Learning and Individual Differences 53 (2017) 122–132

Contents lists available at ScienceDirect

Learning and Individual Differences

journal homepage: www.elsevier.com/locate/lindif

Which one works best? Considering the relative importance of

motivational regulation strategies
Malte Schwinger a,⁎, Nantje Otterpohl b
a
Department of Psychology, Philipps-University Marburg, Gutenbergstr. 18, D-35032 Marburg, Germany
b
Department of Psychology, University of Giessen, Otto-Behaghel-Str. 10F, 35394 Giessen, Germany

a r t i c l e i n f o a b s t r a c t

Article history: Although previous studies have shown that motivational regulation strategies are generally effective in
Received 14 June 2016 sustaining students' effort and persistence, little is known about which kind of motivational strategy
Received in revised form 4 November 2016 works best in general, and for certain students in particular. In this article, we investigated the relative im-
Accepted 15 December 2016
portance of eight different motivational regulation strategies across three samples (N = 531, N = 613, and
Available online xxxx
N = 301, respectively) of German high school and college students under varying conditions. Relative
Keywords:
weights analyses enabled us to control for multicollinearity and to disentangle the unique proportion of
Motivational regulation variance each motivational strategy explained in students' self-reported effort. Moreover, we examined
Regulation of motivation potential moderating effects of gender, conscientiousness, dispositional interest, and achievement goal
Relative weights analysis orientations. Consistently across the three samples, results revealed mastery self-talk as the most effective
Self-regulated learning strategy, followed by proximal goal setting and performance-approach self-talk. Interest enhancement
strategies and performance-avoidance self-talk did not explain a significant amount of variance in self-
reported effort. There were no substantial moderation effects leading to the conclusion that the established
rank order of motivational regulation strategies might be generally applicable to the majority of students.
We discuss theoretical implications of our findings for future studies in motivation regulation research as
well as practical implications for educational practitioners.
© 2016 Published by Elsevier Inc.

1. Introduction
The ability to enhance and uphold one's own motivation represents
an essential feature of self-regulated learning (e.g., Sansone & Thoman,
2005; Schwinger & Stiensmeier-Pelster, 2012). Researchers have there-
fore focused on the concept of motivational self-regulation which can be
defined as the more or less conscious control over one's own motivation
which mostly serves to increase effort and persistence (Wolters, 2003).
A number of studies have shown that students who use certain strate-
gies to regulate their motivation put more effort into learning tasks
when faced with obstacles or difficulties (Leutner, Barthel, &
Schreiber, 2001; Schwinger, Steinmayr, & Spinath, 2009; Wolters &
Benzon, 2013). In contrast to the relatively well-established finding
that frequently using any kind of motivational regulation strategy
often results in higher effort and persistence, we do not know much
about their relative importance, that is, which kind of motivational
⁎ Corresponding author.
E-mail addresses: malte.schwinger@uni-marburg.de (M. Schwinger),
nantje.otterpohl@psychol.uni-giessen.de (N. Otterpohl).
strategy might be the best to choose in general, and/or for certain stu-
dents in particular. Although some authors have focused on this issue
(e.g., Engelschalk, Steuer, & Dresel, 2015; Wolters, 1998), the available
empirical evidence is limited in several ways. First, a considerable num-
ber of studies have not examined a broad set of motivational regulation
strategies simultaneously, which limits the generalizability of their find-
ings. Second, the statistical procedures used were inadequate for con-
trolling the multicollinearity among motivational strategies. Third,
only a few studies have considered personality differences as possible
moderating factors (Sansone, Wiebe, & Morgan, 1999; Schwinger et
al., 2009).
In the present paper, we strive to address these shortcomings
and to further our understanding of the relative importance of mo-
tivational regulation strategies. We report the findings from three
empirical studies in which students' learning effort has been pre-
dicted by eight different motivational regulation strategies. Using
Relative Weights Analysis (Tonidandel & LeBreton, 2011), we exam-
ine the relative importance of each motivational strategy across
the three studies. Moreover, we investigate whether relative strate-
gy importance differs depending on students' gender, conscien-
tiousness, dispositional interest, and achievement goal orientations,

http://dx.doi.org/10.1016/j.lindif.2016.12.003
1041-6080/© 2016 Published by Elsevier Inc.
 M. Schwinger, N. Otterpohl / Learning and Individual Differences 53 (2017) 122–132
which have all been described as important covariates in motivation
regulation contexts.
1.1. Strategies for motivational regulation
Students can use several strategies in order to successfully regulate
their motivation. Wolters (1998, 1999) originally proposed five differ-
ent motivational strategies. More recently, however, Wolters and
Benzon (2013) reported on the reliability and validity of an advanced
version of this instrument assessing six strategies. Schwinger, von
der Laden, and Spinath (2007; see also Schwinger et al., 2009) intro-
duced an adapted version of Wolters' original questionnaire in
which they distinguished eight different strategies for motivational
regulation. This taxonomy of strategies forms the basis for the
present article.
Most of the eight motivational strategies can be divided into one of
two categories that each describe the respective psychological mecha-
nism through which the strategies are supposed to influence students'
task-related motivation. The first group of strategies may be labelled in-
terest-enhancement strategies, whereas the second group consists of
rather goal-based strategies. Although there are certainly different
ways of sorting the various motivational strategies, we think this cate-
gorization makes it easier to grasp the essentials of each strategy as
well as to see similarities and differences between them. With respect
to the group of interest-enhancement strategies, the first motivational
strategy is called enhancement of situational interest and it refers to situ-
ations when students try to modify the task in a way that makes it more
exciting and interesting for them. For example, younger students who
have to copy long text passages at school might use different colors
in order to stay motivated for this rather boring task (Sansone, Weir,
Harpster, & Morgan, 1992; Sansone et al., 1999). While this strategy
focuses on a short-term increase of enjoyment and persistence, the
strategy enhancement of personal significance describes students'
efforts to identify additional reasons for why engaging in certain
tasks could be meaningful for them. That is, they try to find relations
between the tasks at hand and their own individual interests,
preferences, and goals in life (Leutner et al., 2001). The third strate-
gy possibly labelled as an option for interest-enhancement is called
self-consequating. This strategy is among the most frequently applied
strategies for motivation regulation. Herein, students promise
themselves a reward (e.g., socializing, watching a movie) for
successfully accomplishing a certain task.
Regarding the group of goal-based strategies, proximal goal setting
represents a strategy where students split a lengthy task into small
pieces in order to feel more self-efficacious about the task (Bandura &
Schunk, 1981). A great number of students use proximal goal setting
and self-consequating in combination (cf. Wolters, 2003). Overall
three goal-based strategies comprise self-talk related to long-term
goals. Based on the trichotomous model of achievement goals (Elliot,
1999), mastery self-talk refers to remembering and thinking about the
goal to improve one's competencies and to learn as much as possible.
Performance-approach self-talk means that students recall their goal of
being better than their classmates whereas performance-avoidance
self-talk describes self-instructing by thinking about not to be worse
than others. While seven of eight motivational strategies can be easily
categorized as rather interest-enhancement vs. goal-based strategy,
this is not that easy for the last strategy which is named environmental
control. It refers to any kinds of rearranging the learning environment
in a way that it helps to sustain one's motivation and persistence. For
example, many students state to prefer quiet places for learning
(Zimmerman & Martinez-Pons, 1986). Whether using environmental
control for motivation regulation enhances one's effort through increas-
ing interest and task-related value or through boosting self-efficacy to
achieve one's goals may depend on both personal and contextual
factors.
123
1.2. Effectiveness of motivational regulation strategies
Studies testing the effectiveness1 of motivational regulation strate-
gies have mostly focused on two outcome variables, namely academic
effort and achievement. Findings from these studies have indicated
that motivational regulation strategies have only weak, if any, direct ef-
fects on achievement (Schwinger et al., 2009; Schwinger, Steinmayr, &
Spinath, 2012; Wolters, 1998, 1999). However, focusing mainly on di-
rect effects of motivational strategies may underestimate the impor-
tance of motivational regulation for students' achievement. In a later
review of the motivational regulation literature, Wolters clearly stated
that “… one immediate goal of students' regulation of motivation is to in-
crease their effort, persistence, or choice of activities. This adaptive change
in motivation ultimately should positively affect outcomes that more direct-
ly reflect students' learning and achievement” (Wolters, 2003, p. 201).
Thus, strategies for motivational regulation first and foremost aim to
optimize students' learning efforts. Substantive direct effects of motiva-
tional strategies on achievement are not to be expected. This assump-
tion parallels the discussion on the effectiveness of cognitive learning
strategies, which also varies depending on the respective outcome
(e.g., Credè & Kuncel, 2008). In conclusion, supposed positive effects of
motivational regulation strategies on achievement would be mediated
by an increase in task-related effort and persistence. Schwinger et al.
(2009) yielded empirical support for the mediation effect hypothesis.
In a sample of eleventh and twelfth grade German high school students,
they found six of the eight motivational regulation strategies discussed
above to be significantly related to students' effort management, but not
to their school grades. Effort management, in turn, was demonstrated to
be a significant predictor of students' GPA.
With respect to the question which motivational outcome is most
important to consider, we agree with Wolters (2003) and Pintrich
(2004), who have suggested the broad concept of effort to be the rele-
vant outcome criterion in motivational regulation. Other researchers
have argued that the motivational outcome should not be restricted to
effort. Sansone and Thoman (2005), for example, stressed that having
extrinsic reasons to perform a task (e.g., expecting rewards) is not
enough to maintain motivation when a person's interest in the task
has not been regulated as well. If the activity is continued only due to su-
perordinate reasons, the person is presumed to feel stressed and hassled
because there is no positive phenomenological experience while com-
pleting the task. As a consequence, the person will probably quit the
task earlier and/or will perform worse compared to a person who has
experienced a successful regulation of task-related interest. Following
Sansone and Thoman (2005), we would need to examine phenomeno-
logical experiences such as task-related interest as well in our studies.
However, in the three studies reported here, we focused on effort as
the only outcome variable since we believe that investing a suitable
amount of time and energy, which describes the core of the effort con-
struct, represents the most fundamental form of motivation needed to
successfully complete a task. While we agree with Sansone and
Thoman (2005) that experiencing some kind of intrinsic motivation
would be both beneficial and satisfying for the learner, we see it as
just one of many precursors of the finally resulting effort.
Regarding the impact of motivational regulation on students' subjec-
tive effort, several studies have reported positive effects. In a study by
Wolters (1999), high school students' self-reported effort was linked
1
In this article, we use the term “effectiveness” twofold. First, we are interested wheth-
er certain strategies for motivational regulation are generally effective in enhancing stu-
dents' effort. In this regard, a non-significant correlation between a motivational
strategy and effort would indicate an ineffective strategy. The second use of the term effec-
tiveness refers to the question of relative importance, that is, whether particular motiva-
tional regulation strategies appear to be more effective than others. We would like to
point out that we are not able to examine the “efficiency” of motivational strategies in
the studies presented here. This would relate to the question whether using a certain strat-
egy in qualitatively different ways leads to qualitatively different effects. Albeit interesting,
issues of qualitative strategy efficiency are not part of this article.
124 M. Schwinger, N. Otterpohl / Learning and Individual Differences 53 (2017) 122–132
to their use of interest enhancement, mastery and performance self-
talk, self-consequating, and environmental control. Schwinger et al.
(2009, 2012) found the eight motivational regulation strategies de-
scribed above to be associated with an increase in effort expenditure
and current learning motivation, which was, in turn, positively associat-
ed with students' performance. In addition to these findings from field
studies, experimental evidence has been reported for several particular
motivational strategies. Hulleman and Harackiewicz (2009) asked stu-
dents to find connections between their own individual interests and
the contents taught in science class (i.e., they encouraged them to use
the strategy enhancement of personal significance). Compared to an un-
treated control group, these students substantially increased their inter-
est in science over time. Jackson and Molloy (1985) provided evidence
that students could solve more math problems when they promised
themselves rewards (i.e., when they used self-consequating) for being
successful in those tasks. Sansone et al. (1992) instructed students
who had to copy letters by hand to ornament the characters. This proce-
dure, which resembles the strategy enhancement of situational interest,
led to increased persistence compared to not instructed students. Taken
together, there is substantial evidence that motivational strategies are
positively linked to indicators of students' learning motivation, such as
effort and persistence.
1.3. Relative importance of motivational regulation strategies
A number of college and high school students have reported having
problems in successfully regulating their own learning motivation (e.g.,
Pychyl, Lee, Thibodeau, & Blunt, 2000). These problems, in turn, might
increase the risk for maladaptive reactions, such as procrastination or
school dropout (Grunschel, Patrzek, & Fries, 2013; Wolters, 2003). In
order to develop efficient intervention programs that assist those stu-
dents in building up an adaptive motivation regulation profile, it
would be helpful to know which one of the motivational regulation
strategies works best for which kinds of students. However, the avail-
able empirical evidence on the effectiveness of motivational regulation
strategies is limited in several ways. For example, there is a lack of stud-
ies assessing motivation regulation on a behavioral level and/or consid-
ering the quality of strategy use. Altogether, it is difficult to draw valid
conclusions on the relative importance of a particular motivational
strategy. Possible reasons for this specific shortcoming include (a) the
small number of studies examining a broad set of motivational strate-
gies simultaneously, (b) the use of inadequate statistical procedures,
and (c) the non-consideration of personality characteristics as moderat-
ing factors.
1.3.1. Examining a broad set of motivational strategies
Many prior studies examining motivational regulation focused on a
single strategy, or only a small number of strategies. As described
above, Sansone and colleagues (Sansone & Thoman, 2005; Sansone et
al., 1992) have centered exclusively on students' tendency to increase
their situational interest. Similarly, the line of research pioneered by
Zimmerman and Martinez-Pons (1986) has tapped into a number of
different regulatory strategies, but only into one or two (e.g., self-
consequating, environmental structuring) that might arguably be char-
acterized as forms of motivational regulation. Other researchers tended
to rely on higher-order indicators of motivational regulation (e.g.,
McCann & Garcia, 1999; Wolters, 1998). Only a few studies have exam-
ined the respective effects of a larger number of motivational regulation
strategies. Wolters (1999) reported the highest correlation with effort
for mastery self-talk (r = 0.43) and the lowest for self-consequating
(r = 0.25). However, when all strategies were entered simultaneously
as predictors into a multiple regression model, only mastery self-talk
remained a significant predictor of students' effort. Schwinger and
colleagues (Schwinger et al., 2007, 2009, 2012; Schwinger &
Stiensmeier-Pelster, 2012) reported a quite similar correlation pattern
across several studies. Mastery self-talk mostly showed the highest
association to effort, with rs ranging between 0.25 and 0.48. Correla-
tions between performance-approach self-talk and effort were similarly
high, whereas the two interest enhancement strategies and perfor-
mance-avoidance self-talk were rather uncorrelated with effort. Unlike
the study by Wolters (1999), however, Schwinger and colleagues did
not examine the relative effectiveness of all strategies in a multiple
regression. They argued against such analyses due to the high
multicollinearity among the motivational strategies.
1.3.2. Using adequate analysis techniques
As stated above, the two common statistical procedures to evaluate
the importance of a set of motivational regulation strategies are zero-
order correlations and multiple regression estimates. Correlations rep-
resent the unique contribution of each predictor by itself, whereas re-
gression coefficients represent the incremental contribution of each
predictor when combined with all remaining predictors. Unfortunately,
both procedures are not suitable for determining the relative impor-
tance of a particular predictor examined together with a range of
other predictors (Johnson & LeBreton, 2004). Zero-order correlations
are inadequate because a specific motivational strategy may not func-
tion independently from other ones in the real-world setting. Tradition-
al multiple regression estimates are also inadequate because when
estimating the effect of a particular predictor, the effects of all other pre-
dictors need to be held constant. In most cases, however, there is high
multicollinearity among predictors, which leads to biased regression es-
timates. In Wolters' (1999) study, for instance, the strategy of interest
enhancement was positively correlated with effort (r = 0.30), but failed
to predict effort significantly in multiple regression.
Altogether, both kinds of information (zero-order correlations and
regression estimates) are necessary to determine the relative impor-
tance of each predictor in a set of predictor variables. A statistical proce-
dure that integrates this information and that also (mostly) overcomes
the multicollinearity problem is Relative Weights Analysis (Johnson &
LeBreton, 2004; Tonidandel & LeBreton, 2011). When independent pre-
dictor variables in multiple regression are correlated, relative weights
address this problem by using principal components analysis to trans-
form the original independent variables into a set of uncorrelated prin-
cipal components that are most highly correlated with the original
independent variables (Tonidandel & LeBreton, 2011). These compo-
nents are then submitted to two regression analyses. The first analysis
is a regression that predicts the dependent variable from these uncorre-
lated principal components. Next, the original independent variables
are regressed onto the uncorrelated principal components. Finally, rela-
tive weights are computed by multiplying squared regression weights
from the first analysis (regression of dependent variables on compo-
nents) with squared regression weights from the second analysis (re-
gression of independent variables on components). Each weight can
be divided by R2 and multiplied by 100 so that the new weights add
to 100%, with each weight reflecting the percentage of predictable var-
iance. Relative weights are unique as a measure of total effect in that
they provide rank orderings of individual independent variables' contri-
butions to a multiple regression effect in the presence of all other pre-
dictors based on a computational method that addresses associations
between independent variables by creating their uncorrelated “coun-
terparts” (Johnson & LeBreton, 2004). With respect to motivational reg-
ulation, we believe that relative weights analysis represents a suitable
procedure to disentangle the relative importance of each motivational
strategy.
1.3.3. Individual characteristics as moderating factors
The process of motivational regulation seems to be affected by con-
textual and individual factors that function as moderator variables by
influencing how often and how effectively particular motivational strat-
egies are used (Pintrich, 2004; Schwinger & Stiensmeier-Pelster, 2012;
Wolters, 2003). Some studies have shown that personality characteris-
tics can moderate the effectiveness of motivational regulation
 M. Schwinger, N. Otterpohl / Learning and Individual Differences 53 (2017) 122–132
strategies. For instance, Sansone et al. (1999) reported that
hardiness—defined as the tendency to perceive stressful life events as
controllable and as challenging opportunities, rather than as threats
(Kobasa, 1979)—moderated the effectiveness of interest-enhancing
strategies. Moreover, Schwinger et al. (2009) found that students who
have higher cognitive abilities use some, but not all, motivational regu-
lation strategies more effectively compared to students with less cogni-
tive abilities. Despite these and other studies, the overall empirical
evidence on the moderating role of individual characteristics in motiva-
tional regulation is still scarce. However, we think that the question of
relative strategy importance cannot be examined without considering
important individual features. In the three studies presented here, we
therefore assessed different aspects of the individual student.
Our search for potential moderating features was guided by the as-
sumption that motivational regulation is most effective when the strat-
egy chosen matches a learner's basic motivational beliefs and attitudes
(Wolters & Benzon, 2013). Moreover, we sought to assess potential
moderators for each group of motivational strategies, i.e. for interest-
enhancement and for goal-based strategies. Based on the literature re-
view and theoretical considerations, dispositional interest appeared to
be the most suitable moderator for the effectiveness of interest-en-
hancement strategies while conscientiousness and achievement goal
orientations were supposed to moderate the effectiveness of goal-
based strategies. Due to the suggested matching principle, it is plausible
to assume that highly interested individuals profit more from using in-
terest-enhancing strategies since they may find it easier to think about
and/or rearrange certain components of task value which they had
probably been aware of yet. Likewise, highly conscientious individuals
may use motivational strategies that refer to goals more effectively be-
cause these individuals set themselves ambitious goals, stay committed
to these goals during task completion, and work hard and persistently to
achieve their goals (Klein & Lee, 2006). In a similar and even more spe-
cific way, it would be reasonable to suggest that mastery-oriented stu-
dents would mostly benefit from using mastery self-talk, whereas
their classmates pursuing performance-approach goals may rely on per-
formance-approach self-talk as their favorite motivational strategy.
Given that prior research has shown that male and female students
often differ in their motivation beliefs (e.g., Roeser, Midgley, & Urdan,
1996) and their use of self-regulatory strategies (e.g., Cleary & Chen,
2009), we decided to include gender as a further potential moderator
in our analyses. To our knowledge, however, none of these studies in-
vestigated in how far gender moderated the effectiveness of self-regula-
tory strategy use. Thus, we had no specific expectations regarding
moderation effects of gender in our studies.
1.4. The present research
The few available studies that have examined the relative impor-
tance of motivational regulation strategies had several deficits, such as
focusing on a small number of strategies and using inadequate statistical
techniques. Altogether, we think that our knowledge of the relative im-
portance of motivational regulation strategies is limited and needs to be
furthered by more sophisticated methodological designs. In the present
paper, we therefore seek to examine the relative importance of motiva-
tional strategies in three different samples and under varying
conditions. In Study 1, we use a cross-sectional design where the moti-
vational strategies as independent variables and effort expenditure as a
dependent variable are assessed concurrently. In Study 2, we use a lon-
gitudinal design which enables us to control for students' initial motiva-
tion. In Study 3, we also use a longitudinal design, but additionally chose
a more specific level of analysis by examining motivational regulation
shortly before an important exam. Moreover, while we examine univer-
sity students in the first two studies, we choose 12th grade high-school
students as participants in Study 3. We expect that results will be similar
for rather global (Studies 1 and 2) vs. specific (Study 3) research designs
and also for university (Studies 1 and 2) vs. high-school (Study 3)
125
students. If this were the case, it would strengthen the generalizability
of our findings and interpretations on the relative importance of moti-
vational regulation strategies. In addition to the cumulative study de-
signs, there is also a stepwise increase in the number of personality
moderators assessed. Although gender is the only moderator in Study
1, conscientiousness is added in Study 2, and dispositional interest and
achievement goal orientations are further added in Study 3. This step-
wise increase of moderator variables allows us to both replicate and ex-
tend our findings from study to study.
2. Study 1
In Study 1, we want to take a first look at the relative importance of
the eight motivational regulation strategies described in the introduc-
tion. Based on the literature review, we specify the following hypotheses.
Hypothesis 1. All motivational regulation strategies contribute unique-
ly to the explanation of variance in students' self-reported effort.
Hypothesis 2. Goal-based motivational strategies predict a significant-
ly higher amount of variance in students' self-reported effort than inter-
est-enhancement strategies.
Hypothesis 3. Relative importance rank order of motivational strate-
gies will be the same for male and female students.
2.1. Method
2.1.1. Sample and procedure
In May 2010, a total of 531 students (362 female; age: M = 24.19,
SD = 3.87) from several German universities participated in an online
survey. At the beginning of the survey, students were informed that par-
ticipation in this study was voluntary and that they could decide to
withdraw from the study at any time without penalty. They were fur-
ther informed that the questions were anonymous, that their data will
be treated confidentially, that personal allocation from these data to
specific persons were not possible, and that return of the questionnaire
will be considered their consent to participate. About 35% of participants
were enrolled in science subjects, 50% in humanities, and the remaining
students were dispersed across various other subjects. In all, 55% were
in the first five semesters of their studies, whereas the remaining were
in higher semesters.
2.1.2. Measures
The eight motivational regulation strategies were assessed with the
questionnaire by Schwinger et al. (2007, 2009). Each scale is represent-
ed by three to five items which have all to be rated on a scale ranging
from 1 (strongly disagree) to 5 (strongly agree). Former studies have re-
ported good psychometric qualities for this questionnaire (Schwinger et
al., 2007, 2009). Sample items for each motivational regulation strategy
are displayed in Table 1. The Effort subscale of the German version of the
Motivated Strategies for Learning Questionnaire (MSLQ, Pintrich, Smith,
Garcia, & McKeachie, 1993; Wild & Schiefele, 1994) was used to assess
students' effort expenditure. The scale contains eight items, for instance,
“I work hard to do well in this class even if I don't like what we are
doing”. All items were answered on a five-point scale ranging from 1
(rarely) to 5 (very often). We assessed gender as a moderating variable.
2.1.3. Analysis procedure
First, we conducted multiple regression analyses in Mplus 7.4
(Muthén & Muthén, 1998–2015) with the eight motivational regulation
strategies as predictors and the effort subscale as dependent variable.
Next, we conducted relative weight analyses to identify which motiva-
tional regulation strategy represents the best predictor of effort expen-
diture. Relative weight analyses were conducted in R version 2.15.1 (R
Development Core Team, 2012) using the respective R code provided
126 M. Schwinger, N. Otterpohl / Learning and Individual Differences 53 (2017) 122–132

Table 1
Sample items for motivational regulation strategies.
Scales (numbers of items)
Enhancement of sit. interest (5)
Enhancement of pers. signif. (3)
Self-consequating (4)
Proximal goal setting (3)
Mastery self-talk (4)
Performance-approach self-talk (5)
Performance–avoidance self-talk (3)
Environmental control (3)
Sample items
I make learning more pleasant for
me by trying to arrange it playfully.
I look for connections between
the tasks and my life as such.
I promise myself that, after work,
I will do something that I like.
I approach work step-by-step in order
to get the feeling that I proceed well.
I persuade myself to work intensely
for the sake of learning.
I call my attention to the fact of how
important it is to obtain good grades.
I imagine that my classmates make
fun of my poor performance.
I make sure that distractions occur
as seldom as possible.
Study 1 were cross-sectional, so we cannot be sure about the causal re-
lation between motivational regulation strategy use and effort expendi-
ture. In Study 2, we therefore used a longitudinal design with repeated
measurement of effort expenditure as a dependent variable.2 Moreover,
we assessed conscientiousness as a further personal trait that might
moderate the relative effectiveness of motivational regulation strate-
gies. Based on our literature review and the findings from Study 1, we
formulated the following hypotheses.
Hypothesis 1. All motivational regulation strategies, except for the
strategy enhancement of personal significance, contribute uniquely to
the explanation of variance in students' self-reported effort.
Hypothesis 2. Goal-based motivational strategies predict a significant-
ly higher amount of variance in students' self-reported effort than inter-
est-enhancement strategies.
Hypothesis 3. Relative importance rank order remains stable when
controlling for students' initial motivation.

Hypothesis 4. Relative importance rank order of motivational strate-

gies will be only slightly different for male and female students. Highly
by Tonidandel and LeBreton (2011). This program allowed us to (a)
conscientious students will profit more from using goal-based motiva-
evaluate whether a particular relative weight was significantly different
tional strategies than students with low values in conscientiousness.
from zero, (b) examine whether relative weights were significantly dif-
ferent from each other, and (c) test whether a predictor's relative
weight differed significantly across two groups. This feature was used
for testing possible moderating effects of gender. 3.1. Method

3.1.1. Sample and procedure

2.2. Results
2.2.1. Descriptive statistics
Means, standard deviations, reliabilities, and intercorrelations can be
found in Table 2. Reliabilities for all scales were satisfying to good. Akin
to previous studies, the highest means were reported for performance-
approach self-talk and self-consequating, whereas enhancement of sit-
uational interest was the least frequently used strategy by students.
Scale intercorrelations were also in line with prior research. For exam-
ple, performance-approach and performance-avoidance self-talk were
highly associated with each other, but both strategies were uncorrelated
with the two interest enhancement strategies. All motivational strate-
gies were positively related to students' effort expenditure, albeit to a
different degree.
2.2.2. Relative importance of motivational strategies
Results from the relative weights analyses are displayed in Table 3.
Altogether, the eight motivational regulation strategies explained 39%
of the variance in effort expenditure. About one third of the total vari-
ance was explained by the strategy of mastery self-talk (35%), followed
by proximal goal setting (22%), and performance-approach self-talk
(17%). Environmental control accounted for 10% and self-consequating
for 7% of the variance, whereas performance-avoidance self-talk, en-
hancement of situational interest, and enhancement of personal signif-
icance explained b 5% each. Only the relative weight for enhancement of
personal significance was not significantly different from zero.
2.2.3. Moderating factors
Gender had a moderating effect on two strategies. First, the relative
weight of mastery self-talk was significantly higher for male (46%)
than for female students (28%). By contrast, however, the relative im-
portance of performance-avoidance self-talk was slightly higher for fe-
male (6%) compared to male students (1%).
In August 2010, 613 students (487 female; age: M = 23.88, SD =
5.04) from different German universities participated in another online
survey. Prior to the study, they received the same informed consent as
students in Study 1. The majority of students (60%) were enrolled in
psychology, whereas the remaining students were registered in other
subjects (e.g., mathematics, sports, law). All told, 48% of the participants
were in their first semester, whereas the other students were dispersed
across higher semesters. At the first time of measurement, students
responded to the motivational regulation questionnaire and also rated
their effort expenditure. Four months later, participants received an
email invitation for another online survey in which they were asked to
rate their effort expenditure again and also to report their study perfor-
mance. A subsample of 386 students participated at the second time of
measurement. Dropout was not significantly related to gender nor to
the use of any of the eight motivation regulation strategies and effort ex-
penditure at t1 (ps N 0.05). The relative weights analyses were conduct-
ed with the subsample of 386 participants. Data from this sample have
been used in a previous publication by Schwinger, Steinmayr, and
Spinath (2012).
3.1.2. Measures
Motivational regulation strategies and effort expenditure were mea-
sured with the same instruments as in Study 1. However, effort expen-
diture was measured at both times of measurement. Conscientiousness
was measured with the respective subscale from the German version of
the NEO-FFI (Borkenau & Ostendorf, 1993). It includes self-descriptive
statements (e.g., “I keep my belongings neat and clean.”) that partici-
pants respond to using a 1 (strongly disagree) to 5 (strongly agree)
Likert-type scale. As in Study 1, we also assessed gender as a potential
moderating variable.

3. Study 2
With the second study, we sought to replicate and extend the find-
ings from the first study in several ways. Most important, data from
2
We are aware that these are still correlational data, so we cannot be certain about the
causality of relationships. Nevertheless, due to the clear theoretical assumption that moti-
vational strategy use is supposed to influence one's effort exertion and due to the longitu-
dinal design and the option to control for prior effort, we deem it appropriate to use these
data as a cautious starting point for interpretation of the relationships in question.
 M. Schwinger, N. Otterpohl / Learning and Individual Differences 53 (2017) 122–132 127

Table 2
Means, standard deviations, reliabilities, and intercorrelations (Study 1).

α M (SD) (1) (2) (3) (4) (5) (6) (7) (8)

Enhancement of sit. interest (1) 0.83 2.19 (0.86) 1

Enhancement of pers. signif. (2) 0.87 3.01 (1.17) 0.48⁎⁎ 1
Self-consequating (3) 0.90 3.82 (1.07) 0.17⁎⁎ 0.14⁎⁎ 1
Proximal goal setting (4) 0.76 3.31 (1.03) 0.34⁎⁎ 0.27⁎⁎ 0.28⁎⁎ 1
Mastery self-talk (5) 0.75 2.70 (0.91) 0.34⁎⁎ 0.31⁎⁎ 0.22⁎⁎ 0.31⁎⁎ 1
Performance-approach self-talk (6) 0.89 3.48 (1.07) 0.04 0.06 0.27⁎⁎ 0.19⁎⁎ 0.36⁎⁎ 1
Performance–avoidance self-talk (7) 0.78 2.69 (0.95) 0.05 0.04 0.19⁎⁎ 0.08 0.27⁎⁎ 0.66⁎⁎ 1
Environmental control (8) 0.82 3.16 (0.91) 0.33⁎⁎ 0.20⁎⁎ 0.27⁎⁎ 0.44⁎⁎ 0.35⁎⁎ 0.15⁎⁎ 0.15⁎⁎ 1
Effort expenditure (9) 0.94 3.27 (0.79) 0.22⁎⁎ 0.15⁎⁎ 0.28⁎⁎ 0.42⁎⁎ 0.49⁎⁎ 0.34⁎⁎ 0.10⁎ 0.34⁎⁎
⁎ p b 0.05.
⁎⁎ p b 0.01.

3.2. Results scaled variable conscientiousness, the scale was divided by Median-
Split into high vs. low. Conscientiousness appeared to be a significant
3.2.1. Descriptive statistics moderator of the relative importance of proximal goal setting. This par-
All scales showed good reliabilities (see Table 4). The mean use of ticular strategy was more effective for less conscientious (8%) compared
motivational strategies was in a similar rank order as in Study 1, indicat- to highly conscientious students (0.3%).
ing again the most frequent use for the strategies performance-ap-
proach self-talk and self-consequating. Scale intercorrelations
4. Study 3
provided a sound picture of the expected associations among the differ-
ent motivational strategies with, for instance, moderately positive rela-
Unlike Studies 1 and 2, where the items measuring motivational reg-
tionships between mastery self-talk and the two interest enhancement
ulation strategies pointed to students' general strategy use across differ-
strategies. Zero-order correlations between the motivational regulation
ent situations, we sought to choose a more specific level of analysis in
strategies and students' reported effort at t2 were similar to those of
Study 3. Pertinent studies on cognitive regulation strategies have
Study 1. Although different in magnitude, the respective correlations
shown that correlations between strategy use and achievement fluctu-
were positive for all strategies except performance-avoidance self-talk.
ate considerably, depending on the specificity of the survey context
(Butler, 2002; Leopold & Leutner, 2002). As in Study 2, we also con-
3.2.2. Relative importance of motivational strategies trolled for students' initial effort. Regarding potential individual moder-
We followed the same analysis procedure as in Study 1. Effort ex- ators, we focused again on gender and conscientiousness. However,
penditure at t2 was predicted by the eight motivational regulation strat- based on findings by Wolters and Benzon (2013), who stated that stu-
egies and by effort expenditure from t1. Unsurprisingly, relative weights dents' engagement in motivational regulation is a function of their
analyses revealed that, by far, the greatest portion of variance in effort existing motivational beliefs and attitudes, we further added disposi-
expenditure t2 was explained by effort expenditure t1 (69%, see tional interest and achievement goal orientations to our list of potential
Table 5). 31% of the variance could be attributed to the eight motiva- moderators.
tional strategies. Among those, mastery self-talk appeared again as the
Hypothesis 1. In contrast to the remaining strategies, the strategies en-
most powerful predictor (11%), followed by proximal goal setting
hancement of situational interest, enhancement of personal signifi-
(5%), performance-approach self-talk (4%), environmental control
cance, and performance-avoidance self-talk will not contribute
(4%), and self-consequating (3%). The relative weights for enhancement
uniquely to the explanation of variance in students' self-reported effort.
of situational interest, enhancement of personal significance, and per-
formance-avoidance self-talk were not significantly different from Hypothesis 2. Goal-based motivational strategies predict a significant-
zero (b 2%). ly higher amount of variance in students' self-reported effort than inter-
est-enhancement strategies.
3.2.3. Moderating factors
Hypothesis 3. Relative importance rank order remains stable when
Gender did not moderate the relative effectiveness of any motiva-
controlling for students' initial motivation.
tional regulation strategy. To test the moderating effect of the metrically
Hypothesis 4. Relative importance rank order of motivational strate-
Table 3 gies will be the same for male and female students. Highly conscientious
Regression estimates, relative weights and explained criterion variance (%) for all motiva- students will profit more from using goal-based motivational strategies
tional regulation strategies (Study 1). than students with low values in conscientiousness. Highly interested
Motivational strategies students will profit more from using interest-enhancement strategies.
Goal-related self-talk will be most effective if the strategy-specific goal
B (SE) β RW %
matches the general goal orientation of the student (e.g., highly mastery
d
Enhancement of sit. interest 0.01 (0.04) 0.01 0.013 3.36 oriented students will profit most from using mastery self-talk).
Enhancement of pers. signif. −0.05 (0.03) −0.07 0.005 1.17
Self-consequating 0.06 (0.03) 0.09⁎ 0.028c,d 7.30
Proximal goal setting 0.18 (0.03) 0.23⁎⁎ 0.085a,b 21.98
Mastery self-talk 0.30 (0.04) 0.34⁎⁎ 0.135a 34.87 4.1. Method
Performance-approach self-talk 0.22 (0.04) 0.29⁎⁎ 0.068b,c 17.44
Performance–avoidance self-talk −0.19 (0.04) −0.23⁎⁎ 0.015d 3.77
4.1.1. Sample and procedure
Environmental control 0.08 (0.04) 0.09⁎ 0.039c 10.10
Explained variance R2 0.388 0.388 100 In October 2010, N = 301 students (56.5% female; Age: M = 17.55,
SD = 0.69) participated in this study. They were in their twelfth-grade
Notes. Relative weights holding a superscript are significantly different from zero at
p b 0.05. Values with different superscripts are significantly different at p b 0.05.
year from several high schools (German Gymnasium) in two federal
⁎ p b 0.05. German states. Prior to the study, informed consent was obtained
⁎⁎ p b 0.01. from the students as well as from their parents and teachers. The
128 M. Schwinger, N. Otterpohl / Learning and Individual Differences 53 (2017) 122–132

Table 4
Means, standard deviations, reliabilities, and intercorrelations (Study 2).

α M (SD) (1) (2) (3) (4) (5) (6) (7) (8) (9)

Enhancement of sit. interest (1) 0.88 2.57 (0.89) 1

Enhancement of pers. signif. (2) 0.89 3.34 (1.01) 0.54⁎⁎ 1
Self-consequating (3) 0.93 3.51 (1.04) 0.29⁎⁎ 0.15⁎⁎ 1
Proximal goal setting (4) 0.84 3.35 (0.93) 0.43⁎⁎ 0.20⁎⁎ 0.38⁎⁎ 1
Mastery self-talk (5) 0.76 2.97 (0.85) 0.35⁎⁎ 0.33⁎⁎ 0.27⁎⁎ 0.26⁎⁎ 1
Performance-approach self-talk (6) 0.87 3.67 (0.87) 0.00 −0.04 0.27⁎⁎ 0.07 0.37⁎⁎ 1
Performance–avoidance self-talk (7) 0.79 2.29 (0.96) −0.09⁎ −0.12⁎⁎ 0.11⁎⁎ −0.05 0.19⁎⁎ 0.51⁎⁎ 1
Environmental control (8) 0.74 3.22 (0.96) 0.24⁎⁎ 0.19⁎⁎ 0.31⁎⁎ 0.28⁎⁎ 0.28⁎⁎ 0.19⁎⁎ 0.04 1
Effort expenditure t1 (9) 0.83 3.55 (0.70) 0.29⁎⁎ 0.15⁎⁎ 0.27⁎⁎ 0.31⁎⁎ 0.45⁎⁎ 0.34⁎⁎ 0.05 0.28⁎⁎ 1
Effort expenditure t2 (10) 0.83 3.60 (0.70) 0.43⁎⁎ 0.20⁎⁎ 0.31⁎⁎ 0.35⁎⁎ 0.46⁎⁎ 0.31⁎⁎ 0.05 0.31⁎⁎ 0.83⁎⁎
⁎ p b 0.05.
⁎⁎ p b 0.01.

information in the consent forms was similar to the ones in Studies 1
and 2. Students had to bring along the consent form signed by their par-
ents at the day of testing. In their final years in Gymnasium, students can
choose two subjects as advanced-level courses, which means that they
receive more lessons (usually six) per week. Accordingly, the exam
grades from the advanced-level courses have the greatest impact on
the grade of their school exit exam. As a consequence, students in ad-
vanced-level courses are likely to be more intrinsically motivated than
students who have not chosen their own subjects. Participants were
asked to complete questionnaires referring to their preparation for an
upcoming exam in one of their advanced-level courses. A total of
42.6% of the students had mathematics as an advanced-level course,
26.9% had German, 15.9% had English, and 14.6% had history.
Data from this sample have been used in a previous publication by
Schwinger and Stiensmeier-Pelster (2012).
Presumed moderating factors were assessed at the first measure-
ment point. Conscientiousness was assessed with the same instrument
as in Study 2. Dispositional interest in the respective subject was mea-
sured by four self-created items (e.g., “I like mathematics”; “I am inter-
ested in mathematics”). Students' goal orientations were assessed by
means of the German Scales for the Assessment of Learning and Perfor-
mance Goals (Spinath, Stiensmeier-Pelster, Schöne, & Dickhäuser,
2002). This measure assesses mastery and performance-avoidance
goals with eight items each, and performance-approach goals with
seven items in each domain (see Spinath & Steinmayr, 2012, for details).
These scales have proven convincing convergent and discriminate
validities with related constructs (e.g., achievement motivation, self-ef-
ficacy beliefs, and test anxiety), as well as good predictive validity of
goal orientations for school grades. All items were responded to on a
rating scale ranging from 1 (strongly disagree) to 5 (strongly agree).

4.1.2. Measures and procedure
The students responded to items that referred to an upcoming exam
(specific event) in one of their advanced-level courses (specific subject).
Questionnaires were filled in during regular lessons in the respective
advanced-level course. About two weeks before the exam (t1), we
assessed students' current effort (i.e., their willingness to learn for the
upcoming exam) with three items, such as “My motivation to learn
for the upcoming exam is currently... [1] very low to [5] very high.” The
teachers had blocked the last lesson before the exam, so in all courses,
the second measurement point took place in the next to last lesson be-
fore the exam (t2). This resulted in comparable conditions for all sub-
jects, with a similar time lag of two to four days before the exam. At
this second time of measurement, we measured students' subject- and
event-specific use of motivational regulation strategies (i.e., their use
of motivational strategies in the days before the exam) with the ques-
tionnaire by Schwinger et al. (2007, 2009). Moreover, we also asked
the students again to state their current effort for the upcoming exam.
4.2. Results
4.2.1. Descriptive statistics
The inspection of scale means, standard deviations, reliabilities, and
intercorrelations yielded a very consistent impression across the three
samples. In Study 3, performance-approach self-talk and self-
consequating again showed the highest means, Cronbach alphas were
mostly good, the eight motivational strategies were intertwined in a
highly similar pattern compared to Studies 1 and 2, and all motivational
strategies were positively related to students' current effort at t2 (see
Table 6).
4.2.2. Relative importance of motivational strategies
The rank order of the relative weights was highly similar to those in
Studies 1 and 2, independent of whether initial motivation was included
in the regression model or not (see Table 7). Naturally, the highest
amount of variance in current effort t2 was explained by current effort
t1 with 59%. The remaining 41% of the explained variance could be

Table 5
relative weights and percentage of explained criterion variance (%) for all motivational regulation strategies plus initial motivation (Study 2).

Motivational strategies Motivational strategies + initial motivation

B (SE) β RW % B (SE) β RW %
a
Initial motivation (t1) – – – – – 0.77 (0.04) 0.757 0.484 69.28
Enhancement of sit. interest 0.01 (0.05) 0.01 0.016b 5.14 −0.02 (0.03) −0.024 0.013 1.92
Enhancement of pers. signif. −0.08 (0.04) −0.11 0.005 1.54 −0.02 (0.03) −0.024 0.003 0.48
Self-consequating 0.05 (0.04) 0.07 0.032b 10.03 0.02 (0.02) 0.035 0.026c 3.69
Proximal goal setting 0.11 (0.04) 0.15⁎⁎ 0.050a,b 15.83 0.03 (0.03) 0.038 0.036b,c 5.10
Mastery self-talk 0.28 (0.05) 0.35⁎⁎ 0.120a 37.62 0.07 (0.03) 0.083 0.074b 10.69
Performance-approach self-talk 0.12 (0.05) 0.15⁎⁎ 0.048b 15.01 −0.03 (0.03) −0.033 0.029c 4.24
Performance–avoidance self-talk −0.11 (0.04) −0.15⁎⁎ 0.006 2.04 0.01 (0.03) 0.002 0.001 0.23
Environmental control 0.09 (0.04) 0.13⁎⁎ 0.041b 12.78 0.04 (0.03) 0.060 0.030c 4.36
Explained variance R2 0.318 0.318 100 0.699 0.699 100

Notes. Relative weights holding a superscript are significantly different from zero at p b 0.05. Values with different superscripts are significantly different at p b 0.05.
⁎⁎ p b 0.01.
 M. Schwinger, N. Otterpohl / Learning and Individual Differences 53 (2017) 122–132 129

Table 6
Means, standard deviations, reliabilities, and intercorrelations (Study 3).

α M (SD) (1) (2) (3) (4) (5) (6) (7) (8) (9)

Enhancement of sit. interest (1) 0.82 2.37 (0.82) 1

Enhancement of pers. signif. (2) 0.79 2.66 (0.99) 0.56⁎⁎ 1
Self-consequating (3) 0.86 3.40 (0.96) 0.29⁎⁎ 0.20⁎⁎ 1
Proximal goal setting (4) 0.74 2.95 (0.88) 0.39⁎⁎ 0.22⁎⁎ 0.44⁎⁎ 1
Mastery self-talk (5) 0.70 2.86 (0.74) 0.40⁎⁎ 0.36⁎⁎ 0.38⁎⁎ 0.39⁎⁎ 1
Performance-approach self-talk (6) 0.84 3.47 (0.83) 0.08 0.11⁎ 0.33⁎⁎ 0.28⁎⁎ 0.35⁎⁎ 1
Performance–avoidance self-talk (7) 0.71 2.57 (0.79) 0.28⁎⁎ 0.17⁎⁎ 0.16⁎ 0.23⁎⁎ 0.32⁎⁎ 0.51⁎⁎ 1
Environmental control (8) 0.74 3.06 (0.83) 0.32⁎⁎ 0.23⁎⁎ 0.39⁎⁎ 0.46⁎⁎ 0.52⁎⁎ 0.35⁎⁎ 0.22⁎⁎ 1
Current effort t1 (9) 0.78 3.08 (0.97) 0.08 0.09 0.11 0.20⁎⁎ 0.28⁎⁎ 0.23⁎⁎ 0.21⁎⁎ 0.15⁎⁎ 1
Current effort t2 (10) 0.80 3.00 (1.01) 0.17⁎⁎ 0.18⁎⁎ 0.17⁎⁎ 0.29⁎⁎ 0.35⁎⁎ 0.25⁎⁎ 0.18⁎⁎ 0.24⁎⁎ 0.50⁎⁎
⁎ p b 0.05.
⁎⁎ p b 0.01.

attributed mostly to mastery self-talk (13%), followed by proximal goal
setting (9%), and performance-approach self-talk (6%). The relative
weights of the other motivational regulation strategies were not signif-
icantly different from zero.
4.2.3. Moderating factors
Gender was found to moderate the relative effectiveness of environ-
mental control. This particular strategy was more effective for male
(12%) compared to female students (1%). Conscientiousness, disposi-
tional interest, and the three achievement goal orientations did not
moderate the relative effectiveness of any motivational regulation
strategy.
5. Summary of relative weights results
Findings from the relative weights analyses yielded a consistent pat-
tern across the three studies. Among the eight motivational regulation
strategies examined here, mastery self-talk explained most of the vari-
ance in motivational outcomes, followed by proximal goal setting and
performance-approach self-talk. Environmental control and self-
consequating were mostly ranked in middle positions, whereas perfor-
mance-avoidance self-talk, enhancement of situational interest, and en-
hancement of personal significance showed very low relative weights in
each study. In Studies 2 and 3, this basic pattern of relative strategy im-
portance did not change when we controlled for the effects of students'
initial motivation in our analyses. The results across all studies are illus-
trated in Fig. 1.
6. General discussion
In this article, we sought to further our understanding of the relative
importance of motivational regulation strategies. Across three empirical
studies with German high school and college students, and by using rel-
ative weights analysis as a sophisticated statistical procedure, we aimed
to clarify which one of a series of different motivational strategies works
best for sustaining students' effort and persistence. Moreover, we were
interested in whether relative importance might be different depending
on individual differences between students regarding gender, conscien-
tiousness, dispositional interest, and achievement goal orientations.
6.1. Relative importance of motivational regulation strategies
The relative weights procedure used here provided two important
kinds of information, namely (a) whether a relative weight is signifi-
cantly different from zero and (b) significant differences between rela-
tive weights. With respect to the first point, it is interesting to note that
the relative weights of enhancement of personal significance and of per-
formance-avoidance self-talk (except Study 1) did not differ significant-
ly from zero. That is, these two strategies did not work in the expected
way because students who used them did not experience a significant
increase in motivation to learn. At this point, the added value of the rel-
ative weights approach compared to zero-order correlations or tradi-
tional multiple regression models becomes apparent because both
motivational strategies had shown significantly positive correlations
to the effort measures in all three studies. However, these positive rela-
tionships disappeared when the unique contribution of each predictor
was considered by computing relative weights.
With respect to the second point, including information on statistical
differences between predictors' relative weights led to a more fine-
grained interpretation of the top three ranked motivational regulation
strategies. Statistical significance of these three strategies' relative
weights were slightly different in the first two studies compared to
the last study. In Studies 1 and 2, the relative weight for mastery self-
talk was significantly different from the weight for performance-ap-
proach self-talk, but not from the weight for proximal goal setting.
The relative weights for proximal goal setting and performance-
approach self-talk, in turn, were not significantly different. In Study 3,
relative weights for all three strategies did not differ significantly from

Table 7
Relative weights and percentage of explained criterion variance (%) for all motivational regulation strategies, plus initial motivation (Study 3).

Motivational strategies Motivational strategies + initial motivation

B (SE) β RW % B (SE) β RW %
a
Initial motivation (t1) – – – – – 0.44 (0.05) 0.42 0.189 59.60
Enhancement of sit. interest −0.02 (0.09) −0.02 0.006 3.83 0.02 (0.08) 0.01 0.006 1.98
Enhancement of pers. signif. 0.06 (0.07) 0.05 0.009 4.14 0.05 (0.06) 0.05 0.008 2.82
Self-consequating −0.04 (0.07) −0.04 0.006 3.54 −0.02 (0.06) −0.02 0.005 1.66
Proximal goal setting 0.19 (0.08) 0.17⁎⁎ 0.035a 21.83 0.13 (0.07) 0.12⁎ 0.028a 8.78
Mastery self-talk 0.32 (0.09) 0.24⁎⁎ 0.057a 35.22 0.20 (0.09) 0.15⁎ 0.041a 13.09
Performance-approach Self-talk 0.15 (0.08) 0.13 0.025a 15.83 0.10 (0.08) 0.09 0.019a 6.22
Performance–avoidance self-talk 0.01 (0.08) 0.01 0.008 5.09 −0.04 (0.08) −0.03 0.005 1.89
Environmental control −0.01 (0.08) −0.01 0.014 8.51 0.02 (0.08) 0.01 0.012 3.95
Explained variance R2 0.161 0.161 100 0.316 0.316 100

Notes. Relative weights holding a superscript are significantly different from zero at p b 0.05. Values with different superscripts are significantly different at p b 0.05.
⁎ p b 0.05.
⁎⁎ p b 0.01.
130 M. Schwinger, N. Otterpohl / Learning and Individual Differences 53 (2017) 122–132
Fig. 1. Summary of relative weights results across all studies.
each other. The next ranked motivational strategies—environmental
control and self-consequating—had significantly lower relative weights
than did the top three strategies, but only in Studies 1 and 3.
Despite these small limitations with respect to statistical signifi-
cance, we think it is reasonable to conclude that mastery self-talk, prox-
imal goal setting, and performance-approach self-talk are particularly
effective strategies for regulating motivation for an average student
in a typical learning situation. Environmental control and self-
consequating may also be labelled as effective and useful motivational
strategies. Whether they are really least effective compared to the top
three strategies cannot be concluded with certainty, so further studies
are needed to replicate the statistical differences partially found here.
What can be concluded with certainty, however, is that performance-
avoidance self-talk and the two interest enhancement strategies repre-
sent rather ineffective ways to regulate students' learning motivation.
Moreover, it is striking that this rank order of motivational strategies
was found across different samples (high school vs. college), different
designs (cross-sectional vs. longitudinal), different covariates (control-
ling vs. not controlling for initial motivation), and different contexts
(school in general vs. course-specific), so our findings demonstrated
high stability and replicability. It has to be noted, however, that previous
studies provided empirical evidence for the general effectiveness of dif-
ferent forms of interest enhancement as well (Leutner et al., 2001;
Sansone et al., 1992). Future studies should explore the reasons for
this incompatibility of findings.
6.2. Effectiveness of goal-based motivational regulation strategies
How can our results be explained and what are the psychological
mechanisms behind them? Obviously, students were more effective in
regulating their motivation when they relied on strategies that were
based in some way on goals. Numerous studies have shown that mas-
tery and performance-approach goals are adaptive motivational orien-
tations as they are related to several important educational processes
and outcomes, such as higher interest (Hulleman, Durik, Schweigert, &
Harackiewicz, 2008; Vansteenkiste et al., 2010), greater performance,
and adaptive attribution patterns after failure. These positive outcomes
can be easily explained by the fact that pursuing those kinds of approach
goals provides both energy and direction for the individual student
(Elliot, 1999). In this regard, our findings seem to indicate that recalling
and remembering of these goals can re-initiate their inherent power
and energy, making it available again in the current learning situation.
We do not know whether the amount of motivation elicited by the ap-
proach-oriented achievement goals is the same every time the goal is
consciously activated. Two different settings may be possible in this re-
gard. First, the motivational energy provided is equally powerful in
every stage of recalling, thus of regulating motivation. Second, the mo-
tivation-enhancing effects become smaller over time. Although our de-
sign does not allow for testing these competing ideas, the clearly highest
relative weights for goal-related motivational strategies lead us to sug-
gest that the first assumption might describe the self-regulation realities
of most students more accurately. Nevertheless, this issue needs to be
considered in more detail in further studies.
Taken together, it seems reasonable that goal-based motivational
strategies are effective because goals have a strong power to provide en-
ergy and to direct that energy in the right way. As the case of perfor-
mance-avoidance self-talk has demonstrated, however, an important
prerequisite is to formulate goals in positive, approach-related terms,
in order to become effective tools for motivation regulation. But why
are goal-related strategies more effective than interest-enhancement
strategies? We think there are two possible reasons for this. First,
goal-based vs. interest-based strategies might be used at different levels
of motivational deficits. Schwinger and Stiensmeier-Pelster (2012)
noted that students' motivation may only be temporarily lowered or
fundamentally affected. Although it is difficult to categorize the eight
motivational regulation strategies examined here into ones better suit-
ed for temporary vs. fundamental motivational problems, it seems rea-
sonable that students try to solve temporal motivational problems by
remembering an already established goal and by re-focusing on it in-
stead of searching for a new, fundamental reason to persist in the task.
The second argument for why we think it is plausible that goal-related
strategies are more effective than interest-enhancement strategies
points to a similar issue. The strategy of enhancement of personal signif-
icance aims at building a new motivational basis that is intrinsic and
long-lasting. If this strategy is applied successfully, it should largely in-
fluence students' effort and persistence. As we did not observe such ef-
fects, however, we concluded that the strategy does not work in many
cases. A possible explanation for this might be that the task incorporated
in the strategy was too difficult for the students. If they were to have
seen a personal significance in the task, they would have probably
seen it before, and would thus have been intrinsically motivated and
would have had no need for motivational regulation at all. But if a stu-
dent does not see any personal significance in the task, it appears in-
stead to be a rather tremendous undertaking to create one. This
reasoning is in line with research on academic motivation in general.
Hidi and Harackiewicz (2000), for instance, have pointed out that it
may be easier to motivate students who already have external incen-
tives, or by setting short- and long-term goals, rather than trying to in-
crease their intrinsic motivation.
6.3. Moderating effects of individual factors
Our search for personality factors that might moderate the relative
importance of motivational strategies was guided by the assumption
that motivational regulation is most effective when the strategy chosen
matches a learner's basic motivational beliefs and attitudes (Wolters &
Benzon, 2013). We supposed, for example, that highly conscientious
students would be relatively more effective at using goal-related moti-
vational strategies because conscientiousness has been found to be as-
sociated with ambitious goal setting and commitment (Klein & Lee,
2006). Quite surprising, however, our interaction analyses revealed
only single, isolated effects that did not point to substantial moderation
effects for any of the personality factors considered. This means that the
established relative importance of motivational strategies did not seem
to be affected by students' gender, conscientiousness, dispositional in-
terest, or achievement goal orientations. We are not sure how reliable
these findings are because it is possible that the power to detect interac-
tion effects was too low in our studies. Cases with pronounced profiles
(i.e., individuals with high values on one dimension and very low values
on the other) are often rare in observational studies, which negatively
 M. Schwinger, N. Otterpohl / Learning and Individual Differences 53 (2017) 122–132
affects their power to detect meaningful interactions (McClelland &
Judd, 1993). Given this statistical caveat, we are cautious in interpreting
our findings on moderation effects. However, we think that the results
clearly suggest to focus on goal-related strategies in trainings on moti-
vational regulation. Moreover, it does not seem important to take
students' personality profile into account when conducting such train-
ings. We clearly need to be cautious with such interpretations as we
measured several variables only in some of our studies and as we relied
on different samples in those studies. This could mean, for instance, that
achievement goals have a substantial moderating impact for university
students, albeit we found no effects among high-school students in
Study 3. In sum, however, there were no convincing hints for modera-
tion effects of personality variables in our data, so we think it may be
more promising instead to pay attention to contextual differences,
such as boring vs. difficult tasks (Schwinger & Stiensmeier-Pelster,
2012; Wolters, 1998).
6.4. Practical implications
Our findings underline the importance of setting and continuously
endorsing achievement-related goals in both school and university con-
texts. It may therefore be promising that educational practitioners such
as teachers and school psychologists encourage their students to set
goals for their studying as well as to constantly reflect and recall those
goals, especially in times when they feel unmotivated to learn. Results
from a recent meta-analysis support the idea that teachers may exert
significant influence on students' motivation. Lazowski and Hulleman
(2016) examined the effects of motivation interventions in education.
They found that interventions were generally effective, with an average
mean effect size of d = 0.49 (95% confidence interval = [0.43, 0.56])
and that there were no statistically significant differences in effect size
due to theoretical framework of the intervention, age of participants (el-
ementary through postsecondary students), or type of dependent vari-
able (performance, behavior, self-reported motivation). Integrating our
findings reported here and the meta-analytic results provided by
Lazowski and Hulleman (2016) may thus justify our claim that teachers
should instruct their students to adopt certain goals while studying.
However, this recommendation can be further differentiated on the
basis of evidence on achievement goal structures. Several studies in
this field have shown—in accordance with our results on goal-related
self-talk—that promoting mastery and/or approach goals in the class-
room is most adaptive for students' academic adjustment (e.g.,
Rolland, 2012; Wolters, 2004).
6.5. Limitations and suggestions for further research
Our findings are limited in several ways. First, we investigated older
high school and college students, leaving open the question of whether
younger students would show the same relative importance rank order.
There is an overall lack of studies on secondary school students' self-reg-
ulation of motivation, which needs to be addressed in future studies.
Second, our analyses and interpretations are completely based on self-
report data. Many authors have criticized these kind of data since they
only capture the remembered frequency of strategy use and may thus
not reflect a valid measure of students' self-regulated learning (e.g.,
Winne, 2010). It would therefore be an important research endeavor
to find out whether our conclusions on the relative importance of moti-
vation regulation strategies can be supported in designs using behavior-
al, observational, and/or qualitative data. Third, we focused exclusively
on individual moderators, so we do not know how certain individual
characteristics might interact with varying contextual conditions. It
may be possible, for instance, that highly conscientious students react
differently to motivational deficits elicited by a difficult task compared
to less conscientious individuals. Providing a deeper understanding of
these interactions is an important task for future research. However,
such interaction analyses might require very large sample sizes, which
131
points to the fourth limitation of our studies. Although the three sam-
ples were large enough to identify the overall rank order of motivational
strategies, they might have been too small to detect meaningful interac-
tion effects (McClelland & Judd, 1993). Fifth, we do not know much
about why some strategies have been found to be more effective than
others. Possible reasons for this may refer to lower metacognitive
knowledge about how to use certain motivational strategies appropri-
ately (Thillmann, Gößling, Marschner, Wirth, & Leutner, 2013) or to
the fact that some strategies have been applied in a qualitative better
way than others (Schwinger et al., 2009). Future studies should exam-
ine these issues in more detail. Despite these limitations, however, rep-
licating the relative importance across three samples under varying
conditions, and by using relative weights analysis as a sophisticated sta-
tistical technique, clearly represents important strengths of this study.
Our findings might provide a promising starting point for developing
basic training modules for improving students' motivation regulation
capacities.
References
Bandura, A., & Schunk, D. (1981). Cultivating competence, self-efficacy and intrinsic inter-
est through proximal self-motivation. Journal of Personality and Social Psychology, 41,
586–598. http://dx.doi.org/10.1037/0022-3514.41.3.586.
Borkenau, P., & Ostendorf, F. (1993). NEO-Fünf-Faktoren Inventar (NEO-FFI) [NEO five-fac-
tor inventory]. Göttingen, Germany: Verlag für Psychologie.
Butler, D. L. (2002). Qualitative approaches to self-regulated learning: Contributions
and challenges. Educational Psychologist, 37, 59. http://dx.doi.org/10.1207/
S15326985EP3701_7.
Cleary, T. J., & Chen, P. P. (2009). Self-regulation, motivation, and math achievement in
middle school: Variations across grade level and math context. Journal of School
Psychology, 47, 291–314. http://dx.doi.org/10.1016/j.jsp.2009.04.002.
Credè, M., & Kuncel, N. (2008). Study habits, skills, and attitudes: The third pillar
supporting collegiate academic performance. Perspectives on Psychological Science, 3,
425–453. http://dx.doi.org/10.1111/j.1745-6924.2008.00089.x.
Elliot, A. J. (1999). Approach and avoidance motivation and achievement goals.
Educational Psychologist, 34, 169–189. http://dx.doi.org/10.1207/s15326985ep3403_3.
Engelschalk, T., Steuer, G., & Dresel, M. (2015). Wie spezifisch regulieren Studierende ihre
Motivation bei unterschiedlichen Anlässen? [Situation-specific motivation regula-
tion: How specifically do students regulate their motivation for different situations?
]. Zeitschrift für Entwicklungspsychologie und Pädagogische Psychologie, 47, 14–23.
http://dx.doi.org/10.1026/0049-8637/a000120.
Grunschel, C., Patrzek, J., & Fries, S. (2013). Exploring different types of academic delayers:
A latent profile analysis. Learning and Individual Differences, 23, 225–233. http://dx.
doi.org/10.1016/j.lindif.2012.09.014.
Hidi, S., & Harackiewicz, J. M. (2000). Motivating the academically unmotivated: A critical
issue for the 21st century. Review of Educational Research, 70, 151–179. http://dx.doi.
org/10.3102/00346543070002151.
Hulleman, C. S., & Harackiewicz, J. M. (2009). Promoting interest and performance in high
school science classes. Science, 326, 1410–1412. http://dx.doi.org/10.1126/science.
1177067.
Hulleman, C. S., Durik, A. M., Schweigert, S. A., & Harackiewicz, J. M. (2008). Task values,
achievement goals, and interest: An integrative analysis. Journal of Educational
Psychology, 100, 398–416. http://dx.doi.org/10.1037/0022-0663.100.2.398.
Jackson, H., & Molloy, G. (1985). Some effects of feedback alone and four types of self-
consequation on selected measures of problem solving. Perceptual and Motor Skills,
61, 1005–1006. http://dx.doi.org/10.2466/pms.1985.61.3.1005.
Johnson, J. W., & LeBreton, J. M. (2004). History and use of relative importance indices in
organizational research. Organizational Research Methods, 7, 238–257. http://dx.doi.
org/10.1177/1094428104266510.
Klein, H. J., & Lee, S. (2006). The effects of personality on learning: The mediating role
of goal setting. Human Performance, 19, 43–66. http://dx.doi.org/10.1207/
s15327043hup1901_3.
Kobasa, S. C. (1979). Stressful life events, personality, and health: An inquiry into hardi-
ness. Journal of Personality and Social Psychology, 37, 1–11. http://dx.doi.org/10.
1037/0022-3514.37.1.1.
Lazowski, R. A., & Hulleman, C. S. (2016). Motivation interventions in education: A meta-
analytic review. Review of Educational Research, 86, 602–640. http://dx.doi.org/10.
3102/0034654315617832.
Leopold, C., & Leutner, D. (2002). Der Einsatz von Lernstrategien in einer konkreten
Lernsituation [The usage of learning strategies in a concrete learning situation].
Zeitschrift für Pädagogik, 45(Beiheft), 240–258.
Leutner, D., Barthel, A., & Schreiber, B. (2001). Studierende können lernen, sich selbst zum
Lernen zu motivieren: Ein Trainingsexperiment. [Students can learn to motivate
themselves for learning—A training experiment]. Zeitschrift für Pädagogische
Psychologie, 15, 155–167. http://dx.doi.org/10.1024//1010-0652.15.34.155.
McCann, E. J., & Garcia, T. (1999). Maintaining motivation and regulating emotion: Mea-
suring individual differences in academic volitional strategies. Learning and Individual
Differences, 11, 259–279. http://dx.doi.org/10.1016/S1041-6080(99)80003-X.
McClelland, G. H., & Judd, C. M. (1993). Statistical difficulties of detecting interactions and
moderator effects. Psychological Bulletin, 114, 376–390. http://dx.doi.org/10.1037//
0033-2909.114.2.376.
132 M. Schwinger, N. Otterpohl / Learning and Individual Differences 53 (2017) 122–132
Muthén, L. K., & Muthén, B. O. (1998–2015). Mplus user's guide (7th ed.). Los Angeles, CA:
Muthén & Muthén.
Pintrich, P. R. (2004). A conceptual framework for assessing student motivation and self-
regulated learning in college students. Educational Psychology Review, 16, 385–407.
http://dx.doi.org/10.1007/s10648-004-0006-x.
Pintrich, P. R., Smith, D., Garcia, T., & McKeachie, W. (1993). Predictive validity and
reliability of the Motivated Strategies for Learning Questionnaire. Educational
and Psychological Measurement, 53, 801–813. http://dx.doi.org/10.1177/
0013164493053003024.
Pychyl, T. A., Lee, J. M., Thibodeau, R., & Blunt, A. (2000). Five days of emotion: An expe-
rience sampling study of undergraduate student procrastination. Journal of Social
Behavior and Personality, 15, 239–254.
R Development Core Team (2012). R: A language and environment for statistical computing.
Vienna, Austria: R Foundation for Statistical Computing.
Roeser, R. W., Midgley, C., & Urdan, T. C. (1996). Perceptions of the school psychological
environment and early adolescents' psychological and behavioral functioning in
school: The mediating role of goals and belonging. Journal of Educational Psychology,
88, 408–422. http://dx.doi.org/10.1037/0022-0663.88.3.408.
Rolland, R. G. (2012). Synthesizing the evidence on classroom goal structures in middle
and secondary schools a meta-analysis and narrative review. Review of Educational
Research, 82, 396. http://dx.doi.org/10.3102/0034654312464909.
Sansone, C., & Thoman, D. B. (2005). Interest as the missing motivator in self-regulation.
European Psychologist, 10, 175–186. http://dx.doi.org/10.1027/1016-9040.10.3.175.
Sansone, C., Weir, C., Harpster, L., & Morgan, C. (1992). Once a boring task always a boring
task? Interest as a self-regulatory mechanism. Journal of Personality and Social
Psychology, 63, 379–390. http://dx.doi.org/10.1037/0022-3514.63.3.379.
Sansone, C., Wiebe, D. J., & Morgan, C. (1999). Self-regulating interest: The moderating
role of hardiness and conscientiousness. Journal of Personality, 67, 701–733. http://
dx.doi.org/10.1111/1467-6494.00070.
Schwinger, M., & Stiensmeier-Pelster, J. (2012). Effects of motivational regulation on ef-
fort and achievement: A mediation model. International Journal of Educational
Research, 56, 35–47. http://dx.doi.org/10.1016/j.ijer.2012.07.005.
Schwinger, M., von der Laden, T., & Spinath, B. (2007). Strategien zur Motivationsregulation
und ihre Erfassung [Motivational regulation strategies and their measurement].
Zeitschrift für Entwicklungspsychologie und Pädagogische Psychologie, 39, 57–69. http://
dx.doi.org/10.1026/0049-8637.39.2.57.
Schwinger, M., Steinmayr, R., & Spinath, B. (2009). How do motivational regulation strat-
egies affect achievement: Mediated by effort management and moderated by intelli-
gence. Learning and Individual Differences, 19, 621–627. http://dx.doi.org/10.1016/j.
lindif.2009.08.006.
Schwinger, M., Steinmayr, R., & Spinath, B. (2012). Not all roads lead to Rome—Comparing
different types of motivational regulation profiles. Learning and Individual Differences,
22, 269–279. http://dx.doi.org/10.1016/j.lindif.2011.12.006.
Spinath, B., & Steinmayr, R. (2012). The roles of competence beliefs and goal orientations
for change in intrinsic motivation. Journal of Educational Psychology, 104, 1135–1148.
http://dx.doi.org/10.1037/a0028115.
Spinath, B., Stiensmeier-Pelster, J., Schöne, C., & Dickhäuser, O. (2002). Die Skalen zur
Erfassung von Lern-und Leistungsmotivation (SELLMO) [Scales for the Measurement of
Learning and Achievement Motivation]. Göttingen, Germany: Hogrefe.
Thillmann, H., Gößling, J., Marschner, J., Wirth, J., & Leutner, D. (2013). Metacognitive
knowledge about and metacognitive regulation of strategy use in self-regulated sci-
entific discovery learning: New methods of assessment in computer-based learning
environments. In R. Azevedo, & V. Aleven (Eds.), International handbook of metacogni-
tion and learning technologies (pp. 575–588). New York, NY: Springer.
Tonidandel, S., & LeBreton, J. M. (2011). Relative importance analysis: A useful supple-
ment to regression analysis. Journal of Business and Psychology, 26, 1–9. http://dx.
doi.org/10.1007/s10869-010-9204-3.
Vansteenkiste, M., Smeets, S., Soenens, B., Lens, W., Matos, L., & Deci, E. L. (2010). Auton-
omous and controlled regulation of performance-approach goals: Their relations to
perfectionism and educational outcomes. Motivation and Emotion, 34, 333–353.
http://dx.doi.org/10.1007/s11031-010-9188-3.
Wild, K. P., & Schiefele, U. (1994). Lernstrategien im Studium. Ergebnisse zur
Faktorenstruktur und Reliabilität eines neuen Fragebogens. Zeitschrift für
Differentielle und Diagnostische Psychologie, 15, 185–200.
Winne, P. H. (2010). Improving measurements of self-regulated learning. Educational
Psychologist, 45, 267–276. http://dx.doi.org/10.1080/00461520.2010.517150.
Wolters, C. A. (1998). Self-regulated learning and college students' regulation of motiva-
tion. Journal of Educational Psychology, 90, 224–235. http://dx.doi.org/10.1037/0022-
0663.90.2.224.
Wolters, C. A. (1999). The relation between high school students' motivational regu-
lation and their use of learning strategies, effort, and classroom performance.
Learning and Individual Differences, 11, 281–299. http://dx.doi.org/10.1016/
S1041-6080(99)80004-1.
Wolters, C. A. (2003). Regulation of motivation: Evaluating an underemphasized aspect of
self-regulated learning. Educational Psychologist, 38, 189–205. http://dx.doi.org/10.
1207/S15326985EP3804_1.
Wolters, C. A. (2004). Advancing achievement goal theory: Using goal structures and goal
orientations to predict students' motivation, cognition, and achievement. Journal of
Educational Psychology, 96, 236–250. http://dx.doi.org/10.1037/0022-0663.96.2.236.
Wolters, C. A., & Benzon, M. B. (2013). Assessing and predicting college students' use of
strategies for the self-regulation of motivation. The Journal of Experimental
Education, 81, 199–221. http://dx.doi.org/10.1080/00220973.2012.699901.
Zimmerman, B. J., & Martinez-Pons, M. (1986). Development of a structured interview for
assessing student use of self-regulated learning strategies. American Educational
Research Journal, 23, 614–628. http://dx.doi.org/10.3102/00028312023004614.