 1 2 3 4 5 6 7 8 Taken together, these findings shed some light
Number of top performances on the mechanisms by which the achievement
Fig. 6.10 Valence of performance as a function of the
motive fuels task-related efforts in the presence
performance level (number of personal best perfor- of self-evaluative standards of comparison. Two
mances) and the strength of the achievement motive (hope discrepancies are of particular importance here
for success by Heckhausen). Shown are the gradients for (for a historical overview of the discrepancy-
hypothetical individuals who are one standard deviation
above or below the average of the achievement motive
theoretical interpretation of achievement-related
assessed with the TAT (Taken from Brunstein & Maier, behavior, see McClelland et al., 1953). The first
2005, p. 218) discrepancy is a central component of an indi-
vidual’s initial expectation. Potential success is
associated with a high level of satisfaction,
ther = 5) if they managed to score a certain num- whereas potential failure is associated with dis-
ber of personal bests (between 0 and 8 during satisfaction. Because an individual’s mental state
each half of the test). In contrast to earlier stud- depends strongly on the outcome of her or his
ies, the motive- dependent steepness of the engagement, a strong need to perform well is
valence gradient was determined using a multi- present from the beginning. The second discrep-
level data-analytic method (growth curve analy- ancy emerges if an individual realizes during an
sis, cf. Bryk & Raudenbush, 1992). This activity that the actual outcome might differ from
approach was chosen because satisfaction rat- the desired outcome. If things do not go well,
ings were assessed multiple times for the same they indicate a future state of dissatisfaction, and
individuals. Because the ratings before and dur- in reaction to this concern, more effort is put into
ing the experiment were highly correlated, the the activity. The behavioral change is combined
analysis used their average. This led to the fol- with the expectation that potential future dissatis-
lowing findings (Fig. 6.10): faction can be averted by one’s own behavior and
thus transformed into satisfaction. Subjectively,
1. For all participants, expected (dis)satisfaction the individual experiences this as mastering the
depended on the expected number of personal activity. The first discrepancy creates a strong
bests (the higher this number, the more impulse to engage in achievement-related efforts,
satisfied). while the second discrepancy has a correcting
2. This relationship was much stronger in par- function during the activity. This is only the case,
ticipants with a strong achievement motive however, if the achievement motive is strong. If
compared to those with a weak motive. this condition is met, achievement-related behav-
254 J.C. Brunstein and H. Heckhausen
ior serves to ensure that the affective reactions comparisons are used, self-reported achievement
(satisfaction and pride vs. disappointment and motives and the associated self-concept of abil-
shame) to the outcomes of an individual’s actions ity become relevant to the prediction of valence
will be positive. gradients.
A shortcoming of Brunstein and Maier’s
(2005) study is that they completely ignored fear
of failure in their analysis and conceptualized sat- 6.4.2 C
hoice: Product of Incentive
isfaction vs. dissatisfaction as (only) two poles of and Expectancy
the same affective dimension. In another experi-
ment, however, they changed the operationaliza- We now come to the expectancy aspect of the
tion of the concept of personal best performances. risk-taking model. Because incentive value
After each trial, participants received feedback hinges on the level of difficulty (Is = 1 – Ps; If
on whether their performance had been part of = –Ps) and is in turn multiplied by the probabil-
the top 33% of all participants. If a social stan- ity of success, the function for the resultant
dard of excellence was thus applied to individual motivational tendency peaks at a moderate level
performances, the steepness of the valence gradi- of difficulty. This is the point of maximum
ent was predicted by both the TAT measure and approach for success-motivated individuals, but
the questionnaire measure of achievement moti- the point of maximum avoidance for failure-
vation. The steepest gradients were found for motivated individuals. The model thus has a
participants with a strong achievement motive as symmetrical structure, as described in Chap. 5.
reflected in the TAT and a strong self-concept of The symmetry around the horizontal axis (level
achievement as assessed with the self-report of difficulty) is determined by the scores for the
instrument. In accord with Halisch and two achievement motives. Depending on which
Heckhausen’s (1988) idea, the introduction of of these two motives is dominant, an individual
social comparisons resulted in an involvement of either prefers (Ms > Mf ) or avoids (Ms < Mf )
participant’s self-image pertaining to moderately difficult tasks. This symmetry rests
achievement. The question of how directly (TAT) on two assumptions:
and indirectly (self-report) measured motives
interact in the regulation of affect and behavior is 1. The incentive is a function of the level of task
discussed elsewhere in this volume (Chap. 9). difficulty. This assumption is not only intui-
tively reasonable, but has also been confirmed
Summary in numerous studies (Feather, 1959b,
Determining motive-dependent valence gradi- Karabenick, 1972; Meyer, Niepel, & Engler,
ents is relevant to test one of the key assumptions 1987; Schneider, 1973, experiment 2).
of the risk-taking model. There still has not been 2. Approach and avoidance motivation peak at a
a complete confirmation that the strength of the moderate level of task difficulty (P = 0.50),
success motive and the strength of the motive to i.e., the point at which the product of incentive
avoid failure weight the incentives of success and expectancy reaches its maximum.
and failure. Nonetheless, individual studies have
managed to show that the valence of success can Studies seeking to test this assumption have
be conceptualized as the product of difficulty been dogged by numerous difficulties, as summa-
incentive and success motive (Ve = Me × Ie). rized below.
This has not yet been shown for failure motiva-
tion. In general, only TAT variables can uncover 6.4.2.1 Objective and Subjective
differences in the valences of success and fail- Probability of Success
ure. Questionnaire measures are unable to pro- Various methods have been used to measure the
vide such information. If, however, social probability of success. Atkinson (1957) initially
6 Achievement Motivation 255
1.00
• One’s own experience (how well one per-
Subjective Probability of Success
Ps
formed on previous attempts to solve a certain
task) .75
The tendency to overestimate one’s perfor- level and previous attainment), the goal set by
mance (see the study above) seems to be charac- the participant is compared with his or her
teristic of achievement-related behavior. It is prior performance.
almost as if the desire to improve one’s perfor-
mance outcomes was factored into the expec- From the outset, a consistent pattern of results
tancy value. At least, this is the pattern observed emerged. The level of aspiration does not increase
when a task has already been attempted and per- steadily with the strength of success-oriented
formance outcomes are dependent on effort and achievement motivation; rather, there is a prefer-
practice. Expectancies formed without prior ence for high but attainable goals and avoidance
exposure to a task may have to be corrected after of unrealistically high ones. Many of the studies
the first few attempts. The reliability of these using the task-selection paradigm have involved
expectancies is correspondingly low, and they are ring toss games. In a study with kindergarten
not suitable for testing the risk-taking model. children as participants, McClelland (1958c)
Similar problems have emerged for social com- found that success-motivated children preferred
parison norms (e.g., “This task was solved by “calculated risks” and chose tasks that were nei-
50% of the previous participants”). The diver- ther too easy nor too difficult. Figure 6.12 shows
gence of the subjective anticipation of success the distances chosen by success-motivated and
from the stated norm may be more or less pro- failure-motivated students (as measured by
nounced, depending on how an individual ranks nAchievement and TAQ) in a study by Atkinson
his or her task-specific ability relative to that of and Litwin (1960). The preference for intermedi-
the reference group. Furthermore, research has ate distances was much more pronounced among
shown that respondents often have little confi- success-motivated students than among failure-
dence in probabilities of success or failure motivated students. Heckhausen (1963) reported
reported by an experimenter (Feather, 1963, similar findings from an analysis of goal-setting
1966). A certain amount of exposure to a task behavior in a labyrinth task. The difficulty of the
thus seems to be indispensable if reliable data on task was varied by presenting labyrinths of differ-
probabilities of success are to be obtained. ent sizes; the achievement motive was assessed in
terms of a TAT measure of net hope (HS – FF).
6.4.2.2 Level of Aspiration: Task Choice Success-motivated individuals chose goals that
and Goal Setting were comparable to, or moderately higher than,
The risk-taking model was originally developed their previous performance, whereas failure-
to explain how levels of aspiration are set. Two motivated participants were more likely to set
experimental paradigms can be used to examine themselves goals that were either extremely dif-
this mechanism: ficult or extremely easy relative to their earlier
performance (Fig. 6.13).
1. In the task-selection paradigm, participants
Studies designed to test whether the most fre-
choose between tasks of the same type repre- quently chosen difficulty levels fall into a broadly
senting different levels of difficulty (e.g., defined “intermediate” range have produced data
throwing from different distances, shooting substantiating the risk-taking model. Upon closer
from the same distance at goals of different inspection, however, three problems are apparent,
widths, or solving increasingly complex laby- two of them empirical and one theoretical in nature.
rinth problems). When the preferred probabilities of success are
2. In the goal-setting paradigm, participants exe- examined in more detail, a marked deviation from
cute repeated trials on a single task. The goal the risk-taking model is observed. The maximum
is defined in terms of the time required to exe- preference, whether defined in terms of objective or
cute the task, the number of correct solutions, subjective probability of success, falls below the
or the number of mistakes. To determine goal critical level of Ps = 0.50; as a rule, it is between
discrepancy (difference between current goal 0.30 and 0.40. In other words, people do not prefer
6 Achievement Motivation 257
Percentage of Shots
achievement motive
(Based on Atkinson & 15
Litwin, 1960, p. 55)
10
5 10 15
Distance From Target (in Feet)
10
NH (HS - FF )
-5
-10
1-20 21-40 41-60 61-80 81-100
Percentile of Goal Discrepancy Distribution
Fig. 6.13 Goal discrepancies for labyrinth tasks as a function of net hope (hope for success - fear of failure) in
Heckhausen’s TAT measure (Based on Heckhausen, 1963, p. 95)
tasks of moderate difficulty, but opt for somewhat have also been documented for unselected samples
more difficult tasks. Moreover, failure-motivated of school students (McClelland, 1958).
individuals do not choose extremely difficult tasks Besides these two empirical problems, there is
to anything like the extent predicted by the risk- a third problem that is inherent in the risk-taking
taking model. Atkinson speculated that too few of model itself. The model does not predict whether
his student participants were high in failure motiva- a failure-motivated individual will be more likely
tion (Atkinson & Litwin, 1960; Atkinson & Feather, to opt for extremely difficult or extremely easy
1966). Yet findings similar to those reported above tasks. Heckhausen (1963) proposed a possible
258 J.C. Brunstein and H. Heckhausen
solution to this problem, suggesting that the task shows the three indexes for predictions of hits in
choice of failure-motivated individuals depends a motor skills task (goal-shooting game). The
on the strength of their aggregate motivation (AM findings for all three indexes were inconsistent
= HS + FF). If their aggregate motivation is high, with the symmetrical form predicted by the risk-
so goes Heckhausen’s reasoning, they will prefer taking model. Decision time peaked well below
extremely difficult tasks; if it is low, they will the objective probability of 0.50 (when respon-
choose very easy tasks. In other words, failure- dents had chalked up as many successes as fail-
motivated individuals high in aggregate motiva- ures on previous trials). Likewise, confidence
tion will tend to expect too much of themselves, reached its lowest value well below this point.
and those low in aggregate motivation will not Subjective assessments of the probability of suc-
stretch themselves enough. Jopt (1974), Schmalt cess were higher than would be expected on the
(1976a), and Schneider (1971) reported evidence basis of the objective data. Schneider attributes
for the validity of these hypotheses. these findings to a “hope bonus” that people add
It is worth asking whether these discrepancies to their performance level when thinking about
from the risk-taking model are attributable to the future. This bonus may explain why people
shortcomings in the measurement of probability tend to tackle tasks that slightly exceed their cur-
of success (or task difficulty). In addition to self- rent level of performance.
reports, Schneider (1973, 1974; Schneider & There have been many attempts to adapt the
Heckhausen, 1981) used an objective index to risk-taking model to this body of findings
determine the probability of success, namely, the (Hamilton, 1974; Heckhausen, 1968; Nygard,
time it took respondents to decide whether or not 1975; Wendt, 1967). In most cases, additional
they would succeed. Moreover, Schneider asked variables have been specified and incorporated
respondents to state how confident they were in into revisions of the model. Examples of such
this judgment (confidence rating). Figure 6.14 variables are:
objective probability of
success and absolute .70 4.5
difficulty levels (Based
on Schneider, 1974,
Decision Time
.60 4.0
Confidence
p. 154)
.50 3.5
C
.40 3.0
.30 DT 2.5
.20 Ps 2.0
.10 1.5
0
.14 .16 .28 .29 .37 .50 .61 .67 .75
Objective Probability of Success
6 Achievement Motivation 259
• Personal standards defining the difficulty level ration levels after success. The risk-taking model
at which a certain success incentive is reached can explain this seemingly rather peculiar behav-
(Kuhl, 1978b) ior in terms of an interaction between the proba-
• Inertial tendencies resulting from previous bility of success and the failure motive. Atypical
attempts to complete a task, which afford a shifts can be expected when failure-motivated
kind of additional motivation for future tasks individuals experience an unexpected success on
(Atkinson & Cartwright, 1964; Weiner, a difficult task or a surprising failure on an easy
1965a, 1970) one. In both cases, the probability of success
• Future-oriented tendencies that take effect approaches the intermediate range, i.e., precisely
when task attainment entails a number of the range of difficulty that failure-motivated indi-
consecutive steps, e.g., in the context of long- viduals seek to avoid. As a result, the level of
term goals (Raynor, 1969, 1974; Raynor & aspiration shows erratic shifts toward the other
Roeder, 1987) end of the task difficulty scale. The pattern of
results predicted by the risk-taking model is illus-
These revisions are described in detail else- trated in Fig. 6.15.
where (Heckhausen, 1980; Heckhausen et al., Moulton (1965) tested these inferences by
1985). None of them proved a resounding suc- inducing three task difficulty levels (symmetri-
cess, however. cally distributed probabilities of success of 75%,
50%, and 25%, respectively). Respondents were
6.4.2.3 Typical and Atypical Shifts first instructed to select one of the three tasks, but
in the Level of Aspiration they were then all administered the moderately
Moulton (1965) took an apparently paradoxical difficult task. Moulton induced failure for partici-
finding from research on the level of aspiration pants who had chosen the easy task and success
and used it as a test case for the validation of the for participants who had chosen the difficult task.
risk-taking model. He studied the atypical shifts Participants were then free to choose the next
in aspiration levels that are sometimes observed task. As shown in Table 6.6, the results were in
after task accomplishment, namely, increased line with the assumptions of the risk-taking
aspiration levels after failure and decreased aspi- model. In the free-choice condition, the majority
A B C
Choice of A: Ps
0 .25 .50 .75 1.00
A B C
Choice of C: Ps
0 .25 .50 .75 1.00
Fig. 6.15 Atypical shifts in the aspiration levels of failure-motivated individuals who have succeeded on a difficult task
(shift from A to C) or failed on an easy one (shift from C to A), as derived from the risk-taking model
260 J.C. Brunstein and H. Heckhausen
Table 6.6 Initial task preferences and subsequent typical versus atypical shifts in the level of aspiration of success- and
failure-motivated individuals
Difficulty level of task initially chosen
Easy Intermediate Difficult Shift in level of aspiration
(Ps = 0.75) (Ps = 0.50) (Ps = 0.25) Atypical Typical
Success-motivated individuals (N = 31) 1 23 7 1 30
Failure-motivated individuals (N = 31) 9 14 8 11 20
Based on Moulton (1965, pp. 403–404)
of success-motivated individuals chose moder- tion) thus predict a preference for moderately
ately difficult tasks, whereas a substantial propor- difficult tasks.
tion of the failure-motivated participants opted Trope (1975, 1980, 1986b; Trope & Brickman,
for easy or difficult tasks. The results also sub- 1975; for overviews, see Trope, 1983, 1986a)
stantiated predictions on change in the level of compared and contrasted the two principles in an
aspiration. Relatively few participants made attempt to determine which is decisive for task
atypical choices, and all but one of those who did selection. To this end, he assigned higher diag-
belonged to the failure-motivated group. nosticity for self-evaluation of ability to either
easy or difficult tasks. In other words, respon-
6.4.2.4 Striving to Maximize Affect or dents were told that certain tasks either distin-
to Obtain Information? guished very clearly between people high and
According to the risk-taking model, the prefer- low in ability (high diagnosticity) or barely dis-
ence for moderate levels of difficulty maximizes tinguished between the two groups (low diagnos-
the anticipated affect, be it pride at success or ticity). Trope found that respondents generally
shame at failure. Success-motivated individuals preferred high to low diagnosticity tasks. He
thus prefer moderately difficult tasks because interpreted these findings as indicating that peo-
they promise the highest degree of satisfaction; ple strive to reduce uncertainty about their ability
failure-motivated individuals avoid these tasks levels. Notably, individuals high in achievement
because they risk the highest degree of shame. motivation are even more likely to choose highly
The behavior of the former group is geared at diagnostic tasks than those low in achievement
maximizing positive self-evaluative emotions motivation (Trope, 1980).
that of the latter group at reducing negative self- From Trope’s (1986b) interpretation, it fol-
evaluative emotions (Sect. 6.5). Other authors lows that achievement-motivated individuals
have pitted the principle of maximizing affect seek to obtain realistic and valid information
against the principle of obtaining information, about their abilities as a matter of principle,
based on Festinger’s (1954) theory of social com- whether this information proves to be positive
parison processes. According to this second prin- (success) or negative (failure; see also Meyer &
ciple, people have a fundamental need to acquire Starke, 1982). This need for self-assessment can
insights into their own attitudes, opinions, and be compared with the need for “self-
skills and to evaluate these attitudes, opinions, enhancement,” another fundamental motive of
and skills in social comparison with others. self-evaluation (Sedikides & Strube, 1997). Some
Accordingly, they prefer moderately difficult authors have postulated that achievement-
tasks that split populations into high vs. low abil- motivated individuals are primarily interested in
ity groups of approximately equal size and thus demonstrating their superior abilities (Kukla,
have the greatest information value with respect 1972a, 1972b, 1978), implying that they prefer
to one’s own ability levels (Meyer, 1973; tasks that afford them the opportunity to
Schneider, 1973; Weiner et al., 1971). Both prin- emphasize positive aspects of the self and thus to
ciples (maximizing affect and obtaining informa- enhance their self-esteem.
6 Achievement Motivation 261
Consensus has not yet been reached on which often perform worse than those with failure
of these two needs (self-assessment or self- motivation (Szeto et al., 2011).
enhancement) is dominant in determining which
tasks will be selected by achievement-motivated Attempts have also been made to relate differ-
individuals. Sorrentino (Sorrentino & Hewitt, ing needs for self-evaluation to features of the
1984; Sorrentino, Roney, & Hewitt, 1988) assessment situation (Taylor, Neter, & Wayment,
reported that both needs influence task choice, 1995). Brunstein and Maier (2005) found that
but that they are associated with different per- individuals who describe themselves as achievers
sonality characteristics. The achievement act according to the principle of self-enhancement
motive (nAchievement) is oriented to maximiz- when the ability being tested is socially desirable
ing the affective value of a task and predicts how and according to the principle of self-assessment
much value individuals attach to obtaining feed- in less ego-involving situations. As Sedikides and
back on high abilities (success-motivated indi- Strube (1997) pointed out, the relations between
viduals) or avoiding feedback on low abilities achievement motives – whether assessed by the
(failure-
motivated individuals). As such, TAT or by questionnaire measures – and different
nAchievement can be interpreted as a motive needs for self-evaluation warrant careful exami-
geared to affect maximization. However, nation in future research.
Sorrentino identified another motive, labeled
uncertainty orientation, that can also be assessed
using the TAT (nUncertainty; cf. Sorrentino, 6.4.3 Persistence
Hanna, & Roney, 1992) or related techniques
(cf. Szeto, Sorrentino, Yasunaga, Kouhara, & Persistence is the second major criterion
Lin, 2011). People high in uncertainty orienta- against which the risk-taking model has been
tion generally strive to obtain information about tested. Persistence can manifest itself in vari-
themselves and their social environment. This ous forms:
cognitive need is expressed in the tendency to
choose tasks that promise to provide as much • Duration of uninterrupted pursuit of a task
new information as possible, whether it is indic- • Resumption of an interrupted or unsuccessful
ative of high or of low ability levels (Sorrentino activity
& Hewitt, 1984). As such, nUncertainty can be • Long-term pursuit of a superordinate goal
interpreted as a motive geared to the self-assess- (e.g., career success)
ment of abilities.
Heckhausen and Kuhl (1985) have questioned
• Integral to the theory of uncertainty orienta- whether the motivation to choose a task can be
tion (Sorrentino, Smithson, Hodson, Roney, & equated with the motivation that occurs when
Walker, 2003) is the idea that achievement engaged in a task. Where long-term persistence is
motives (hope for success and fear of failure) concerned, Raynor (1969, 1974) was quick to
are only activated if the current certainty vs. point out that the risk-taking model would have
uncertainty of a behavioral situation (e.g., a to be extended to yield valid predictions in this
pre-structured vs. an open lesson) fits the domain as well. More specifically, actions would
uncertainty orientation of the individual. If have to be broken down into a series of more or
this condition is met, people with strong suc- less interconnected subactions, the outcome of
cess motivation perform better than their each determining whether or not a person is per-
counterparts with strong failure motivation. If mitted to continue along the path in question
the condition is not met, however, both (e.g., passing academic exams is the prerequisite
motives are deactivated, resulting in a reversal for entering a graduate career). This model is par-
of motive-dependent performance differences: ticularly suited to predicting persistence in the
in this situation, success-motivated people pursuit of long-term, superordinate goals
262 J.C. Brunstein and H. Heckhausen
An interrupted action leaves a residual tension tion can be derived for the resultant motivational
that becomes manifest as soon as it is no longer tendency (Tr):
suppressed by another, stronger action tendency.
Tr = ( Ms × Ps × Is + TGi ) + ( Mf × Pf × If + T− Gi )
Atkinson and Cartwright (1964) integrated these
ideas into the risk-taking model, adding to the
success tendency (Ts) the “inertial tendency” The resultant inertial tendency increases the
(TGi) that results from not having completed an motivation of success-oriented individuals to
earlier achievement-related activity: engage in achievement-related activities and
inhibits the motivation of failure-oriented indi-
Ts = Ms × Ps × Is + TGi ,
viduals to resume failed activities or related
where T denotes an action tendency, G (“goal”) a activities. In this point, Weiner’s model departs
particular class of action goals (here, achieve- from the Atkinson and Cartwright conception of
ment), and “i” (“inertial”) the fact that the ten- inertia: after failure, success-motivated individu-
dency in question derives from an unfinished or als are expected to experience a gain in motiva-
failed activity. As soon as the individual embarks tion and failure-motivated individuals to
on an activity relating to the same theme, this experience a loss. In line with this hypothesis,
persistent inertial tendency is added to the moti- Weiner (1965b, 1979) found that success-
vation already activated. In other words, Atkinson motivated individuals performed better after fail-
and Cartwright (1964) assumed that inertial ten- ure than after success, whereas failure-motivated
dencies can be transferred to the entire spectrum individuals showed better performance after suc-
of action tendencies in the same thematic cate- cess than after failure.
gory. Both the classic literature on the substitute
value of actions (Henle, 1944; Lissner, 1933;
Mahler, 1933) and more recent works on the 6.4.4 Performance Outcomes
topic ( Wicklund & Gollwitzer; 1982; Brunstein,
1995) suggest that it is unrealistic to assume such It is a daring undertaking to predict not only task
a broad level of generalizability. Nevertheless, choice but also performance outcomes on the
Atkinson and Cartwright can be commended for basis of resultant motivational strength.
expanding the perspective on individual episodes Motivation is a variable better suited to explaining
of achievement-related behavior to cover multi- intraindividual variation in performance than
ple action tendencies. This perspective only came interindividual differences in performance out-
to full fruition in the theory of the dynamics of comes. These interindividual differences derive
action, which was developed by Atkinson and primarily from differences in task-related abili-
Birch (1970, 1974; see also Revelle, 1986; ties, which often have little to do with motive vari-
Revelle & Michaels, 1976) to explain the inter- ables (a highly motivated novice will not be able
play of different action tendencies competing for to match the performance of an expert in a given
the access to behavior. domain, even if the expert makes no great effort).
Atkinson and Cartwright (1964) only postu- But even when individual differences in ability
lated an (positive) inertial tendency for the suc- are controlled, there is still no coherent theory to
cess tendency. Weiner (1965a, 1970) extended explain how achievement motivation influences
this conceptualization to the tendency to avoid the individual steps involved in task performance
failure. After a failure, the previous success ten- or the associated patterns of information
dency (TGi) and failure tendency (T–Gi) both con- processing.
tinue to exist (the minus sign indicates that the Krau (1982) noted that the motivation to
persistent failure tendency has an inhibiting select a task should not be equated with the moti-
effect on achievement behavior). Building on the vation that occurs when engaged in a task. Goal
original risk-taking model, the following equa- setting and goal pursuit refer to different action
6 Achievement Motivation 265
phases that are determined by different variables. performance gains than their failure-motivated
Specifically, Krau distinguished the following classmates. Weiner (1967) reported comparable
action-phase and associated variables: data for college students, with success-motivated
students benefiting most from ability grouping.
Action-phases Variables Gjesme (1971) presented similar findings,
Goal setting Estimated task difficulty; strength having taken a somewhat different approach. He
of the individual achievement
motive assigned students from mixed-ability classes to
Preparation Planned effort expenditure aptitude groups based on their intelligence scores
Execution Actual effort expenditure and and found, as expected, that it was only in the
work-related attitudes moderate-ability group that the success motive
was positively, and the failure motive negatively,
As expected, Krau found that the achievement related to school performance. Assuming that
motive does not have an impact on persistence instructional demands fell in the moderate diffi-
and performance directly, but that it affects per- culty range for students of moderate intelligence
formance outcomes indirectly by increasing the only, these findings are consistent with the risk-
amount of effort that people plan to expend (or taking model.
are willing to invest). It seems rather rash, in view These data should not be interpreted as sup-
of these findings, to assume that achievement porting ability grouping in schools, however.
motivation (or indeed the achievement motive First, instruction can be individualized to ensure
itself) has direct and unmediated effects on task that the tasks assigned are neither too easy nor
performance. Nevertheless, achievement motiva- too difficult (“principle of fit”; Heckhausen,
tion research has generated various noteworthy 1969). Second, when cooperative learning meth-
models and findings concerning the relationship ods are applied, heterogeneity of the student
between motivation and performance. Krau’s body is no impediment to creating realistic, com-
arguments were later integrated within the petitive classroom settings that do not over- or
Rubicon model of action phases (Chap. 11). understretch students (Slavin, 1995). Moreover,
the opportunity to select and work on tasks inde-
6.4.4.1 School Performance pendently can have positive effects on task moti-
It would seem logical for researchers to examine vation, at least when students are predominantly
the relationship between achievement motivation success motivated (and thus choose moderately
and school performance. Studies of this type difficult tasks). McClelland (1980) attributed the
must control for both motivational dispositions low (to nonexistent) correlations found between
(e.g., hope for success and fear of failure) and the achievement motive (nAchievement) and
task difficulty. Researchers can only expect to school performance to the fact that the incentives
find substantial relations between motive mea- essential for activating the achievement motive
sures and performance measures when character- (difficulty, novelty, variation, self-determination,
istics of the instructional setting and the tasks informative feedback) are often not present in the
assigned are taken into account (unless the classroom, in contrast to occupational settings,
achievement motivation data also reflect differ- where they are either more easily accessible or
ences in school performance). One way of get- can be actively sought out. These arguments are
ting around this problem is to examine all based on the assumption that motives are dis-
ability-based groups. It can be assumed that most positional variables. However, expectancy-value
students in these classes find the work assigned theories have also been successfully applied to
moderately difficult. O’Connor, Atkinson, and predict school performance, as illustrated in the
Horner (1966) found that success-motivated stu- excursus on this page based on the research of
dents in homogeneous classes showed greater Eccles and Wigfield.
266 J.C. Brunstein and H. Heckhausen
.80
Success Motivated carefully modeled through the in-depth analysis
of an individual’s ongoing feelings, thoughts, and
Failure Motivated
.60 actions during task performance. This approach to
.40
the analysis of task performance would require to
Outcome Measure
a task is very complex (C), motivational strength where (overly) high success motivation would
reaches its optimal level even sooner. Hence, a have detrimental effects.
given motivational strength can have very different
effects on performance outcomes depending on • What distinguishes this new approach is the
the type of task at hand (in other words, more assumption that the effects of motivational
motivation does not automatically mean better strength on performance are moderated by
performance). task complexity.
The motivational strength to perform a task is
determined by three variables: The model was tested with data from empiri-
cal studies addressing the effects of multiple
1 . The person’s motives motives and incentives on task performance. The
2. The perceived difficulty of the task (probabil- idea behind this approach was very simple: the
ity of success) interaction of multiple motives and incentives
3. The situational incentives (e.g., consequences can easily result in a state of overmotivation that
of self and other evaluation after success and impairs subsequent task performance. Most of
failure) these studies were summarized in the volume
edited by Atkinson and Raynor (1974) and based
These assumptions are largely in line with on the reanalysis of published data.
those of the risk-taking model. Atkinson contin- Entin (1974) measured the achievement and
ued to suggest that the relationship between the affiliation motives of student respondents (per-
tendency to avoid failure and the tendency to son characteristics) presented with simple or
approach success is subtractive, leading to the complex calculations (task characteristics). The
logical, though seemingly paradoxical, hypothe- situational context was endowed with achieve-
sis that high failure motivation can have favor- ment-related (private feedback) or affiliation-
able effects on performance on complex tasks, related (public feedback) incentives (situational
6 Achievement Motivation 271
characteristics). In the private feedback condi- Table 6.8 Mean number of anagrams solved as a func-
tion, success-motivated students performed bet- tion of the resultant achievement motive (nAchievement –
TAQ), affiliation motive (TAT), and three incentive
ter than failure-motivated students, regardless of conditions (N = 88 male students; scores were standard-
the complexity of the task. In the public feedback ized to have a mean of 50 and a standard deviation of 10)
condition, respondents with high scores in both Condition
motives (achievement and affiliation) showed No Female
marginal performance deficits as a result of over- Motive competitor competitor Male
motivation. Again, no differences were found constellation (alone) competitor
between simple and complex tasks. High affiliation motive
Atkinson’s (1974b) reanalysis of studies High success 46.5 53.9 48.4
motive
reported by Atkinson and Reitman (1956) and
High failure 41.8 53.6 56.1
Reitman (1960) was rather more convincing. motive
Participants were given math tasks in a multi- Low affiliation motive
thematic incentive situation (group competition, High success 48.4 53.4 53.7
encouragement by the experimenter, and promise motive
of reward). Success-motivated respondents per- High failure 40.8 47.7 46.7
formed less well under these conditions than in a motive
situation with few extrinsic achievement incen- Based on Horner (1974a, p. 249)
tives. The reverse held for participants with a low
resultant achievement motive, who benefited from
the introduction of additional incentives and per- The most convincing evidence to date for over-
formed better under these conditions. Findings motivation leading to performance decrements was
from further studies confirm that multi-thematic reported by Short and Sorrentino (1986).
incentives soon lead to performance decrements in Participants worked on a rule construction task,
success-motivated individuals, whereas less- either alone or in small groups. When the incentive
motivated or failure-motivated participants tend to of group work was added, a combination of high
benefit from the provision of additional incentives. success and high affiliation motives predicted a
Horner (1974b) asked male students to solve performance decrement, whereas a high failure
math problems and anagrams, either alone or in motive was associated with enhanced performance.
competitive situations with a male or a female This is one of the few studies that has succeeded in
opponent. Again, the resultant achievement demonstrating that the failure motive has a subtrac-
motive and the affiliation motive were assessed. tive effect on the achievement tendency and can
Table 6.8 documents the findings for the anagram thus diminish the effects of overmotivation.
tasks (the pattern of results obtained for the math Nevertheless, three points warrant further
problems was similar). When working indepen- consideration:
dently, success-motivated students performed
much better than failure-motivated students. 1. There has been surprisingly little empirical
When competitive incentives were added, a dif- investigation of Atkinson’s hypothesis that
ferent picture emerged, particularly for respon- task complexity moderates the effects of moti-
dents competing with a same-sex opponent (i.e., vational strength on performance. This
in this case with a male). Under these conditions, endeavor would doubtless be facilitated by a
participants high in both the success and the affil- taxonomy permitting more precise definitions
iation motive performed just as poorly as partici- of task complexity and the associated infor-
pants low in both of these motives. In the former mation processing demands (cf. Wood, 1986).
case, the performance decrement was attributed Strictly speaking, the core premise of the
to the effects of overmotivation and, in the latter achievement model described above remained
case, to the effects of undermotivation. untested.
272 J.C. Brunstein and H. Heckhausen
IMMEDIATE ENVIRONMENT
AS GUIDE TO ACTION
PERSONALITY
NATURE OF THE TASK (A) CUMULATIVE EFFECT
EFFICIENCY
MOTIVES
TIME SPENT
STRENGTH OF AT WORK ON THE PERSON
MOTIVATION (TA) INCREASE IN ABILITY,
KNOWLEDGE, AND BELIEFS
FORMATIVE CHANGE IN CONCEPTIONS
ENVIRONMENT KNOWLEDGE, STRENGTHENING OR
BELIEFS, AND WEAKENING OF MOTIVES
CONCEPTIONS
INCENTIVES AND
OPPORTUNITIES
STRENGTH OF
MOTIVATION FOR
ALTERNATIVES
(TB-----TZ)
IMMEDIATE ENVIRONMENT
AS GOAD TO ACTION
Fig. 6.20 The dual role of motivation as a determinant of of task performance. Second, strength of motivation has a
cumulative achievement. Besides individual ability on a direct influence on the time devoted to a task, although
specific task (A), strength of motivation (TA ) influences strength of motivation for alternative activities (TB ... TZ )
performance gains in two ways. First, together with the must also be taken into account here (Based on Atkinson,
specific demands of the task, it determines the efficiency Lens, & O’Malley, 1976, p. 51)
motivation and inhibit performance (e.g., by acti- effort in successive phases of an activity. Atkinson
vating fears), this condition is unlikely to be met assumed an almost linear relationship between
in real-life contexts. Scores on ability tests thus the strength of the (activated) achievement motive
represent a mixture of true ability and motivation- and the time devoted to an activity. In the long
dependent efficiency that is difficult to run, high efficiency coupled with high time
disentangle. Simply instructing test takers to do investments results in a high level of cumulative
their best does not suffice to neutralize these achievement.
influences, as research showing that scores on In other words, because quality of performance
mental concentration tests are influenced by the is dependent on both ability and efficiency, it fol-
induction of success- and failure-related motiva- lows that cumulative achievement is the product
tional states has demonstrated (cf. Brunstein & of performance quality and time invested in a
Gollwitzer, 1996; Brunstein & Hoyer, 2002). task. The latter is determined by the strength of
In Atkinson’s view, individuals high in motive the success motive and by the presence of envi-
strength are at particular risk for becoming over- ronmental incentives capable of arousing this
motivated and suffering from performance decre- motive. Of course, incentives and motives relating
ments in high arousal situations such as exams. to alternative activities (e.g., meeting up with
Yet under everyday working conditions, where friends rather than doing one’s homework) may
achievement-relevant incentives are less preva- also take effect. Which activity is performed, and
lent, these individuals benefit from high motive how much time is invested, ultimately depends on
strength. In these contexts, their high motiva- the relative strengths of the competing motives.
tional strength is within the range of optimal Motivation thus serves a dual function in cumula-
efficiency and fosters the investment of time and tive achievements. First, it influences the effi-
276 J.C. Brunstein and H. Heckhausen
ciency with which a task is performed. Second, it acquisition of new knowledge. Entirely convinc-
influences the time invested in that task. ing evidence for this hypothesis has yet to be pre-
This model has important implications, not sented, however.
only for predicting cumulative achievements but
also for the long-term acquisition of knowledge Summary
and skills. Besides having an impact on current A good deal of research on achievement motiva-
performance, the multiplicative interaction tion has drawn on Atkinson’s risk-taking model.
between performance quality and working (or Although studies of task choice and persistence
learning) time affects the individual himself or provided evidence in support of this model, the
herself in the sense that it furthers the develop- insights it provided into achievement levels and
ment of important competences and skills. As the learning trajectories were rather limited. Whereas
proverb says, “practice makes perfect.” Hence, quantity of performance is dependent on strength
Atkinson anticipated an idea that was later devel- of motivation, the same only applies to quality of
oped in expertise research (Ericsson, 1996): performance under very specific conditions.
excellence, in any given area of expertise, Therefore, Atkinson developed various models to
requires intensive and regular practice, with a predict the effects of motivational strength on the
focus on insightful learning (“deliberate prac- efficiency of performance at various levels of dif-
tice”) rather than routine drills. ficulty. In doing so, he established that both
Given its complexity, the model has mainly undermotivation and overmotivation can cause
been used as a framework theory for explaining performance decrements. In the case of cumula-
multiply determined performance trajectories tive achievement, ability levels have to be taken
(e.g., the development of scholastic achieve- into account as well; it is the interaction between
ments; cf. Helmke & Weinert, 1997). Yet detailed ability and motivation that determines the quality
empirical analyses are scarce. Sawusch (1974) of long-term performance. It has not been possi-
could validate the model’s key assumptions in a ble to confirm the predictions of the risk-taking
computer simulation. Because this analysis drew model for the effects of failure motivation, prob-
on artificial data, its results should be interpreted ably because fear of failure is not purely an
with caution. Atkinson et al. (1976) assessed avoidance motive.
resultant achievement motivation (nAchieve- As yet, there have been relatively few efforts
ment – TAQ) and intelligence levels of sixth and to test the core assumptions of the risk-taking
ninth graders and used these data to predict aca- model. Findings on the valences of success and
demic performances at the end of their partici- failure and on subjective evaluations of the prob-
pants’ school career (grade 12). Overall, ability of success indicate either that achievement
differences in intelligence explained more vari- behavior in real-life contexts deviates from the
ance in students’ final grades than did motiva- model’s assumptions of symmetry (with respon-
tional differences. There was also an interaction dents preferring rather difficult tasks to tasks of
effect between strength of motivation and intelli- moderate difficulty) or that researchers have not
gence. High motivation predicted better school yet succeeded in measuring the critical variables
performance only among students in the upper (e.g., the probability of success) with a sufficient
range of the intelligence distribution. This find- degree of accuracy.
ing is consistent with Atkinson’s idea of cumula-
tive achievement: it is only at high ability levels
that motivational strength – mediated by effi- 6.5 Achievement Motivation
ciency – can have positive effects on performance and Self-evaluation
quality. Furthermore, the relationship between
motive strength and ability level was more pro- How can the findings on achievement motivation
nounced in grade 9 than in grade 6. This finding theory described above best be integrated and
might indicate that motive strength – mediated by interpreted? As impressive and differentiated as
the time spent on school work – promotes the these data may be, the question remains of how
6 Achievement Motivation 277
Governing Directive:
Enhance One's Competence
Affective Balance:
Anticipatory Emotion:
Positive Self-Evaluative
Hope for Success
Emotions Predominate
Success Motivation
as a Self-Reinforcing
System
Attribution:
Goal Setting:
Success: Ability/Effort
Challenging Task
Failure: Lack of Effort
Outcomes:
Balanced Failure/Success Ratio
intermediate difficulty, like their success-moti- conflict between competence striving and threat to
vated counterparts, they would be punished by self-esteem can have detrimental consequences for
recurrent negative self-evaluative emotions. student learning).
Simply describing what failure-motivated indi- The self-evaluation model was welcomed as a
viduals do not do cannot provide a satisfactory heuristic framework that unifies and clarifies many
understanding of how failure motivation affects the of the findings produced in decades of research on
regulation of achievement behavior. The adaptive achievement motivation. Heckhausen’s idea of
functions of failure-driven behavior must also be describing success- and failure-motivated behav-
identified. Heckhausen (1975a) proposed that the ior in terms of a self-regulating and self-reinforc-
directive governing the behavior of failure-moti- ing system has since generated much further
vated individuals is markedly different from the research, the effects of which are most apparent in
directive hypothesized for success-motivated indi- applied motivation psychology. One of the mod-
viduals. Its ultimate aim is to reduce threats to self- el’s major implications is that any attempts to
esteem or, if possible, to avoid them altogether. The transform failure motivation into success motiva-
behavior of failure- motivated individuals is not tion (e.g., in training programs) must target three
driven by the goal of doing things better and better, subprocesses at once:
but gives priority to the goal of protecting one’s
self-esteem. Because failure-motivated individuals • Goal setting
associate achievement-related behavior with nega- • Causal attributions
tive self-evaluative emotions (fear of failure prior to • Achievement-related affect
an achievement-related activity and shame when a
failure actually occurs), the only possible self- A focus on just one of these three subpro-
reinforcing factor is a form of negative reinforce- cesses would risk the intervention’s success being
ment, namely, avoiding experiences that will compromised by the effects of the neglected
threaten self-esteem. Choosing extremely difficult elements.
or extremely easy tasks, low persistence, and aban- On this basis, Rheinberg and Krug (2004; see
doning achievement-related activities are just a few also Rheinberg & Engeser, 2010) have devel-
of the many measures that can help to diminish or oped student training programs that have been
avert threats to self-esteem (see Higgins, Snyder, & shown to bring about a sustained increase in
Berglas, 1990, and Schwinger, Wirthwein, hope for success and a corresponding decline in
Lemmer, & Steinmayr, 2014, for further self-hand- fear of failure. Furthermore, Fries (2002; Fries,
icapping strategies, people use to shield their self- Lund, & Rheinberg, 1999) showed that a train-
esteem in threatening achievement situations). All ing program targeting all three subprocesses can
these approaches serve either to minimize the prob- increase the efficacy of treatments to enhance
ability of failure (selecting very easy tasks) or to cognitive skills. Indeed, it is vital that training
prevent negative self-evaluations after failure (the programs aiming to increase actual perfor-
task was so difficult that failure has not to be attrib- mance, as well as motivation, do not overlook
uted to personal inadequacies). Thus, although the the strategies necessary for the task at hand.
behavior of failure-motivated individuals may This principle is congruent with Atkinson’s
seem strange and inappropriate from the perspec- model of cumulative achievement outlined
tive of the “improve one’s capabilities” directive, it above; its effects have already been demon-
is in fact adaptive and entirely functional from that strated in training programs designed to increase
of the “reduce threats to self-esteem” directive. the economic activities of small business entre-
Nonetheless, the failure-related directive remains preneurs (McClelland & Winter, 1969).
detrimental to the acquisition of knowledge and The research presented in the two following
skills. It is associated with defensive and sporadic sections shows how a change in the reference
achievement behavior and is incompatible with the norm used to evaluate achievement is associated
goal of increasing personal competence (see with marked changes in each of the three subpro-
Covington, 1992, for a clear account of how the cesses identified above.
280 J.C. Brunstein and H. Heckhausen
6.5.2 T
he Role of Reference Norms different phases of skill acquisition. In a study
in the Motivation Process conducted by Brackhane (1976), participants
were asked to evaluate their own performance at
McClelland et al. (1953) and Heckhausen (1963) a dart-throwing task. At first, they based their
defined achievement motivation as the striving to judgments on the characteristics of the task, i.e.,
meet standards of excellence. Yet they did not on the scores displayed on the rings of the target
specify which standard is used to evaluate an (criterion norm). As they gained more experi-
action outcome. Three such standards are the fol- ence, they developed a personal reference system
lowing (see Heckhausen, 1974) for assessing change in their performance (indi-
Each of these reference norms can also be vidual norm). With increasing practice, the crite-
applied to evaluate the performance of others. rion for a good outcome was shifted gradually
This is particularly relevant for occupations (e.g., upward. Finally, some participants inquired about
the teaching profession) involving the routine their cohorts’ performance, indicating that they
evaluation of others’ performance (see below). were interested in how their performance com-
The three reference norms are not mutually pared with that of others (social reference norm).
exclusive, but have been shown to take effect in The advantages of this sequence of reference-
norm application are clear (see also Zimmerman
& Kitsantas, 1997). At first, attention is focused
Reference Norms (Based on Dickhäuser on the task itself. People then begin to register
& Rheinberg, 2003) improvements in their performance and only start
to evaluate that performance in social compari-
1. Individual reference norms: son after gaining a good deal of practice. The
Individuals compare their perfor- reverse sequence of norm application could only
mance outcomes with previous outcomes lead to frustration and discourage people to per-
to determine whether their performance sist in their task-related efforts until they have
has improved, worsened, or remained acquired a new skill.
unchanged over time. The comparative The distinction between different reference
perspective is temporal change in an norms (or standards of excellence) was long
individual’s development (e.g., learning neglected in achievement motivation research
gains on a new type of task). (for an exception cf. Veroff, 1969), but has
2. Social reference norms: attracted increasing attention since the 1980s. It
Individuals compare their perfor- is no coincidence that researchers investigating
mance outcomes with those of others. motivational issues in the context of developmen-
The comparative perspective is the per- tal and educational psychology have played a
formance distribution within a social pioneering role here: Rheinberg (1980;
reference group (e.g., a student’s posi- Heckhausen & Rheinberg, 1980) in the German-
tion in a class). In a snapshot cross- speaking countries and Nicholls (1984a, 1984b,
sectional comparison, the individual’s 1989), Dweck (1986; Dweck & Elliot, 1983), and
current ranking on a certain task is Ames and Ames (1984) in the English-speaking
determined by relating it to the achieve- countries, to name just a few.
ments of relevant others. But how do the different reference norms have
3. Objective or criterial reference norms: an impact on the motivation process?
Performance outcomes are measured The primate of individual reference norms.
against absolute criteria inherent in the Individual reference norms occupy a preeminent
task itself. A solution may be correct or position among aspects promoting motivation.
incorrect; an intended outcome may be People assessing their own performance levels
achieved to a specifiable extent. on the basis of their previous achievements gen-
erally find that effort and persistence, on the one
6 Achievement Motivation 281
hand, covary with gradual improvements in per- more positive light, probably because this
formance, on the other. Moreover, the perfor- appraisal has less to do with social comparison
mance level attained tends to be in the (doing better than others) than with individual
intermediate range of (subjective) difficulty, learning gains (improving one’s own knowledge
which – according to the risk-taking model – is and skills). Learners who notice the progress they
maximally motivating. By contrast, comparison are making see the effects of their efforts at first
with social reference norms tells an individual hand and gain more enjoyment from the learning
only that he or she is better or worse than a cer- process (Jagacinski & Nicholls, 1987). This pat-
tain percentage of others and does not reflect tern of results has been observed in natural condi-
performance gains (assuming the reference tions (unaffected by outside influences), as well as
group to be making comparable progress). in intervention studies in which teachers were
Individual progress does not imply an improve- trained to apply individual reference norms, and
ment in relative ranking, as reflected in the high in training studies seeking to modify students’
stability of school grades. Individual reference self-evaluations. Transforming a social frame of
norms focus attention on improvements in per- reference in the classroom to an individual one (or
sonal performance and the effort made to achieve at least enriching it by aspects of intraindividual
learning gains. Effort is a factor that is under the comparison) creates a motivational climate that
voluntary control of the individual and for which has an impact on students’ self-evaluations, with
he or she can thus be held responsible. Social favorable effects on precisely those subprocesses
reference norms, on the other hand, focus atten- (goal setting, causal explanations, and achieve-
tion on a relative ranking – e.g., relative to the ment-related emotions) identified above as being
rest of a class – that tends to be relatively stable relevant to success-motivated achievement behav-
and that barely correlates with effort and persis- ior (Rheinberg et al., 2000).
tence. Social reference norms thus reflect differ- A study by Krampen (1987) showed that indi-
ences in ability. Especially when assessed in vidual reference norms have particularly positive
social comparison, ability is generally seen as a effects on the outcomes of weaker students.
determinant of achievement that is very difficult Mathematics teachers in 13 ninth and tenth grade
to influence in the short term. classes were trained to provide students with writ-
ten comments about their work, based on either
• Instructional experiments conducted by individual, social, objective (curriculum-based), or
Rheinberg and Krug (Rheinberg, 1980; no reference norms. The students were assigned at
Rheinberg & Krug, 2004; see also Rheinberg & random to one of the four reference-norm condi-
Engeser, 2010) provided strong evidence for the tions. After 6 months, findings showed significant
idea that individual reference norms, in terms of interactions between the baseline performance
both self-evaluations (student ratings) and other level (school grade) and the reference norm on
evaluations (teacher ratings), enhance motiva- which the teachers’ comments were based. As
tion to learn. School classes characterized by shown in Fig. 6.22, the expectancies of students
individual reference norms show higher levels with relatively poor achievements were highest
of hope for success, willingness to exert effort, when feedback was based on individual reference
and student responsibility. norms, whereas feedback based on social com-
parison was associated with a much more pessi-
Furthermore, individual reference norms are mistic view. The performance gains observed over
associated with more realistic levels of aspiration a 6-month interval were even more remarkable.
and performance expectations and with increased Here again, individual reference norms had the
effort attributions. Achievement-related affect is most favorable effect; social reference norms had
dominated by joy and pride rather than by shame none. The lower the student’s baseline perfor-
and fear of failure. Heckhausen (1975b) even mance, the more conducive an individual refer-
found that people evaluate their own abilities in a ence norm was to learning gains (Fig. 6.23).
282 J.C. Brunstein and H. Heckhausen
-0.5
-1.0
-1.5
-2.0
type and purpose of the evaluation. Teachers mean success motive of their students to be
with an individual reference-norm orientation quite substantial (r = 0.54). Rheinberg, Schmalt,
have proved to be much more flexible in this and Wasser (1978) found that the failure motive
respect, varying the reference norm applied was relatively pronounced in classes whose
according to the context of evaluation (e.g., teachers preferred social reference norms.
using objective and social norms when writing Interestingly, a longitudinal study by Rheinberg
report cards, but using individual norms in the (1980) showed that the introduction of individ-
context of student discussions and everyday ualized feedback led to a more pronounced
feedback). Teachers with a social reference- reduction in the initial level of failure motiva-
norm orientation have proved to be less flexible, tion in educationally disadvantaged students
applying a social frame of reference regardless who could barely compete with their class-
of the purpose of the evaluation (report cards, mates. The sample consisted of fifth graders
praise for good work, etc.). from the lowest track of the three-tier German
The reference norm applied in the classroom secondary system (Hauptschule). Students had
also provides a certain amount of insight into teach- been allocated to new classes at the beginning
ers’ causal attributions of student performance: of the school year. Half of the classes were
assigned a teacher who applied social reference
• Teachers with a social reference-norm orien- norms and the other half a teacher who applied
tation tend to ascribe success and failure at individual reference norms. Within each class,
school to stable, internal factors (e.g., ability) students were categorized into three groups
and to form stable expectations of student per- based on their scores on an intelligence test.
formance. They only reward achievements Figure 6.24 shows how the failure motive (mea-
that are above average. They set all students sured by Schmalt’s, 1976a, Achievement
the same tasks, and their praise and criticism Motive Grid) changed over the school year.
are dependent on class-average performance. Students exposed to individual reference norms
“Very good” students will be praised even if experienced a reduction in the failure motive,
they could have done better, as long as their and this effect was most pronounced among
performance is above the class average. students whose intelligence scores were in the
• Teachers with an individual reference- lowest tertile.
norm orientation attribute students’ learn- Corresponding patterns of results were
ing outcomes largely to effort. Their praise found for test and manifest anxiety. Moreover,
and criticism is dependent on learning gains, students exposed to an individual reference
regardless of a student’s absolute achieve- norm reported an increase in self-perceived
ment level. Progress is consistently rewarded ability, regardless of their intelligence. They
(by praise) and supported by informative were also much less likely than students
feedback. Moreover, these teachers adapt the exposed to a social reference norm to attribute
difficulty level of task assignments to their failure to a lack of ability. This finding has
students’ individual knowledge level. since been replicated in numerous further stud-
ies (Rheinberg & Krug, 2004).
In view of all these correlates, an individual
reference-norm orientation in the classroom • Individual reference norms in the classroom
can be expected to have positive effects on stu- are conducive to the development of students’
dents’ learning motivation. Indeed, empirical hope for success and reduce fear of failure.
evidence indicates this to be the case. For 16 These effects are not limited to the instruc-
third-grade classes, Brauckmann (1976) tional situation, but extend to the level of
reported the correlation between the individual personality dispositions as they develop and
reference-norm orientation of teachers and the become increasingly stable.
284 J.C. Brunstein and H. Heckhausen
Fig. 6.24 Developmental
change in the failure
motive (FM-1, Grid test)
35
during the fifth grade for
classes whose teachers
applied individual vs.
Family context and achievement motivation. (e.g., at school). As a result, the child experi-
Trudewind and Husarek (1979) presented some ences fear of failure and helplessness when con-
of the most compelling findings on the relation- fronted with scholastic demands, particularly
ship between family background and the devel- when outcomes are under par. This pattern of
opment of motive dispositions. The authors results is supported by the findings of Hodoka
investigated how mothers’ behavior in homework and Fincham (1995), who studied mother-child
situations was associated with the development interactions in students classified as “helpless”
of hope for success and fear of failure from first (teacher rating), again in homework situations.
to second grade. Mothers of children who feared Their findings confirm those of Trudewind and
failure were found to differ from mothers of chil- Husarek to the letter. A practical conclusion to
dren who were confident of success in the follow- be drawn from these insights is that interventions
ing respects: designed to combat fear of failure or to boost
hope for success must take both the school and
• They were more likely to apply social norms family contexts into consideration (for a parent
than individual and objective norms and training program of this kind, see Lund,
tended to expect too much of their children. Rheinberg, & Gladasch, 2001).
• They interfered in the homework process and
showed little respect for their child’s wishes or
6.5.3.2 The Achievement Motive
autonomy. and Preferences for Reference
• They criticized failure, but responded neu- Norms
trally to success. Extrapolating from these findings, it seems rea-
• They attributed failure to a lack of ability, but
sonable to assume that people scoring high on
success to the ease of the task. success motivation instinctively use individual
reference norms to evaluate their own
These findings clearly show that fear of fail- performance. There have been few investigations
ure, as described in Heckhausen’s (1975a) self- of this assumption, but at least three studies have
evaluation model, is transferred from the provided findings to support it. In a study with
(negative) model of the mother to the child. 124 students aged between 11 and 13, Rheinberg,
Failure-centered interactions may be internal- Duscha, and Michels (1980) found a significant
ized in the form of inner dialogs and thus affect correlation of r = 0.39 between hope for success
the child’s behavior in other situations as well (AM Grid; Schmalt, 1976a) and preference for an
6 Achievement Motivation 285
individual reference norm over a social reference motive was associated with mastery goals and the
norm in a motor skills game. Brunstein and failure motive with social comparison goals.
Hoyer (2002; see also Brunstein & Maier, 2005; Findings on the hierarchical model of motiva-
and Sect. 9.2.2 of this volume) took a different tion (see the excursus) correspond with the ideas
approach, but their pattern of results was similar. of Breckler and Greenwald (1986), who argued
In an experimental study involving a mental con- that achievement-motivated individuals, as
centration test, student participants were given defined by McClelland et al. (1953), have the
feedback on both their individual performance capacity to regulate their behavior autonomously.
gains (self-referenced feedback) and their rank- Achievement-motivated individuals strive con-
ing relative to the performances of other partici- stantly to improve their knowledge and skills,
pants (norm-referenced feedback). The applying their own standards of excellence, and
achievement motive was measured by means of with no need for social norms and feedback.
the TAT (nAchievement). Change in performance Against this background, it makes perfect sense
subsequent to the provision of feedback served as that de Charms et al. (1955) found striving for
the dependent variable. In this situation, the independence and low conformity to be close
achievement motive did not interact with norma- correlates of the achievement motive. Failure-
tive feedback but was highly responsive to self- motivated individuals, on the other hand, seem to
referenced feedback. As soon as their performance be hounded by concerns about the social evalua-
decreased below the level expected on the basis tion of their achievements and its implications.
of their previous performance, participants high Failure-motivated individuals are thus dependent
in achievement motivation redoubled their efforts on the recognition of others. For them, the striv-
and showed an immediate improvement in per- ing to achieve is a means to the end of gaining the
formance. Thrash and Elliot (2002) investigated acceptance and appreciation of the social
how success and failure motives, assessed by environment.
means of projective tests, are related to To the casual observer, these findings seem to
achievement-related goal orientations, assessed contradict a distinction that Nicholls (1984a,
by questionnaire measures. Student participants 1984b) made between two forms of achievement
were asked to state their goals for an upcoming motivation. Nicholls proposed the first form of
exam: achievement motivation to be activated in situa-
tions where the aim is to master a task, make a
• Outperforming other students (achievement- personal effort, and improve one’s performance.
approach goals) In these “task-involving” situations, ability is
• Avoiding failure (achievement-avoidance equated with the capacity to improve one’s per-
goals) sonal performance. In “ego-involving” situations,
• Mastering the tasks as well as possible (mas- in contrast, the main aim is to compare one’s abil-
tery goals) ity with that of others and to do as well as possi-
ble or, at the very least, to conceal one’s
Multiple regression analyses showed that weaknesses. There are obvious parallels between
success-motivated students tended to prefer mas- Nicholls’ distinction between task and ego
tery goals, whereas failure-motivated students involvement, on the one hand, and Rheinberg’s
pursued both achievement-approach goals and distinction between individual and social refer-
achievement-avoidance goals. The latter finding ence norms, on the other. These norms, along
reemphasizes the two sides of the failure- with the respective incentives (self-improvement
avoidance motive (active vs. passive coping with vs. demonstrating one’s superior abilities), are
failure). Findings were similar, though not identi- indeed key components of both forms of motiva-
cal, when questionnaires were used to assess the tional involvement (Butler, 1993). The potential
two achievement motives. Again, the success contradiction is that Nicholls assumed classical
286 J.C. Brunstein and H. Heckhausen
research questions that address the very core of distinguish satisfactorily between active and
the human striving for excellence and self- passive forms of coping with failure.
improvement. We conclude this chapter by high- Nevertheless, it is fortunate that this important
lighting four of those questions: distinction was detected and acknowledged
early in achievement motivation research.
1. Since the beginning of research on achieve- Little is known, however, about the validity of
ment motivation, questions regarding how to FF measures (TAT) or more precisely about
measure motives have stayed highly relevant. how this motive affects behavior depending
The TAT was criticized fairly early due to its on the situation. There is certainly no lack of
inadequate psychometric properties. However, ideas on how the existing findings are to be
there is no other instrument that has been interpreted and integrated into theories a pos-
developed with similar care (based on experi- teriori (Schultheiss & Brunstein, 2005). What
mental studies of motive arousal) and deliv- is now needed is the development of more
ered a comparable amount of insightful results elaborated theories enabling researchers to
(McClelland, 1985b). Recent findings sug- make accurate predictions about the occur-
gesting that reactions in the TAT can be rence and the behavioral effects of each form
explained with the help of stochastic test the- of avoidance. To this end, the focus of theories
ory are encouraging. Such findings have and empirical research must be shifted to the
recently been further elaborated and advanced connections between motivation and strate-
in order to include basic assumptions about gies for coping with failure.
the dynamics of motivation processes (Lang, 3. The complexity of the risk-taking model
2014). If all of its facets are taken into account, should not be underestimated. From a purely
a construct such as that of “the” achievement algebraic perspective, achievement motiva-
motive might be too complex to reduce it to tion is seen as a function of the product con-
only 1 or 2 numerical values. Although sisting of motive, incentive, and expectancy.
Heckhausen (1977a, 1977b) suggested that The tendencies to strive for success and to
the “summary construct” of achievement avoid failure are distinguished from each
motive should be divided into its single incen- other. A regression equation attempting to
tive- and expectancy-related components, this represent Atkinson’s formula in its entirety
has not yet been realized in the field of motive would require 14 different predictors: 6 first-
measurement. The finding, however, that indi- order predictors, 6 two-way interactions, and
rect (TAT) and direct (questionnaires) meth- 2 three-way interactions. The specific effects
ods for measuring achievement motivation are of all of these terms would have to be tested.
barely correlated and thus might not assess the Testing such an equation would require a very
same construct has received much more atten- large sample in order to ensure a reasonable
tion. It might give some comfort that the mea- test power. Studies on the risk-taking model
surement of other personality constructs that dramatically simplified this situation by run-
are difficult to access has suffered from simi- ning preliminary calculations. A dependence
lar problems (Bosson, Swann & Pennebaker, between incentive and expectation is created
2000). It would, however, be better if by means of an additional assumption (Ae =
Heckhausen’s recommendations were heeded 1 – We; this is not the case in other models of
and multidimensional tests for measuring achievement motivation, e.g., Eccles &
basic motives developed. Wigfield, 2002). Thus these two variables are
2. The status of fear of failure in the context of turned into a single one. In effect, the respec-
achievement motivation research remains tive research has almost exclusively focused
uncertain. The mere attempt to measure “the” on task difficulty as the incentive for achieve-
failure motive has proven problematic, at least ment behavior. In order to get a relatively easy
when using the TAT method, which does not measure for the “resulting” achievement
6 Achievement Motivation 289
motive, fear of failure is subtracted from the is often associated or even equated with com-
success motive a priori. This subtraction is petence motivation (see Koestner &
based on a fairly arbitrary convention for McClelland, 1990; Schultheiss & Brunstein,
which alternatives have been suggested (see 2005). For filling this gap the analysis of moti-
Covington & Roberts, 1994). Whether or not vation needs to be linked more closely to cog-
this convention is truly appropriate might nitive and emotional processes that occur
require further investigation (Do the interac- during an activity and interact with each other.
tion terms for success and failure tendencies In an analogous manner, studies investigating
really have opposite signs if they are tested the long-term relationship between achieve-
independently from each other?). If the vali- ment motivation and the development of com-
dation of the risk-taking model is to be placed petency would provide crucial insights.
on an empirically supported foundation, the One-shot studies that only look at the relation-
theoretically assumed interactions between ship between motivation and performance at a
the components of the model need to be tested single point in time can only deliver a momen-
more precisely. tary snapshot of how this interaction is seen
4. Very little is yet known about how achieve- from the outside. Further analyses of how
ment motives influence the acquisition of motivational influences affect learning and
knowledge. Achievement motivation research performance will be needed in the future.
has, for decades, focused on performance cri- Once more, Atkinson (1974a, 1974b) was a
teria and neglected to clarify the relationship pioneer in this regard. His work on the rela-
between motivation and learning. This neglect tionship of motivation and performance
is surprising, because achievement motivation deserves new empirical attention.
5. How can the relationship between the levels) is steeper in people high in success
amount of achievement-related content in motivation than in people low in success
textbooks and differences in educational motivation. This means that individuals
achievements in different German states high in success motivation are more sen-
be explained? sitive to achievement differences than
Due to their semantic connotations, less success-motivated individuals.
achievement-related statements (master- Accordingly, their satisfaction is more
ing something, improving on something) dependent on the level of achievement
and key words (diligent, successful) can attained.
arouse the achievement motive. In turn,
the aroused achievement motive mobi- 8. According to the predictions of the risk-
lizes resources, such as persistence and taking model, which difficulty levels do
effort, that are required to improve one’s success-motivated and failure-motivated
skills and master academic challenges. individuals prefer when choosing tasks?
Outline the actual empirical findings.
6. Which neuroendocrine features are found According to the risk-taking model,
in achievement-motivated individuals success-motivated individuals prefer mod-
when they work on tasks without being erately difficult tasks (Ps = 0.50), whereas
certain whether they can succeed? failure-motivated individuals avoid this
In this situation, a strong achievement range of difficulty, opting instead for
motive (assessed with the TAT) is associ- extremely difficult or extremely easy
ated with stress-response-dampening tasks. Empirical findings show that
effects. Saliva cortisol is comparatively low success-motivated individuals tend to pre-
in achievement-motivated individuals fer more difficult tasks falling below the
under stress compared to individuals with critical value of Ps = 0.50 predicted by the
a weaker achievement motive. risk-taking model. Failure-motivated indi-
Achievement-motivated individuals sub- viduals are more likely than success-moti-
jectively perceive an activity character- vated individuals to choose either extremely
ized by high difficulty or uncertainty of easy or extremely difficult tasks, but they do
success as a challenge that indicates an not purposely avoid the intermediate range
opportunity to master something. of difficulty.
7. How does the risk-taking model define the 9. How does the risk-taking model explain
valence of success and how is it atypical shifts in the level of aspiration in
measured? failure-motivated individuals?
The valence of success is defined as After failure on a simple task or suc-
the product of the success incentive and cess on a difficult task, the probability of
the success motive: Vs = Ms × Is. It is success approaches the critical level of
measured in terms of satisfaction judg- Ps = 0.50, i.e., precisely the range of dif-
ments for achievements at different diffi- ficulty that failure-motivated individuals
culty levels. The more anticipated seek to avoid. As a result, there are
satisfaction increases with the difficulty erratic shifts in the level of aspiration
of the task, the higher the valence of suc- toward the other end of the task diffi-
cess. The valence gradient for success culty scale (i.e., from very easy to very
(satisfaction across different difficulty difficult tasks or vice versa).
6 Achievement Motivation 291
10. Which experimental paradigm did Feather linear relationship between motivation
use to predict the level of persistence on and performance can only be assumed
the basis of the risk-taking model? for very easy, speed-dependent tasks.
Two tasks are administered in Due to the speed/accuracy trade-off,
Feather’s experimental paradigm. high levels of motivation on complex,
Participants are told that the first is either error-prone tasks can lead to decreased
difficult or easy, but it is in fact impossi- performance.
ble. Over repeated trials, the probability Second, the individual’s cognitive and
of success thus approaches Ps = 0.50 self-
regulatory skills are important. A
(“simple” task) or recedes from Ps = 0.50 lack of ability cannot be offset by high
(“difficult” task). The probability of suc- motivation. Self- regulatory skills are
cess on the second task is also stated. needed to ensure the optimal level of
Success-motivated individuals are motivation for the task.
expected to show more persistence when
the first task has a moderate probability 13. How does Covington explain the phenom-
of success and the second an extremely enon of overmotivation?
high or low probability of success. The By a combination of high success
reverse is expected to hold for failure- motivation and high failure motivation.
motivated individuals. For them, the more Covington calls individuals meeting this
extreme the difficulty level of the second description “overstrivers.” They invest a
task, and the nearer the probability of suc- great deal of time and effort, but because
cess on the first task to Ps = 0.50, the their approach tends to be ill-considered
more likely a switch to the second task and superficial, they remain ineffective.
becomes.
14. What is the function of motivational
11. How can the contradiction between the strength in Atkinson’s model of cumula-
risk-
taking model (Atkinson) and goal tive achievement?
theory (Locke) in terms of the relation- Motivation fulfills a dual function in
ship between task difficulty and perfor- this model. Together with ability, it influ-
mance levels be explained? ences the efficiency of task performance.
The risk-taking model is primarily Optimal, rather than maximum, motiva-
concerned with task choice. Tasks of tion facilitates good performance.
moderate difficulty are generally pre- Motivation also influences the time
ferred. Goal theory, in contrast, is con- invested in an activity. From a long-term
cerned with the realization of selected perspective, high motivation thus has a
goals. Effort expenditure is automatically positive effect on the acquisition of new
adjusted to task difficulty level (difficulty knowledge and skills.
law of motivation) until the point of max-
imum potential motivation is exceeded. 15. Why might it not be advisable to calculate
the “resultant motivation tendency” in
12. Name at least two factors that moderate the terms of the difference between success
strength of the relationship between achieve- and failure motivation?
ment motivation and task performance. Four arguments are relevant:
First, the strength of the relationship Success and failure motives represent
depends on the demands of the task. A theoretically independent constructs. By
292 J.C. Brunstein and H. Heckhausen
calculating difference scores, two dimen- one’s own previous performances; social
sions are artificially combined in a single reference norms, comparison of one’s
bipolar dimension of achievement performance with the performance of
motivation. others; and objective norms, task-
Difference scores do not reflect which immanent criteria of success, such as
variable is responsible for the predicted solving vs. not solving a task or attaining
effects. vs. failing to attain a given learning goal.
When difference scores are calculated,
individuals high in both motives have the 18. Individual reference norms are known to
same resultant score as people low in be conducive to achievement-motivated
both motives. behavior. What are the mediating pro-
Failure motivation does not always cesses involved in this relationship?
undermine the success tendency; it can Effort attributions of success and fail-
also facilitate proactive approaches to ure: individual reference norms empha-
coping with failure. size that the level of achievement is
contingent on the amount of effort
16. What is the role of achievement-related invested.
affect in Heckhausen’s self-evaluation Realistic goal setting: the goals set are
model of achievement motivation? based on individual ability or individual
It reinforces the behavioral directives learning trajectories.
that govern success-related vs. failure- A sense of achievement and progress:
related behavior: to increase competence weaker students, in particular, experience
in the case of success motivation and to more success when exposed to individual
protect self-esteem in the case of failure than to social reference norms. The result
motivation. Causal attributions provide the is increased pride, which in turn rein-
link between performance outcomes and forces feelings of competence and
the affective reactions of success-moti- efficacy.
vated vs. failure- motivated individuals.
Failure-motivated individuals avoid chal- 19. Which characteristics of the mother-child
lenges in order to protect their self-esteem interaction are associated with the devel-
against detrimental effects associated with opment of a strong failure motive in ele-
the attribution of failure experiences to mentary school children?
internal and stable factors (e.g., lack of In a homework situation, Trudewind
ability). and Husarek (1979) identified four such
In contrast, success-motivated indi- characteristics:
viduals prefer challenging tasks because
their attributions are conducive to self- Using social reference norms
esteem and enhanced feelings of compe- Expecting too much of the child and
tence. Even when the ratio of successes to having unrealistically high goals and
failures is balanced, the affective balance expectations
remains positive (with pride outweighing Attributing failure to a lack of ability
shame) for this group of individuals. Criticizing failure and ignoring success
17. Which reference norms can be used to 20. How do avoidance goals inhibit achieve-
evaluate a performance outcome? ment and enjoyment of learning?
Individual reference norms, temporal Avoidance goals tend not to have clear
comparison of one’s performance with criteria; progress on such ill-defined goals
6 Achievement Motivation 293
Covington, M. V., & Roberts, B. W. (1994). Self-worth Elliot, A. J., & Church, M. (1997). A hierarchical model
and college achievement: Motivational and personal- of approach and avoidance achievement motivation.
ity correlates. In P. R. Pintrich, D. R. Brown, & C. E. Journal of Personality and Social Psychology, 72,
Weinstein (Eds.), Student motivation, cognition, and 218–232.
learning (pp. 157–188). Hillsdale, NJ: Erlbaum. Elliot, A. J., & Haraciewicz, J. (1996). Approach and
Cronbach, L. J. (1990). Essentials of psychological testing avoidance achievement goals and intrinsic motivation:
(5th ed.). New York: Harper and Row. A mediational analysis. Journal of Personality and
Darwin, C. (1872). The expression of the emotions in Social Psychology, 70, 461–475.
man and animals. London, UK: John Murray. (1965, Elliot, A. J., & McGregor, H. A. (2001). A 2×2 achieve-
Chicago: University of Chicago Press). ment goal framework. Journal of Personality and
De Castella, K., Byrne, D., & Covington, M. (2013). Social Psychology, 80, 501–519.
Unmotivated or motivated to fail? A cross-cultural Engeser, S., Rheinberg, F., & Möller, M. (2009).
study of achievement motivation, fear of failure, Achievement motive imagery in German schoolbooks:
and student engagement. Journal of Educational A pilot study testing McClelland’s hypothesis. Journal
Psychology, 105, 861–880. of Research in Personality, 43, 110–113.
DeCharms, R., & Moeller, G. H. (1962). Values expressed Engeser, S., Hollricher, I., & Baumann, N. (2013). The
in American children’s readers: 1800–1950. Journal stories children’s books tell us: Motive-related imag-
of Abnormal and Social Psychology, 64, 136–142. eries in children’s books and their relation to academic
DeCharms, R., Morrison, H. W., Reitman, W., & performance and crime rates. Journal of Research in
McClelland, D. C. (1955). Behavioral correlates Personality, 47, 421–426.
of directly and indirectly measured achievement Engeser, S., Euen, W., & Bos, B. (2015).
motivation. In D. C. McClelland (Ed.), Studies Leistungsthematischer Gehalt von Schulbüchern und
in motivation (pp. 414–423). New York, NY: Bildungsleistung in der Grundschule. Zeitschrift für
Appleton-Century-Crofts. Pädagogische Psychologie, 29, 65–75.
Dickhäuser, O., & Rheinberg, F. (2003). Engeser, S., Baumann, N., & Baum, I. (2016). Schoolbook
Bezugsnormorientierung: Erfassung, Probleme, texts: Behavioral achievement in math and language.
Perspektiven. In J. Stiensmeier-Pelster & F. Rheinberg PLoS One, 11(3).
(Eds.), Diagnostik von Motivation und Selbstkonzept Entin, E. E. (1974). Effects of achievement-oriented and
(pp. 41–55). Göttingen, Germany: Hogrefe. affiliative motives on private and public performance.
Dweck, C. S. (1986). Motivational processes affect- In J. W. Atkinson & J. O. Raynor (Eds.), Motivation
ing learning. American Psychologist, 41, and achievement (pp. 219–236). Washington, DC:
1040–1048. Winston.
Dweck, C. S., & Elliott, E. S. (1983). Achievement moti- Entwisle, D. R. (1972). To dispel fantasies about
vation. In E. M. Hetherington (Ed.), Socialization, fantasy- based measures of achievement motivation.
personality, and social development (pp. 643–691). Psychological Bulletin, 77, 377–391.
New York, NY: Wiley. Ericsson, K. A. (Ed.). (1996). The road to excellence.
Eccles, J. S., & Wigfield, A. (2002). Motivational beliefs, Mahwah, NJ: Erlbaum.
values, and goals. Annual Review of Psychology, 53, Feather, N. T. (1959). Success probability and choice
109–132. behavior. Journal of Experimental Psychology, 58,
Eccles, J. S., Wigfield, A., Harold, R., & Blumenfeld, 257–266.
P. B. (1993). Age and gender differences in children’s Feather, N. T. (1961). The relationship of persistence at a
self- and task perceptions during elementary school. task to expectation of success and achievement related
Child Development, 64, 830–847. motives. Journal of Abnormal and Social Psychology,
Eccles, J. S., Wigfield, A., & Schiefele, U. (1998). 63, 552–561.
Motivation to succeed. In W. Damon & N. Eisenberg Feather, N. T. (1962). The study of persistence.
(Eds.), Handbook of child psychology (Vol. Bd. 3, 5. Psychological Bulletin, 59, 94–115.
Aufl. ed., pp. 1017–1095). New York, NY: Wiley. Feather, N. T. (1963). The relationship of expectation
Eibl-Eibesfeldt, I. (1984). Die Biologie des menschlichen of success to reporter probability, task structure and
Verhaltens. Grundriss der Humanethologie. München, achievement-related motivation. Journal of Abnormal
Germany: Piper. and Social Psychology, 66, 231–238.
Elliot, A. J. (1997). Integrating the “classic” and “con- Feather, N. T. (1966). Effects of prior success and fail-
temporary” approaches to achievement motivation: ure on expectations of success and subsequent perfor-
A hierarchical model of achievement motivation. In mance. Journal of Personality and Social Psychology,
M. Maehr & P. Pintrich (Eds.), Advances in motivation 3, 287–298.
and achievement (Vol. 10, pp. 243–279). Greenwich, Feather, N. T. (1967). Valence of outcome and expectation
CT: JAI Press. of success in relation to task difficulty and perceived
Elliot, A. J. (1999). Approach and avoidance motivation locus of control. Journal of Personality and Social
and achievement goals. Educational Psychologist, 34, Psychology, 7, 372–386.
169–189. Festinger, L. (1954). A theory of social comparison pro-
cesses. Human Relations, 7, 117–140.
296 J.C. Brunstein and H. Heckhausen
Fineman, S. (1977). The achievement motive construct Greenwald, A. G., Banaji, M. R., Rudman, L. A.,
and its measurement: Where are we now? British Farnham, S. D., Nosek, B. A., & Mellott, D. S.
Journal of Psychology, 68, 1–2. (2002). A unified theory of implicit attitudes, stereo-
Fisch, R., & Schmalt, H.-D. (1970). Vergleich von TAT und types, self-esteem, and self-concept. Psychological
Fragebogendaten der Leistungsmotivation. Zeitschrift Review, 109, 3–25.
für Experimentelle und Angewandte Psychologie, 17, Gruber, N., & Kreuzpointner, L. (2013). Measuring the
608–633. reliability of picture story exercises like the TAT. PLoS
Fodor, E. M., & Carver, R. A. (2000). Achievement and One, 8(11).
power motives, performance feedback, and creativity. Haber, R. N., & Alpert, R. (1958). The role of situation and
Journal of Research in Personality, 34, 380–396. picture cues in projective measurement of the achieve-
French, E. G. (1955). Some characteristics of achieve- ment motive. In J. W. Atkinson (Ed.), Motives in fan-
ment motivation. Journal of Experimental Psychology, tasy, action, and society (pp. 644–663). Princeton, NJ:
50, 232–236. Van Nostrand.
French, E. G. (1958a). Development of a measure of com- Halisch, F. (1986). Operante und respondente Verfahren
plex motivation. In J. W. Atkinson (Ed.), Motives in zur Messung des Leistungsmotivs. München,
fantasy, action, and society (pp. 242–248). Princeton, Germany: Max-Planck-Institut für psychologische
NJ: Van Nostrand. Forschung.
French, E. G. (1958b). Effects of the interaction of moti- Halisch, F., & Heckhausen, H. (1988). Motive-dependent
vation and feedback on task performance. In J. W. vs. ability-dependent valence functions for success
Atkinson (Ed.), Motives in fantasy, action, and society and failure. In F. Halisch & J. van den Bercken (Eds.),
(pp. 400–408). Princeton, NJ: Van Nostrand. Intentional perspectives on achievement and task moti-
French, E. G., & Lesser, G. S. (1964). Some characteris- vation. Lisse, The Netherlands: Swets & Zeitlinger.
tics of the achievement motive in women. Journal of Hall, J. L., Stanton, S. J., & Schultheiss, O. C. (2010).
Abnormal and Social Psychology, 68, 119–128. Biopsychological and neural processes of implicit
Freud, S. (1952). Die Abwehrneuropsychosen. (GW, Bd. motivation. In O. C. Schultheiss & J. C. Brunstein
I, 1894). Frankfurt, Germany: Fischer. (Eds.), Implicit motives (pp. 279–307). New York, NY:
Frey, R. S. (1984). Does n-achievement cause economic Oxford University Press.
development? A cross-lagged panel analysis of the Hamilton, J. O. (1974). Motivation and risk-taking behav-
McClelland thesis. Journal of Social Psychology, ior. Journal of Personality and Social Psychology, 29,
122, 67–70. 856–864.
Fries, S. (2002). Wollen und Können. Münster, Germany: Heckhausen, H. (1960). Die Problematik des
Waxmann. Projektionsbegriffs und die Grundlagen und
Fries, S., Lund, B., & Rheinberg, F. (1999). Lässt sich Grundannahmen des Thematischen Auffassungstests.
das Training induktiven Denkens durch gleichze- Psychologische Beiträge, 5, 53–80.
itige Motivförderung optimieren? Zeitschrift für Heckhausen, H. (1963). Hoffnung und Furcht in der
Pädagogische Psychologie, 13, 37–49. Leistungsmotivation. Meisenheim, Germany: Hain.
Frijda, N. H. (1986). The emotions. Cambridge, UK: Heckhausen, H. (1964). Über die Zweckmäßigkeit eini-
Cambridge University Press. ger Situationsbedingungen bei der inhaltsanalytischen
Fyans, L. J., Salili, M., Maehr, M. L., & Desai, K. A. Erfassung der Motivation. Psychologische Forschung,
(1983). A cross-cultural exploration into the mean- 27, 244–259.
ing of achievement. Journal of Personality and Social Heckhausen, H. (1968). Achievement motive research:
Psychology, 44, 1000–1013. Current problems and some contributions toward a
Geppert, U., & Heckhausen, H. (1990). Ontogenese general theory of motivation. In W. J. Arnold (Ed.),
der Emotion. In K. R. Scherer (Ed.), Enzyklopädie Nebraska symposium on motivation (pp. 103–174).
der Psychologie: Psychologie der Emotion (Vol. IV, Lincoln, NE: University of Nebraska Press.
pp. 115–213). Göttingen, Germany: Hogrefe. Heckhausen, H. (1969). Allgemeine Psychologie in
Gjesme, T. (1971). Motive to achieve success and motive Experimenten. Göttingen, Germany: Hogrefe.
to avoid failure in relation to school performance for Heckhausen, H. (1972). Die Interaktion der
pupils of different ability levels. Scandinavian Journal Sozialisationsvariablen in der Genese des
of Educational Research, 15, 81–99. Leistungsmotivs. In C. F. Graumann (Ed.), Handbuch
Gjesme, T., & Nygard, R. (1970). Achievement-related der Psychologie (Vol. Bd. 7/2, pp. 955–1019).
motives: Theoretical considerations and construction Göttingen, Germany: Hogrefe.
of a measuring instrument. Unpublished manuscript, Heckhausen, H. (1974). Leistung und Chancengleichheit.
University of Oslo. Göttingen, Germany: Hogrefe.
Greenwald, A. G., McGhee, D. E., & Schwartz, J. L. Heckhausen, H. (1975a). Fear of failure as a self-
K. (1998). Measuring individual differences in reinforcing motive system. In I. G. Sarason &
implicit cognition: The implicit association test. C. Spielberger (Eds.), Stress and anxiety (Vol. II,
Journal of Personality and Social Psychology, 74, pp. 117–128). Washington, DC: Hemisphere.
1464–1480.
6 Achievement Motivation 297
Heckhausen, H. (1975b). Effort expenditure, aspira- Hesse, F. W., Spies, K., & Lüer, G. (1983). Einfluß moti-
tion level and self-evaluation before and after unex- vationaler Faktoren auf das Problemlöseverhalten
pected performance shifts. Unpublished manuscript, im Umgang mit komplexen Problemen. Zeitschrift
Ruhr University, Institute of Psychology, Bochum, für experimentelle und angewandte Psychologie, 30,
Germany. 400–424.
Heckhausen, H. (1977a). Achievement motivation and Higgins, R. L., Snyder, C. R., & Berglas, S. (Eds.). (1990).
its constructs: A cognitive model. Motivation and Self-handicapping: The paradox that isn’t. New York,
emotion (Vol. 1, 4, pp. 283–329). New York, NY: NY: Plenum.
Plenum. Hillgruber, A. (1912). Fortlaufende Arbeit und
Heckhausen, H. (1977b). Motivation: Willensbetätigung. Leipzig, Germany: Quelle &
Kognitionspsychologische Aufspaltung eines sum- Meyer.
marischen Konstrukts. Psychologische Rundschau, Hodoka, A., & Fincham, F. D. (1995). Origins of chil-
28, 175–189. dren’s helpless and mastery patterns in the family.
Heckhausen, H. (1980). Motivation und Handeln. Journal of Educational Psychology, 87, 375–385.
Lehrbuch der Motivationspsychologie. Heidelberg, Hofer, J. (2010). Research on implicit motives across cul-
Germany: Springer. tures. In O. C. Schultheiss & J. C. Brunstein (Eds.),
Heckhausen, H. (1984). Emergent achievement behav- Implicit motives (pp. 433–467). New York, NY:
ior: Some early developments. In J. Nicholls (Ed.), Oxford University Press.
Advances in achievement motivation (pp. 1–32). Hofer, J., & Chasiotis, A. (2004). Methodological con-
Greenwich, CT: JAI Press. siderations of applying a TAT-type picture-story-test
Heckhausen, H. (1986). Why some time out might benefit in cross-cultural research: A comparison of German
achievement motivation research. In J. H. L. van den and Zambian adolescents. Journal of Cross-Cultural
Bercken, E. E. J. DeBruyn, & T. C. M. Bergen (Eds.), Psychology, 35, 224–241.
Achievement and task motivation (pp. 7–39). Lisse, Hofer, J., Busch, H., Bender, M., Ming, L., & Hagemeyer,
The Netherlands: Swets & Zeitlinger. B. (2015). Arousal of achievement motivation among
Heckhausen, H. (1987). Perspektiven einer Psychologie student samples in three different cultural contexts:
des Wollens. In H. Heckhausen, P. M. Gollwitzer, & Self and social standards of evaluation. Journal of
F. E. Weinert (Eds.), Jenseits des Rubikon: Der Wille Cross-Cultural Psychology, 41, 758–775.
in den Humanwissenschaften (pp. 121–142). Berlin, Hogenraad, R. (2005). What the words of war can tell us
Germany: Springer. about the risk of war. Peace and Conflict: Journal of
Heckhausen, H., & Kuhl, J. (1985). From wishes to Peace Psychology, 11, 137–151.
action: The dead ends and short cuts on the long Horner, M. S. (1974a). Performance of men in noncom-
way to action. In M. Frese & L. Sabini (Eds.), Goal- petitive and interpersonal competitive achievement-
directed behavior: Psychological theory and research oriented situations. In J. W. Atkinson & J. O. Raynor
on action (pp. 134–160., 367–395). Hillsdale, NJ: (Eds.), Motivation and achievement (pp. 237–254).
Erlbaum. Washington, DC: Winston.
Heckhausen, H., & Rheinberg, F. (1980). Horner, M. S. (1974b). The measurement and behav-
Lernmotivation im Unterricht, erneut betrachtet. ioral implications of fear of success in women. In
Unterrichtswissenschaft, 8, 7–47. J. W. Atkinson & J. O. Raynor (Eds.), Motivation
Heckhausen, H., & Roelofsen, I. (1962). Anfänge and achievement (pp. 91–117). Washington, DC:
und Entwicklung der Leistungsmotivation: (I) Im Winston.
Wetteifer des Kleinkindes. Psychologische Forschung, Hyland, M. E., Curtis, C., & Mason, D. (1985). Fear of
26, 313–397. success: Motive and cognition. Journal of Personality
Heckhausen, H., & Strang, H. (1988). Efficiency under and Social Psychology, 49, 1669–1677.
maximal performance demands: Exertion control, an Jacobs, B. (1958). A method for investigating the cue
individual-difference variable? Journal of Personality characteristics of pictures. In J. W. Atkinson (Ed.),
and Social Psychology, 55, 489–498. Motives in fantasy, action, and society (pp. 617–629).
Heckhausen, H., Schmalt, H.-D., & Schneider, K. (1985). Princeton, NJ: Van Nostrand.
Achievement motivation in perspective. New York, Jagacinski, C. M., & Nicholls, J. G. (1987). Competence
NY: Academic. and affect in task involvement and ego involvement:
Helmke, A., & Weinert, F. E. (1997). Bedingungsfaktoren The impact of social comparison information. Journal
schulischer Leistungen. In F. E. Weinert (Ed.), of Educational Psychology, 79, 107–114.
Psychologie des Unterrichts und der Schule, Jones, E. E., Rock, L., Shaver, K. G., Goethals, G. R.,
Enzyklopädie der Psychologie, Themenbereich & Ward, L. M. (1968). Pattern of performance and
D, Serie I: Pädagogische Psychologie (Vol. 3, ability attribution: An unexpected primacy effect.
pp. 71–176). Göttingen, Germany: Hogrefe. Journal of Personality and Social Psychology, 10,
Henle, M. (1944). The influence of valence on substitu- 317–340.
tion. Journal of Psychology, 17, 11–19. Jopt, U.-J. (1974). Extrinsische Motivation und
Hermans, H. J. M. (1970). A questionnaire measure Leistungsverhalten. Unveröffentlichte Dissertation,
of achievement motivation. Journal of Applied RUB, Fakultät für Philosophie, Pädagogik,
Psychology, 54, 353–363. Psychologie, Bochum, Germany.
298 J.C. Brunstein and H. Heckhausen
Karabenick, S. A. (1972). Valence of success and failure and challenges for future research (pp. 111–169).
as a function of achievement motives and locus of con- New York, NY: Academic Press.
trol. Journal of Personality and Social Psychology, 21, Kuhl, J. (2001). Motivation und Persönlichkeit. Die
101–110. Interaktion psychischer Systeme. Göttingen, Germany:
Karabenick, S. A., & Yousseff, Z. I. (1968). Performance Hogrefe.
as a function of achievement level and perceived dif- Kuhl, J., & Scheffer, D. (1999). Der operante Multi-
ficulty. Journal of Personality und Social Psychology, Motiv- Test (OMT): Manual. Osnabrück, Germany:
10, 414–419. Universität Osnabrück.
Kaufmann, I. C., & Rosenblum, L. A. (1969). The reaction Kukla, A. (1972a). Attributional determinants of
of separation from mother on the emotional behavior achievement-related behavior. Journal of Personality
of infant monkeys. Animals of the New York Academy and Social Psychology, 21, 166–174.
of Science, 159, 681–695. Kukla, A. (1972b). Foundations of an attributional theory
Kirschbaum, C., Pirke, K. M., & Hellhammer, D. H. of performance. Psychological Review, 79, 454–470.
(1993). The ‘Trier Social Stress Test’ – A tool for Kukla, A. (1978). An attributional theory of choice. In
investigating psychobiological stress responses in a L. Berkowitz (Ed.), Advances in experimental social
laboratory setting. Neuropsychobiology, 28, 76–81. psychology (Vol. 11, pp. 113–144). New York, NY:
Klinger, E. (1967). Modelling effects on achieve- Academic.
ment imagery. Journal of Personality and Social Lang, J. W. B. (2014). A dynamic Thurstonian item
Psychology, 7, 49–62. response theory of motive expression in the Picture
Koestner, R., & McClelland, D. C. (1990). Perspectives on Story Exercise: Solving the internal consistency para-
competence motivation. In L. Pervin (Ed.), Handbook dox of the PSE. Psychological Review, 121, 481–500.
of personality theory and research (pp. 527–548). Langens, T. A., & Schmalt, H.-D. (2008). Motivational
New York, NY: Guilford. traits: New directions and measuring motives with the
Koestner, R., Weinberger, J., & McClelland, D. C. (1991). Multi-Motive-Grid (MMG). In G. Boyle, G. Matthews,
Task-intrinsic and social-extrinsic sources of arousal & D. Saklowske (Eds.), The Sage handbook of person-
for motives assessed in fantasy and self-report. Journal ality theory and assessment, Vol. 1: Personality theo-
of Personality, 59, 57–82. ries and models (pp. 523–544). Thousand Oaks, CA:
Krampen, G. (1987). Differential effects of teacher Sage.
comments. Journal of Educational Psychology, 79, van Lawick-Godall, J. (1968). The behavior of free-living
137–146. chimpanzees in the Gombe Stream Area. Animal
Krau, E. (1982). Motivational feedback loops in the Behavior Monographs, 1, 161–312.
structure of action. Journal of Personality and Social Lewin, K. (1926a). Untersuchungen zur Handlungs- und
Psychology, 43, 1030–1040. Affekt-Psychologie, I: Vorbemerkungen über die psy-
Krohne, H. W. (1988). Erziehungsstilforschung: Neuere chischen Kräfte und Energien und über die Struktur
theoretische Ansätze und empirische Befunde. der Seele. Psychologische Forschung, 7, 294–329.
Zeitschrift für Pädagogische Psychologie, 2, 157–172. Lewin, K. (1926b). Untersuchungen zur Handlungs- und
Kubinger, K. D., & Ebenhöh, H. (1996). Arbeitshaltungen Affekt-Psychologie, II.: Vorsatz, Wille und Bedürfnis.
(AHA): Objektiver Persönlichkeitstest. Mödling, Psychologische Forschung, 7, 330–385.
Austria: Schuhfried. Liebert, R. M., & Morris, L. W. (1967). Cognitive and emo-
Kubinger, K. D., & Litzenberger, M. (2003). Zur tional components of text anxiety: A distinction and
Validität der Objektiven Persönlichkeits-Test-Batterie some initial data. Psychological Reports, 20, 975–978.
“Arbeitshaltungen”. Zeitschrift für Differentielle und Lissner, K. (1933). Die Entspannung von Bedürfnissen
Diagnostische Psychologie, 24, 119–133. durch Ersatzhandlungen. Psychologische Forschung,
Kuhl, J. (1977). Miß- und prozeßtheoretische Analysen 18, 218–250.
einiger Person- und Situationsparameter der Litwin, G. H. (1966). Achievement motivation, expec-
Leistungsmotivation. Bonn, Germany: Bouvier. tancy of success, and risk-taking behavior. In
Kuhl, J. (1978a). Situations-, reaktions- und personbezo- J. W. Atkinson & N. T. Feather (Eds.), A theory of
gene Konsistenz des Leistungsmotivs bei der Messung achievement behavior (pp. 103–115). New York,
mittels des Heckhausen TAT. Archiv für Psychologie, NY: Wiley.
130, 37–52. Locke, E. A. (1968). Toward a theory of task motivation
Kuhl, J. (1978b). Standard setting and risk preference: and incentives. Organizational Behavior and Human
An elaboration of the theory of achievement motiva- Performance, 3, 157–189.
tion and an empirical test. Psychological Review, 85, Locke, E. A. (1975). Personnel attitudes and motivation.
239–248. Annual Review of Psychology, 26, 457–480.
Kuhl, J. (1983). Motivation, Konflikt und Locke, E. A., & Latham, G. P. (1990). A theory of goal
Handlungskontrolle. Berlin, Germany: Springer. setting and task performance. Englewood Cliffs, NJ:
Kuhl, J. (2000). A functional-design approach to motiva- Prentice Hall.
tion and volition: The dynamics of personality sys- Locke, E., & Latham, G. (Eds.). (2012). New develop-
tems interactions. In M. Boekaerts, P. R. Pintrich, ments in goal setting and task performance. New York,
& M. Zeidner (Eds.), Self-regulation: Directions NY: Taylor & Francis.
6 Achievement Motivation 299
Locke, E. A., & Shaw, K. N. (1984). Atkinson’s McClelland, D. C. (1985a). How motives, skills, and val-
inverse-U curve and the missing cognitive variables. ues determine what people do. American Psychologist,
Psychological Reports, 55, 403–412. 41, 812–825.
Lowell, E. L. (1950). A methodological study of projectively McClelland, D. C. (1985b). Human motivation. Glenview,
measured achievement motivation. Unveröffentlichte IL: Scott, Foresman.
Magisterarbeit, Wesleyan University. McClelland, D. C. (1995). Achievement motivation in
Lowell, E. L. (1952). The effect of need for achievement relation to achievement–related recall, performance,
on learning and speed of performance. Journal of and urine flow, a marker associated with release of
Psychology, 33, 31–40. vasopressin. Motivation and Emotion, 19, 59–76.
Lund, B., Rheinberg, F., & Gladasch, U. (2001). McClelland, D. C., & Franz, C. E. (1992). Motivational
Ein Elterntraining zum motivationsförderlichen and other sources of work accomplishment in mid-
Erziehungsverhalten in Leistungskontexten. Zeitschrift life: A longitudinal study. Journal of Personality, 60,
für Pädagogische Psychologie, 15, 130–142. 680–707.
Lundy, A. (1985). The reliability of the thematic apper- McClelland, D. C., & Liberman, A. M. (1949). The effects
ception test. Journal of Personality Assessment, 49, of need for achievement on recognition of need related
141–145. words. Journal of Personality, 18, 236–251.
Lundy, A. (1988). Instructional set and thematic appercep- McClelland, D. C., & Winter, D. G. (1969). Motivating
tion test validity. Journal of Personality Assessment, economic achievement. New York, NY: Free.
52, 309–320. McClelland, D. C., Clark, R. A., Roby, T. B., & Atkinson,
Mahler, W. (1933). Ersatzhandlungen verschiedenen J. W. (1949). The projective expression of need for
Realitätsgrades. Psychologische Forschung, 18, achievement on thematic apperception. Journal of
27–89. Experimental Psychology, 39, 242–255.
Mandler, G., & Sarason, S. B. (1952). A study of anxi- McClelland, D. C., Atkinson, J. W., Clark, R. A., & Lowell,
ety and learning. Journal of Abnormal and Social E. L. (1953). The achievement motive. New York, NY:
Psychology, 47, 166–173. Appleton-Century-Crofts.
Marsh, H. W., Byrne, B. M., & Shavelson, R. J. (1988). McClelland, D. C., Koestner, R., & Weinberger, J. (1989).
A multifaceted academic self-concept: Its hierarchi- How do self-attributed and implicit motives differ?
cal structure and its relation to academic achieve- Psychological Review, 96, 690–702.
ment. Journal of Educational Psychology, 80, McKeachie, W. J. (1961). Motivation, teaching methods,
366–380. and college learning. In M. R. Jones (Ed.), Nebraska
Martin, A. J., Marsh, H. W., & Debus, R. L. (2001). A symposium on motivation (pp. 111–142). Lincoln, NE:
quadripolar need achievement representation of self- University of Nebraska Press.
handicapping and defensive pessimism. American Mehrabian, A. (1969) Measures of achieving tendency.
Educational Research Journal, 38, 583–610. Educational and Psychological Measurement, 29,
Martire, J. C. (1956). Relationship between the self- 445–451.
concept and differences in the strength and generality Meyer, W.-U. (1972). Überlegungen zur Konstruktion
of achievement motivation. Journal of Personality, 24, eines Fragebogens zur Erfassung von Selbstkonzepten
364–375. der Begabung. Unveröffentlichtes Manuskript, RUB,
McClelland, D. C. (1958). Risk taking in children with Psychologisches Institut, Bochum, Germany.
high and low need for achievement. In J. W. Atkinson Meyer, W.-U. (1973). Anstrengungsintention in
(Ed.), Motives in fantasy, action, und society (pp. 306– Abhängigkeit von Begabungseinschätzung und
321). Princeton, NJ: Van Nostrand. Aufgabenschwierigkeit. Archiv für Psychologie, 125,
McClelland, D. C. (1961). The achieving society. 245–262.
Princeton, NJ: Van Nostrand. Meyer, W.-U. (1984a). Das Konzept von der eigenen
McClelland, D. C. (1976). New introduction. In D. C. Begabung. Bern, Germany: Huber.
McClelland (Ed.), The achieving society. New York, Meyer, W.-U. (1984b). Das Konzept von der eigenen
NY: Irvington. Begabung: Auswirkungen, Stabilität und voraus-
McClelland, D. C. (1980). Motive dispositions: The mer- laufende Bedingungen. Psychologische Rundschau,
its of operant and respondent measures. In L. Wheeler 35, 136–150.
(Ed.), Review of personality and social psychology Meyer, W.-U. (1987). Perceived ability and achievement-
(Vol. 1, pp. 10–41). Beverly Hills, CA: Sage. related behavior. In F. Halisch & J. Kuhl (Eds.),
McClelland, D. C. (1984a). Motives as sources of long- Motivation, intention, and volition (pp. 73–86). Berlin,
term trends in life and health. In D. C. McClelland Germany: Springer.
(Ed.), Motives, personality, and society (pp. 343–364). Meyer, W.-U., & Starke, E. (1982). Seeking information
New York, NY: Praeger. about one’s own ability in relation to self- concept
McClelland, D. C. (1984b). The empire-building motiva- of ability: A field study. Personality and Social
tional syndrome. In D. C. McClelland (Ed.), Motives, Psychology Bulletin, 8, 501–507.
personality, und society: Selected papers (pp. 147– Meyer, W.-U., Heckhausen, H., & Kemmler, L. (1965).
174). New York, NY: Praeger. Validierungskorrelate der inhaltsanalytisch erfaßten
300 J.C. Brunstein and H. Heckhausen
Leistungsmotivation guter und schwacher Schüler des Nygard, R. (1982). Achievement motives and individ-
dritten Schuljahres. Psychologische Forschung, 28, ual differences in situational specificity of behavior.
301–328. Journal of Personality and Social Psychology, 43,
Meyer, W.-U., Niepel, M., & Engler, U. (1987). 319–327.
Erwartung, Affekt und Attribution: Untersuchungen O’Connor, P., Atkinson, J. W., & Horner, M. S. (1966).
zur Beziehung zwischen Erwartung und Anreiz Motivational implications of ability grouping in
und zur Attributionsabhängigkeit von Affekten. schools. In J. W. Atkinson & N. T. Feather (Eds.),
Psychologische Beiträge, 29, 227–258. A theory of achievement motivation (pp. 231–248).
Mierke, K. (1955). Wille und Leistung. Göttingen, New York, NY: Wiley.
Germany: Hogrefe. Orpen, C. (1983). Risk-taking attitudes among Indian,
Morgan, C. D., & Murray, H. A. (1935). A method for United States, and Japanese managers. Journal of
investigating fantasies: The thematic appercep- Social Psychology, 120, 283–284.
tive test. Archives of Neurological Psychiatry, 34, Pang, J. S. (2006). A revised content-coding measure
289–306. for hope of success (HS) and fear of failure (FF).
Moulton, R. W. (1958). Notes for a projective measure Unpublished dissertation. University of Michigan,
for fear of failure. In J. W. Atkinson (Ed.), Motives in Ann Arbor, MI.
fantasy, action, and society (pp. 563–571). Princeton, Pang, J. S. (2010). The achievement motive: A review
NJ: Van Nostrand. of theory and assessment of n Achievement, hope of
Moulton, R. W. (1965). Effects of success and failure on success, and fear of failure. In O. C. Schultheiss &
level of aspiration as related to achievement motives. J. C. Brunstein (Eds.), Implicit motives (pp. 30–70).
Journal of Personality and Social Psychology, 1, New York, NY: Oxford University Press.
399–406. Pang, J. S., & Schultheiss, O. C. (2005). Assessing
Mücher, H., & Heckhausen, H. (1962). Influence of implicit motives in U.S. College students: Effects
mental activity and achievement motivation on skel- of picture type and position, gender and ethnicity,
etal muscle tonus. Perceptual and Motor Skills, 14, and cross-cultural comparisons. Journal of Personal
217–218. Assessment, 85, 280–294.
Müller, E. F., & Beimann, M. (1969). Die Beziehung Pekrun, R., Elliot, A. J., & Maier, M. A. (2006).
der Harnsäure zu Testwerten der nach Heckhausen Achievement goals and discrete achievement emo-
gemessenen Leistungsmotivation. Zeitschrift für tions: A theoretical model and prospective test.
Experimentelle und Angewandte Psychologie, 16, Journal of Educational Psychology, 98, 583–597.
295–316. Pennebaker, J. W., & Francis, M. E. (1999). Linguistic
Murray, H. A. (1933). The effect of fear upon estimates inquiry and word count: LIWC. Mahwah, NJ:
of the maliciousness of other personalities. Journal of Erlbaum.
Social Psychology, 4, 310–329. Pennebaker, J. W., & King, L. A. (1999). Linguistic styles:
Murray, H. A. (1938). Explorations in personality. Language use as an individual difference. Journal of
New York, NY: Oxford University Press. Personality and Social Psychology, 77, 1296–1312.
Murray, H. A. (1943). Thematic apperceptive test manual. Peterson, B. E., & Stewart, A. J. (1993). Generativity and
Cambridge, MA: Harvard University Press. social motives in young adults. Journal of Personality
Murstein, B. I., & Pryer, R. S. (1959). The concept of and Social Psychology, 65, 186–198.
projection: A review. Psychological Bulletin, 56, Plutchic, R. (1980). Emotion: A psychoevolutionary syn-
353–374. thesis. New York, NY: Harper & Row.
Nicholls, J. G. (1984a). Achievement motivation: Rasch, G. (1960). Probabilistic models for some intelli-
Conceptions of ability, subjective experience, task gence and attainment tests. Kopenhagen, Denmark:
choice, and performance. Psychological Review, 91, Nielson & Lydicke.
328–346. Raynor, J. O. (1969). Future orientation and motivation
Nicholls, J. G. (1984b). Conceptions of ability and of immediate activity: An elaboration of the theory of
achievement motivation. In R. Ames & C. Ames achievement motivation. Psychological Review, 76,
(Eds.), Student motivation (pp. 39–73). Orlando, FL: 606–610.
Academic Press. Raynor, J. O. (1974). Future orientation in the study of
Nicholls, J. G. (1989). The competitive ethos and demo- achievement motivation. In J. W. Atkinson & J. O.
cratic education. Cambridge, MA: Harvard University Raynor (Eds.), Motivation and achievement (pp. 121–
Press. 154). Washington, DC: Winston.
Niitamo, P. (1999). “Surface” and “depth” in human Raynor, J. O., & Entin, E. E. (1982). Motivation, career
personality: Relations between explicit and implicit striving and aging. Washington, DC: Hemisphere.
motives. Helsinki, Finland: Finish Institute of Raynor, J. O., & Roeder, G. P. (1987). Motivation and
Occupational Health. future orientation: Task and time effects for achieve-
Nygard, R. (1975). A reconsideration of the achieve- ment motivation. In F. Halisch & J. Kuhl (Eds.),
ment motivation theory. European Journal of Social Motivation, intention, and volition (pp. 61–71). Berlin,
Psychology, 5, 61–92. Germany: Springer.
Nygard, R. (1977). Personality, situation, and persistence. Reitman, W. R. (1960). Motivational induction and the
Oslo, Norway: Universitetsforlaget. behavioral correlates of the achievement and affiliation
6 Achievement Motivation 301
motives. Journal of Abnormal and Social Psychology, Sawusch, J. R. (1974). Computer simulation of the
60, 8–13. influence of ability and motivation on test perfor-
Reuman, D. A. (1982). Ipsative behavioral variability and mance and cumulative achievement and the relation
the quality of thematic apperceptive measurement of between them. In J. W. Atkinson & J. O. Raynor
the achievement motive. Journal of Personality and (Eds.), Motivation and achievement (pp. 425–438).
Social Psychology, 43, 1098–1110. Washington, DC: Winston.
Revelle, W. (1986). Motivation and efficiency of cogni- Scheffer, D. (2003). Die Messung impliziter Motive.
tive performance. In D. R. Brown & J. Veroff (Eds.), Göttingen, Germany: Hogrefe.
Fronties of motivational psychology (pp. 107–127). Schiefele, U., & Rheinberg, F. (1997). Motivation and
Berlin, Germany: Springer. knowledge acquisition: Searching for mediating
Revelle, W., & Michaels, E. J. (1976). The theory of processes. In M. L. Maehr & P. R. Pintrich (Eds.),
achievement motivation revisited: The implications Advances in motivation and achievement (Vol. 10,
of inertial tendencies. Psychological Review, 83, pp. 251–301). Greenwich, CT: JAI Press.
394–404. Schiefele, U., & Urhahne, D. (2000). Motivationale
Rheinberg, F. (1980). Leistungsbewertung und und volitionale Bedingungen der Studienleistung.
Lernmotivation. Göttingen, Germany: Hogrefe. In U. Schiefele & K.-P. Wild (Eds.), Interesse und
Rheinberg, F. (2004). Motivational competence and flow- Lernmotivation: Untersuchungen zu Entwicklung,
experience. Paper presented at the 2nd European con- Förderung und Wirkung (pp. 183–205). Münster,
ference of positive psychology, Verbania, Italy. Germany: Waxmann.
Rheinberg, F., & Engeser, S. (2010). Motive training and Schmalt, H.-D. (1973). Die GITTER-Technik – ein objek-
motivational competence. In O. C. Schultheiss & J. C. tives Verfahren zur Messung des Leistungsmotivs bei
Brunstein (Eds.), Implicit motives (pp. 510–548). Kindern. Zeitschrift für Entwicklungspsychologie und
New York, NY: Oxford University Press. Pädagogische Psychologie, 5, 231–252.
Rheinberg, F., & Krug, S. (2004). Motivationsförderung Schmalt, H.-D. (1976a). Das
im Schulalltag: Psychologische Grundlagen und LM-GITTER. Handanweisung. Göttingen, Germany:
praktische Durchführung (3. Aufl. ed.). Göttingen, Hogrefe.
Germany: Hogrefe. Schmalt, H.-D. (1976b). Die Messung des Leistungsmotivs.
Rheinberg, F., Schmalt, H., & Wasser, I. (1978). Ein Göttingen, Germany: Hogrefe.
Lehrerunterschied, der etwas ausmacht. Zeitschrift Schmalt, H.-D. (1999). Assessing the achievement motive
für Entwicklungspsychologie und Pädagogische using the Grid technique. Journal of Research in
Psychologie, 10, 3–7. Personality, 33, 109–130.
Rheinberg, F., Duscha, R., & Michels, U. (1980). Schmalt, H.-D. (2003). Leistungsmotivation im
Zielsetzung und Kausalattribution in Abhängigkeit Unterricht: Über den Einsatz des LM-Gitters in der
vom Leistungsvergleich. Zeitschrift für Schule. In J. Stiensmeier-Pelster & F. Rheinberg
Entwicklungspsychologie und Pädagogische (Eds.), Diagnostik von Motivation und Selbstkonzept
Psychologie, 12, 177–189. (pp. 105–127). Göttingen, Germany: Hogrefe.
Rheinberg, F., Vollmeyer, R., & Burns, B. D. (2000). Schmalt, H.-D., & Sokolowski, K. (2000). Zum gegen-
Motivation and self-regulated learning. In wärtigen Stand der Motivdiagnostik. Diagnostica, 46,
J. Heckhausen (Ed.), Motivational psychology of 115–123.
human development: Developing motivation and moti- Schmalt, H.-D., Sokolowski, K., & Langens, T.
vating development (pp. 81–108). Amsterdam, The (2000). Das Multi-Motiv-Gitter (MMG). Lisse, The
Netherlands: Elsevier. Netherlands: Swets.
Riskind, J. H. (1984). They stoop to conquer: Guiding Schneider, K. (1971). Leistungs- und Risikoverhalten
and self-regulatory functions of physical posture after in Abhängigkeit von situativen und überdauernden
success and failure. Journal of Personality and Social Komponenten der Leistungsmotivation: Kritische
Psychology, 47, 479–493. Untersuchungen zu einem Verhaltensmodell.
Rogers, E. M., & Svenning, L. (1969). Modernization Unveröffentlichte Dissertation, RUB, Abt. für
among peasants: The impact of communication. Philosophie, Pädagogik, Psychologie, Bochum,
New York, NY: Holt, Rinehart & Winston. Germany.
Sader, M., & Keil, W. (1968). Faktorenanalytische Schneider, K. (1973). Motivation unter Erfolgsrisiko.
Untersuchungen zur Projektion der Leistungsmotivation. Göttingen, Germany: Hogrefe.
Archiv für die gesamte Psychologie, 120, 25–53. Schneider, K. (1974). Subjektive Unsicherheit und
Sader, M., & Specht, H. (1967). Leistung, Motivation Aufgabenwahl. Archiv für Psychologie, 126, 147–169.
und Leistungsmotivation: Korrelationsstatistische Schneider, K., & Heckhausen, H. (1981). Subjective
Untersuchungen zur Leistungsmotivmessung nach uncertainty and task preference. In H. I. Day (Ed.),
Heckhausen. Archiv für die gesamte Psychologie, 119, Advances in intrinsic motivation und aesthetics
90–130. (pp. 149–167). New York, NY: Plenum.
Sanford, R. N. (1937). The effects of abstinence from Schneider, K., & Kreuz, A. (1979). Die Effekte unter-
food upon imaginal processes: A preliminary experi- schiedlicher Anstrengung auf die Mengen- und
ment. Journal of Psychology, 3, 145–159. Güteleistung bei einer einfachen und schweren
302 J.C. Brunstein and H. Heckhausen
Zahlensymbolaufgabe. Psychologie und Praxis, 23, Scott, W. A. (1956). The avoidance of threatening
34–42. material in imaginative behavior. Journal of Abnormal
Schneider, K., Wegge, J., & Konradt, U. (1993). Motivation and Social Psychology, 52, 338–346.
und Leistung. In J. Beckmann, H. Strang, & E. Hahn Sedikides, C., & Strube, M. J. (1997). Self-evaluation: To
(Eds.), Aufmerksamkeit und Energetisierung: Facetten thine own self be good, to thine own self be sure, to
von Konzentration und Leistung (pp. 101–131). thine own self be true, and to thine own self be bet-
Göttingen, Germany: Hogrefe. ter. Advances in Experimental Social Psychology, 29,
Schroth, M. L. (1988). Relationships between 209–269.
achievement-related motives, extrinsic conditions, and Seidenstücker, G., & Seidenstücker, E. (1974).
task performance. Journal of Social Psychology, 127, Contribution to a computer evaluation of the the-
39–48. matic achievement motivation test by Heckhausen.
Schuler, H., & Prochaska, M. (2000). Entwicklung Psychologische Beiträge, 16, 68–92.
und Konstruktvalidierung eines berufsbezogenen Shantz, A., & Latham, G. P. (2009). An exploratory field
Leistungsmotivationstests. Diagnostica, 46, 61–72. experiment of the effect of subconscious and conscious
Schüler, J., Brandstätter, V., Wegner, M., & Baumann, goals on employee performance. Organizational
N. (2015). Testing the convergent and discriminant Behavior and Human Decision Processes, 109, 9–17.
validity of three implicit motive measures: PSE, OMT, Shoda, & Mischel, Y. (1995). A cognitive-affective system
MMG. Motivation and Emotion, 39, 839–857. theory of personality: Reconceptualizing situations,
Schultheiss, O. C. (2001b). Manual for the assessment dispositions, dynamics, and invariance in personality
of hope of success and fear of failure: English trans- structure. Psychological Review, 102, 246–268.
lation of Heckhausens need achievement measure. Short, J.-A. C., & Sorrentino, R. M. (1986). Achievement,
Unpublished manuscript, University of Michigan at affiliation, and group incentives: A test of the overmotiva-
Ann Arbor. tion hypothesis. Motivation and Emotion, 10, 115–131.
Schultheiss, O. C. (2001a). An information processing Sijtsma, K. (2009). On the use, the misuse, and the
account of implicit motive arousal. In M. L. Maehr very limited usefulness of Cronbach’s alpha.
& P. Pintrich (Eds.), Advances in motivation and Psychometrika, 74, 107–120.
achievement, Vol. 12: New directions in measures and Singh, S. (1979). Relationships among projective and
methods (pp. 1–41). Greenwich, CT: JAI Press. direct verbal measures of achievement motivation.
Schultheiss, O. C. (2013). Are implicit motives revealed Journal of Personality Assessment, 43, 45–49.
in mere words? Testing the marker-word hypoth- Slabbinck, H., De Houwer, J., & Van Kenhove, P. (2012).
esis with computer-based text analysis. Frontiers in The pictorial attitude implicit association test for need
Psychology, 4, 748. for affiliation. Personality and Individual Differences,
Schultheiss, O. C., & Brunstein, J. C. (2001). Assessing 53, 838–842.
implicit motives with a research version of the tat: Slabbinck, H., De Houwer, J., & Van Kenhove, P. (2013).
Picture profiles, gender differences, and relations to Convergent, discriminant, and incremental validity
other personality measures. Journal of Personality of the pictorial attitude Implicit Association Test and
Assessment, 77, 71–86. the Picture Story Exercise as measures of the implicit
Schultheiss, O. C., & Brunstein, J. C. (2005). An implicit power motive. European Journal of Personality, 27,
motive perspective on competence motivation. In A. J. 30–38.
Elliot & C. S. Dweck (Eds.), Handbook of competence Slavin, R. E. (1995). Cooperative learning (2. Aufl. ed.).
and motivation (pp. 31–51). New York, NY: Guilford Boston, MA: Allyn & Bacon.
Press. Smith, C. P. (Ed.). (1992). Motivation and personality:
Schultheiss, O. C., Liening, S., & Schad, D. (2008). Handbook of thematic content analysis. New York,
The reliability of a Picture Story Exercise measure NY: Cambridge University Press.
of implicit motives: Estimates of internal consis- Smith, C. P., & Feld, S. C. (1958). How to learn the
tency, retest stability, and ipsative stability. Journal of method of content analysis for n achievement, n affili-
Research in Personality, 42, 1560–1571. ation, and n power. In J. W. Atkinson (Ed.), Motives in
Schultheiss, O. C., Wiemers, U. S., & Wolf, O. T. (2014). fantasy, action, und society (pp. 685–818). Princeton,
Implicit need for achievement predicts attenuated cor- NJ: Van Nostrand.
tisol responses to difficult tasks. Journal of Research Sokolowski, K., Schmalt, H.-D., Langens, T., & Puca,
in Personality, 48, 84–92. R. M. (2000). Assessing achievement, affiliation,
Schwarz, N. (1990). Feeling as information: Informational and power motives all at once: The Multi-Motive-
and motivational functions of affective states. In E. T. Grid (MMG). Journal of Personality Assessment, 74,
Higgins & R. M. Sorrentino (Eds.), Handbook of moti- 126–145.
vation and cognition: Foundations of social behavior Sorrentino, R. M., & Hewitt, E. C. (1984). The uncertainty-
(Vol. 2, pp. 527–561). New York, NY: Guilford. reducing properties of achievement tasks revisited.
Schwinger, M., Wirthwein, L., Lemmer, G., & Steinmayr, Journal of Personality and Social Psychology, 47,
R. (2014). Academic self-handicapping and achieve- 884–899.
ment: A meta-analysis. Journal of Educational Sorrentino, R. M., & Short, J. (1977). The case of the
Psychology, 106, 744–761. mysterious moderates: Why motives sometimes fail
6 Achievement Motivation 303
to predict behavior. Journal of Personality and Social Trope, Y. (1986b). Self-enhancement and self-assessment
Psychology, 35, 478–484. in achievement behavior. In R. M. Sorrentino & E. T.
Sorrentino, R. M., Roney, C. J. E., & Hewitt, E. C. Higgins (Eds.), Handbook of motivation and cogni-
(1988). Information value versus affective value and tion: Foundations of social behavior (pp. 350–378).
achievement behavior. In F. Halisch & J. H. L. van den New York, NY: Guilford.
Bercken (Eds.), International perspectives on achieve- Trope, Y., & Brickman, P. (1975). Difficulty and diag-
ment and task motivation. Lisse, The Netherlands: nosticity as determinants of choice among tasks.
Swets & Zeitlinger. Journal of Personality and Social Psychology, 31,
Sorrentino, R. M., Hanna, S. E., & Roney, C. J. R. 918–926.
(1992). Uncertainty orientation. In C. P. Smith (Ed.), Trudewind, C., & Husarek, B. (1979). Mutter-Kind-
Motivation and personality: Handbook of thematic Interaktion bei der Hausaufgabenbetreuung und die
content analysis (pp. 428–439). New York, NY: Leistungsmotiventwicklung im Grundschulalter:
Cambridge University Press. Analyse einer ökologischen Schlüsselsituation.
Sorrentino, R. M., Smithson, M., Hodson, G., Roney, C. J. In H. Walter & R. Oerter (Eds.), Ökologie und
R., & Walker, A. M. (2003). The theory of uncertainty Entwicklung (pp. 229–246). Stuttgart, Germany: Klett.
orientation: A mathematical reformulation. Journal of Tuerlinckx, F., De Boeck, P., & Lens, W. (2002).
Mathematical Psychology, 47, 132–149. Measuring needs with the thematic apperception test:
Spangler, W. D. (1992). Validity of questionnaire and A psychometric study. Journal of Personality and
TAT measures of need for achievement: Two meta- Social Psychology, 82, 448–461.
analyses. Psychological Bulletin, 112, 140–154. Veroff, J. (1969). Social comparison and the develop-
Spence, J. T., & Helmreich, R. L. (1978). Masculinity and ment of achievement motivation. In C. P. Smith (Ed.),
femininity. Austin, TX: University of Texas Press. Achievement-related motives in children (pp. 46–101).
Stewart, A. J., & Chester, N. L. (1982). Sex differences New York, NY: Sage.
in human social motives: Achievement, affiliation, and Weber, M. (1904). Die protestantische Ethik und der Geist
power. In A. J. Stewart (Ed.), Motivation and society des Kapitalismus. Archiv für Sozialwissenschaft und
(pp. 172–218). San Francisco, CA: Jossey-Bass. Sozialpolitik, 20, 1–54.
Stiensmeier-Pelster, J., & Rheinberg, F. (Eds.). (2003). Weinberger, J., & McClelland, D. C. (1990). Cognitive
Diagnostik von Motivation und Selbstkonzept. versus traditional motivational models: Irreconcilable
Göttingen, Germany: Hogrefe. or complementary? In E. T. Higgins & R. M.
Stone, P. J., Dumphy, D. C., Smith, M. S., & Ogilvie, Sorrentino (Eds.), Handbook of motivation and cogni-
D. M. (1966). The general inquirer. Cambridge, MA: tion: Foundations of social behavior (Vol. 2, pp. 562–
MIT Press. 597). New York, NY: Guilford.
Szeto, A. C. H., Sorrentino, R. M., Yasunaga, S., Weiner, B. (1965a). Need achievement and the resump-
Kouhara, S., & Lin, L. (2011). Motivation and perfor- tion of incompleted tasks. Journal of Personality and
mance: Uncertainty regulation in Canada and Japan. Social Psychology, 1, 165–168.
Motivation and Emotion, 35, 338–350. Weiner, B. (1965b). The effects of unsatisfied achieve-
Taylor, S. E., Neter, E., & Wayment, H. A. (1995). ment motivation on persistence and subsequent per-
Self-evaluation processes. Personality and Social formance. Journal of Personality, 33, 428–442.
Psychology Bulletin, 21, 1278–1287. Weiner, B. (1967). Implications of the current theory
Thomas, E. A. C. (1983). Notes on effort and achievement of achievement motivation for research and perfor-
oriented behavior. Psychological Review, 90, 1–20. mance in the classroom. Psychology in the School, 4,
Thrash, T. M., & Elliot, A. J. (2002). Implicit and self- 164–171.
attributed achievement motives: Concordance and pre- Weiner, B. (1970). New conceptions in the study of
dictive validity. Journal of Personality, 70, 729–755. achievement motivation. In B. Maher (Ed.), Progress
Thurstone, L. L. (1937). Ability, motivation, and speed. in experimental personality research (Vol. 5,
Psychometrika, 2, 249–254. pp. 67–109). New York, NY: Academic Press.
Trope, Y. (1975). Seeking information about one’s own Weiner, B. (1974). Achievement motivation and attribu-
ability as a determinant of choice among tasks. Journal tion theory. Morristown, NJ: General Learning.
of Personality and Social Psychology, 32, 1004–1013. Weiner, B. (1979). A theory of motivation for some class-
Trope, Y. (1980). Self-assessment, self-enhancement, room experiences. Journal of Educational Psychology,
and task preference. Journal of Experimental Social 71, 3–25.
Psychology, 16, 116–129. Weiner, B. (1985). An attributional theory of achievement
Trope, Y. (1983). Self-assessment in achievement behavior. motivation and emotion. Psychological Review, 92,
In J. M. Suls & A. G. Greenwald (Eds.), Psychological 548–573.
perspectives on the self (Vol. 2, pp. 93–121). Hillsdale, Weiner, B., Frieze, I. H., Kukla, A., Reed, L., Rest, S., &
NJ: Erlbaum. Rosenbaum, R. M. (1971). Perceiving the causes of
Trope, Y. (1986a). Testing self-enhancement and self- success and failure. New York, NY: General Learning.
assessment theories of achievement motivation: A Weisfeld, G. E., & Beresford, J. M. (1982). Erectness of
reply to Sohn’s critique. Motivation and Emotion, 10, posture as an indicator of dominance or success in
247–261. humans. Motivation and Emotion, 6, 113–131.
304 J.C. Brunstein and H. Heckhausen
Wendt, H. W. (1955). Motivation, effort, and per- Winter, D. G., & Stewart, A. J. (1977). Power motive
formance. In D. C. McClelland (Ed.), Studies reliability as a function of retest instructions. Journal
in motivation (pp. 448–459). New York, NY: of Consulting and Clinical Psychology, 45, 436–440.
Appleton-Century-Crofts. Wolters, C. A. (2003). Regulation of motivation:
Wendt, H. W. (1967). Verhaltensmodelle des Evaluating an underemphasized aspect of self-regu-
Nichtwissenschaftlers: Einige biographische und lated learning. Educational Psychologist, 38, 189–205.
Antriebskorrelate der wahrgenommenen Beziehung Wood, R. E. (1986). Task complexity: Definition of the
zwischen Erfolgswahrscheinlichkeit und Zielanreiz. construct. Organizational Behavior and Decision
Psychologische Forschung, 30, 226–249. Processes, 37, 60–82.
Wicklund, R. A., & Gollwitzer, P. M. (1982). Symbolic Wood, R. E., Mento, A. J., & Locke, E. A. (1987). Task
self-completion. Hillsdale, NJ: Erlbaum. complexity as a moderator of goal effects: A meta-
Wigfield, A., & Eccles, J. S. (2000). Expectancy-value analysis. Journal of Applied Psychology, 72, 416–425.
theory of achievement motivation. Contemporary Wright, R. A. (1996). Brehm’s theory of motivation as a
Educational Psychology, 25, 68–81. model of effort and cardiovascular response. In P. M.
Wilson, T. (2002). Strangers to ourselves: Discovering Gollwitzer & J. A. Bargh (Eds.), The psychology of
the adaptive unconscious. Cambridge, MA: Harvard action: Linking cognition and motivation to behavior
University Press. (pp. 424–453). New York, NY: Guilford.
Wilson, T., Lindsey, S., & Schooler, T. Y. (2000). A Yang, F., Ramsay, R. E., Schultheiss, O. C., & Pang, J. S.
model of dual attitudes. Psychological Review, 107, (2015). Need for achievement moderates the effect of
101–126. motive-relevant challenge on salivary cortisol changes.
Wine, J. (1971). Test anxiety and direction of attention. Motivation and Emotion, 39, 321–334.
Psychological Bulletin, 76, 92–104. Yerkes, R. M., & Dodson, J. D. (1908). The relation of
Winter, D. G. (1991a). Manual for scoring motive imag- strength of stimulus to rapidity of habit-formation.
ery in running text (3. Aufl.). Unpublished scoring Journal of Comparative and Neurological Psychology,
manual, University of Michigan at Ann Arbor. 18, 459–482.
Winter, D. G. (1991b). Measuring personality at a Zeigarnik, B. (1927). Über das Behalten von erledig-
distance: Development of an integrated system ten und unerledigten Handlungen. Psychologische
for scoring motives in running text. In R. Hogan, Forschung, 9, 1–85.
D. Ozer, J. M. Healy, & A. J. Stewart (Eds.), Zimmerman, B. J., & Kitsantas, A. (1997). Developmental
Perspectives in personality. Vol. 3B: Approaches phases in self-regulation: Shifting from process goals
to understanding lives (pp. 59–89). London, UK: to outcome goals. Journal of Educational Psychology,
Jessica Kingsley. 89, 29–36.