Learning and Individual Differences 50 (2016) 252–259

Contents lists available at ScienceDirect

Learning and Individual Differences

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

The longitudinal influences of peers, parents, motivation, and

mathematics course-taking on high school math achievement
John Mark Froiland a,⁎,1, Mark L. Davison b
a
University of Northern Colorado, United States
b
Department of Educational Psychology, University of Minnesota, United States

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

Article history: Little is known about how parent expectations and math motivation work in concert with math course taking to
Received 11 March 2015 promote math achievement. This longitudinal structural equation modeling (SEM) study examined expectancy-
Received in revised form 30 June 2016 value and self-determination theory motivation constructs and math development among a nationally represen-
Accepted 30 July 2016
tative sample of U.S. high school students. The role of mathematics course-taking was also examined. As predict-
ed, parent expectations, student expectations, and peer interest predicted math intrinsic motivation in 9th grade,
Keywords:
Adolescents
which predicted student mathematics achievement in 11th grade, even when controlling for SES, race/ethnicity,
Expectations gender and prior math achievement. Intrinsic motivation for math and parent expectations also predicted taking
Mathematics higher-level math courses (e.g., trigonometry or calculus) over the next 2.5 years, which predicted further math
Motivation achievement. Parent expectations were a stronger predictor than student expectations of intrinsic motivation for
Parent involvement math, course taking, and achievement. Implications for math achievement interventions are discussed.
© 2016 Elsevier Inc. All rights reserved.

1. Introduction robust predictor of adolescent achievement (e.g., Froiland, Peterson, &

Preparation in mathematics is essential to both college and career
readiness (Adelman, 2003; Maruyama, 2012). The strongest predictors
of mathematics achievement in high school are prior achievement and
high school mathematics course rigor (Davenport et al., 2013;
Riegl-Crumb, 2006). Both achievement in high school and math course
taking rigor are predicted by motivation to learn math (Froiland &
Worrell, 2016a; Gottfried, Marcoulides, Gottfried, & Oliver, 2013). In
this study, we used extensive longitudinal data to investigate the effects
of peer interest, parent expectations, intrinsic motivation, and mathe-
matics course rigor on mathematics achievement after controlling for
effects of prior achievement, SES, gender, and ethnicity.
Students' intrinsic motivation promotes the development of class-
room engagement (Froiland & Oros, 2014; Froiland & Worrell, 2016a;
Jang, Kim, & Reeve, 2012), positive emotions toward homework and
school (Froiland, 2011), happiness, and achievement among students
(Froiland, Mayor, & Herlevi, 2015; Froiland, Oros, Smith, & Hirchert,
2012; Lepper, Corpus, & Iyengar, 2005; Reiss, 2009). Due to the fact
that intrinsic motivation is vital for children's wellbeing, it is important
to examine factors that influence student motivation, the influence that
motivation has on achievement, and the ways in which motivation has
indirect effects on achievement. Furthermore, parent expectations are a
⁎ Corresponding author at: Pearson Clinical Assessment San Antonio, 19500 Bulverde
Road, San Antonio, TX 78259-3701, United States.
E-mail address: john.froiland@pearson.com (J.M. Froiland).
1
John Mark Froiland is now at Pearson Clinical Assessment.
Davison, 2013). Expectancy-value theory posits that parent expecta-
tions largely exert an effect on achievement via elevated student expec-
tancy (Wigfield & Eccles, 2000). Parent expectations are also thought to
only indirectly affect value for learning (i.e., intrinsic value, such as in-
terest in learning; attainment value or the perceived importance of
doing well; and utility value, such as studying in order to obtain a better
job in the future) via student expectancies (Wigfield & Eccles, 2000).
But, the current study indicates that parent expectations have stronger
direct effects than student expectations on math intrinsic motivation,
math course-taking, and math achievement. This suggests that self-de-
termination theorists should no overlook long-term parent expecta-
tions and that expectancy-value theorists may need to consider that
parent expectations may have multiple pathways to elevating achieve-
ment related choices and achievement.
2. Intrinsic motivation for math and math achievement
According to self-determination theory, intrinsic motivation to learn
involves engaging with learning opportunities due to finding them en-
joyable or interesting (Author; Ryan & Deci, 2000). Students are most
likely to experience intrinsic motivation when their needs for autono-
my, competence, and relatedness are met by parents, peers, and other
adults (Froiland & Worrell, 2016b; Ryan & Deci, 2000). However, self-
determination theorists focused on parenting have articulated that
structure (e.g., providing clear expectations about homework) also sup-
port student's intrinsic motivation to learn (Farkas & Grolnick, 2010).
Gottfried, Fleming, and Gottfried (2001) add that intrinsic motivation

http://dx.doi.org/10.1016/j.lindif.2016.07.012
1041-6080/© 2016 Elsevier Inc. All rights reserved.
 J.M. Froiland, M.L. Davison / Learning and Individual Differences 50 (2016) 252–259
for math includes the enjoyment of challenging tasks, which is impor-
tant due to the of complex math problems and the fact that many stu-
dents describe math as difficult (Gottfried et al., 2001). Another
important aspect of autonomous motivation is integration, which in-
volves synthesizing a pursuit with one's sense of self (Ryan & Deci,
2000). Seeing oneself as a math person is an example of integration.
Both integration and intrinsic motivation have strong, positive relation-
ships to each other, and both are associated with many positive out-
comes (Ryan & Deci, 2000). Intrinsic motivation in self-determination
theory is nearly identical to intrinsic value in expectancy-value theory
(Wigfield & Eccles, 2000), although self-determination theory posits in-
trinsic motivation to be the highest form of motivation, whereas expec-
tancy-value theory does not emphasize a value hierarchy (e.g., both a
high intrinsic value and a high utility value can be seen as motivationally
advantageous to students). Many studies have found a positive associa-
tion between intrinsic motivation for math and math achievement (e.g.,
Areepattamannil, Freeman, & Klinger, 2011; Zhu & Leung, 2011). Chil-
dren who enjoy math employ strategies for deeper comprehension of
what they learn (Chatzistamatiou, Dermitzaki, Efklides, & Leondari,
2013).
3. Mathematics course-taking and math achievement
Researchers have found that the level of mathematics courses taken
in high school plays an important role in math achievement (Davenport
et al., 2013; Davison & Davenport, 2002; Ma & Wilkins, 2007). Gottfried
et al. (2013) found that intrinsic motivation for math and math achieve-
ment both predict that students will take higher level math courses in
high school and that this, in turn, contributes to educational attainment
in adulthood. Their study involved 114 students from California and
employed very few control variables. Harackiewicz, Rozek, Hulleman,
and Hyde (2012) developed an intervention, based upon expectancy-
value theory, that empowers parents to support students' science and
math-course taking by helping students see the relevance of math and
science in everyday life, as well as the way in which math and science
prepare students for college and STEM careers. This intervention led to
increases in parents seeing STEM courses as important for their high
school students' life, which in turn promoted student development of
greater utility value for science and math. Perhaps parent and student
expectations for success in college were also elevated through this inter-
vention. In general, intrinsic motivation plays a crucial role in promoting
high levels of diverse high school students' behavioral engagement with
learning (Froiland & Worrell, 2016a) and math-course taking might be
seen as another vital aspect of engagement.
4. Intrinsic motivation for math among peers
Radel, Sarrazin, Legrain, and Wild (2010) employed social contagion
theory and self-determination theory to assess how intrinsic motivation
spreads among peers. Among 48 high school physical education stu-
dents in France, half were randomly assigned to be tutored by peers
with higher intrinsic motivation and the other half were tutored by
less motivated peers. Perceptions of peer tutors' motivation predicted
the interest and enjoyment of participants in the study, indicating that
intrinsic motivation can spread among students, because intrinsically
motivated peers communicate about learning in an autonomy support-
ive way and exhibit more positive emotions while learning (Radel et al.,
2010). If peers, who may or may not be friends, can influence each
other's intrinsic motivation, there is an even greater potential for friends
to influence each other's intrinsic motivation, due to friends more
strongly meeting the need for relatedness, according to self-determina-
tion theory (Ryan & Deci, 2000).
In a cross-sectional study involving approximately two hundred
eighth grade students from a rural school in the U.S., researchers exam-
ined whether friends' intrinsic motivation for math would predict stu-
dents' own intrinsic motivation, while controlling for gender and
253
scores from a group administered mathematics test (Bissell-Havran &
Loken, 2009). Findings indicated that students' intrinsic motivation for
math was higher when they perceived that their friends' had higher in-
trinsic motivation. These results suggest that friends may affect each
other's intrinsic motivation; however, larger studies with individually
administered achievement tests are needed that replicate such findings
longitudinally with probability samples so that we can be confident that
such findings generalize to students across the population.
5. Parent expectations, math achievement, and intrinsic motivation
for math
Parent involvement is an important contributor to children's
achievement (e.g., Froiland, Powell, & Diamond, 2014; Froiland,
Powell, Diamond, & Son, 2013) and development of mathematics skills
in particular (e.g., Powell, Son, File, & Froiland, 2012). Numerous studies
have found that parents' expectations for their children's long-term ed-
ucational attainment predict better achievement development among
adolescents (e.g., Froiland & Davison, 2014; Froiland, Peterson, &
Davison, 2013). In fact, parent expectations have the largest effects out
of any parent involvement variables in numerous longitudinal studies
(Jeynes, 2012).
Parent expectations predict the autonomy of young adults, as well as
positive educational outcomes (Doren, Gau, & Lindstrom, 2012). In a na-
tionally representative study in the U.S., Fan and Williams (2010) found
that parent expectations for their high school students' long-term edu-
cational attainment predicted students' intrinsic motivation for math,
while controlling for SES and gender. Perhaps the direct relation be-
tween parent expectations and intrinsic motivation is due to the high
value on education conveyed by parents who expect that their students
will obtain a college or graduate degree. Namely, inherent in the expec-
tation that one's high school student will obtain high levels of postsec-
ondary education is a deep value placed upon education. Self-
determination theory considers parent expectations in general (e.g.,
providing clear expectations about homework) as a part of structure,
which promotes children's motivation (Farkas & Grolnick, 2010). How-
ever, measures of structure based on self-determination theory have
been focused more on expectations involving everyday life, rather
than years into the future.
6. Youth expectations, motivation, and achievement
According to expectancy-value theory, parental expectations are
transmitted to children and children's expectations directly affect
achievement (Eccles & Wigfield, 2002). In accordance, Froiland,
Peterson, and Davison (2013) found that parental expectations predict-
ed student expectations years later, which in turn predicted achieve-
ment, but parent expectations also had a direct effect on achievement
that was equivalent to the effect of student expectations. Likewise,
Bandura, Barbaranelli, Caprara, and Pastorelli (2001) found that parent
expectations directly predict achievement and student expectations. Al-
though expectancy-value theory most often focuses on specific expecta-
tions within a domain or task (e.g., “How well do you expect to do at
math this year?”, p. 70, Wigfield & Eccles, 2000), long-term expectations
for educational attainment hold within them numerous expectations
for adequate success in every class in high school (for those that believe
they will finish high school, college, or a graduate degree). Whereas
most expectancy-value theory researchers measure expectancy in
terms of self-concept (e.g., “I am good at math”; Harackiewicz et al.,
2012; Nagy, Trautwein, Baumert, Köller, & Garrett, 2006), long-term ex-
pectations for educational attainment can provide a robust indicator of
expectancy over long periods of time. This may be especially appropri-
ate for multi-year longitudinal studies.
Research on expectancy-value theory has found that expectancy
for success and value are moderately positively associated (e.g.,
Nagengast et al., 2011). Likewise, Suárez-Álvarez, Fernández-
254 J.M. Froiland, M.L. Davison / Learning and Individual Differences 50 (2016) 252–259
Alonso, and Muñiz (2014) found that youth expectations and moti-
vation work in concert to predict math achievement in Spain.
Vansteenkiste, Lens, Witte, and Feather (2005) have shown that ex-
pectancy-value theory and self-determination theory can be used si-
multaneously to more effectively predict job search efforts, than can
either theory alone. For instance, intrinsic motivation (from self-de-
termination theory), such as enjoying learning about the job market
during the search, adds explained variance above and beyond the ex-
pectation of finding a job and importance placed upon finding a job
in predicting job search efforts. They concluded by suggesting that
education researchers could also utilize both theories simultaneous-
ly to better explain academic outcomes. This integrative approach
has also been recommended by Brophy (2013).
7. Control variables related to motivation, expectations, and math
achievement
Girls, on average, express less intrinsic motivation than boys in
mathematics (Fan & Williams, 2010; Lee & Kim, 2014). Boys and girls
usually have similar levels of math achievement (within 0.15 standard
deviations of each other; Else-Quest, Hyde, & Linn, 2010; Hyde,
Lindberg, Linn, Ellis, & Williams, 2008; Skaalvik & Rankin, 1994). On
the other hand, girls in the U.S. benefit from higher parent expectations
that they will finish high school, attend college, and/or obtain an ad-
vanced degree (Froiland & Davison, 2014).
Asian American parents often hold higher long-term expectations
for their adolescents' educational attainment than other parents
(Froiland & Davison, 2014). In some studies, African American and Lati-
no American parents hold higher expectations than European American
families, whereas in other studies the reverse is true (Yamamoto &
Holloway, 2010). Asian American high school students usually have
the highest math achievement in the U.S. followed by European Amer-
ican, Latino and African American students (Mau & Lynn, 1999). Asian
American students reported more interest in math and were more likely
than European American students to take calculus and other high level
mathematics courses (Chen & Stevenson, 1995; Davenport et al., 1998,
2013).
Socio-economic status (SES) also has consistent relationships with
achievement and course-taking (Attewell & Domina, 2008; Davenport
et al., 1998). SES achievement gaps in mathematics often persist
throughout the school years (Shin, Davison, Long, Chan, & Heistad,
2013), making it important to examine malleable factors that contribute
to growth in mathematics. Furthermore, SES is positively associated
with parent expectations in various studies (e.g., Froiland & Davison,
2014; Froiland, Peterson, & Davison, 2013).
8. The current study
This longitudinal study utilizes a nationally representative prob-
ability sample with rigorous statistical controls to examine the ef-
fects of parent expectations, student expectations, peer interest in
school, intrinsic motivation for math, and math course-taking on
the development of math achievement from 9th to 11th grade. The
vast majority of studies that have examined math course-taking
and achievement have not examined intrinsic motivation within
their models. The beauty of including intrinsic motivation is that it
predicts persistence with challenging tasks (Froiland, 2015; Ryan &
Deci, 2000). This study adds to the literature on peer influences on
intrinsic motivation by testing whether peer interest in 9th grade is
indirectly related to math achievement in 11th grade via intrinsic
motivation. Previous studies of peer interest have been with non-
representative samples and have not included as many key control
variables. This study integrates key variables that have rarely been
considered simultaneously.
9. Hypotheses
The following hypotheses each involve SES, race/ethnicity (African
American, Asian American, and Latino), gender, and prior math achieve-
ment as control variables:
1) Intrinsic motivation for math in 9th grade will predict stronger math
achievement in 11th grade, both directly and indirectly via 9th grade
math achievement;
2) Intrinsic motivation for math will predict the highest level of math
course taken by 11th grade, which will, in turn, predict 11th grade
math achievement;
3) Parent expectations and peer interest in school will both be positive-
ly associated with intrinsic motivation for math;
4) Parent expectations will predict math achievement in 9th and 11th
grade, as well as highest level math course taken by 11th grade;
5) Peer interest will have an indirect positive effect on math achieve-
ment in 11th grade via intrinsic motivation for math.
In further accordance with expectancy-value theory, we
hypothesized:
6) Parent expectations will be positively associated with student ex-
pectations;
7) Student expectations will predict the highest level math course
taken by 11th grade;
8) Student expectations will be positively associated with intrinsic mo-
tivation for math;
9) Student expectations will predict math achievement in 11th grade
directly as well as via intrinsic motivation and highest level math
course taken.
10. Method
10.1. Data source and participants
The data examined in this study were extracted from the dataset col-
lected for the National Center for Education Statistics (NCES) High
School Longitudinal Study of 2009 (HSLS:09) that began in the fall of
2009 and followed high school students until the spring of 2012, ap-
proximately two and a half years (Ingels et al., 2014).2 This study in-
volved 18,623 children from across the U.S. and their parents who
remained in the study since ninth grade. A total of 49.7% of the youth
participating were female, 51.7% were European American, 22.3% were
Hispanic, 13.7% were African-American or Black, 3.4% were Asian Amer-
ican, 0.5% were native Hawaiian or Pacific Islanders, 0.6% were Native
American or Alaska Natives and 7.7% of participants were of more
than one race, non-Hispanic. This analysis included the nationally repre-
sentative longitudinal weight (W2W1stu), so results can be generalized
to high school students across the U.S.
10.2. Instruments
10.2.1. Intrinsic motivation
This latent variable was composed of four self-report questions stu-
dents answered in ninth grade. Students were asked the extent to which
2
The following germane studies also used data from the High School Longitudinal
Study of 2009: Middleton (2013) examined motivational variables that predicted achieve-
ment, but did not include parent expectations, student expectations, or highest level math
course. Andersen and Ward (2014) found that science intrinsic motivation predicted per-
sistence in STEM courses among high ability students, but they did not examine math
achievement. Using latent profile analysis, Andersen and Cross (2014) found that 41% of
high ability students had high math motivation and 15% had low motivation in math. They
did not examine expectations, parent expectations, or math-course taking. Alvarado and
An (2015) found that having close friends that are college-bound predicted greater college
readiness, but they did not examine how this is mediated by intrinsic motivation and
course-taking, nor did they examine parent expectations.
 J.M. Froiland, M.L. Davison / Learning and Individual Differences 50 (2016) 252–259 255

they see themselves as a math person (1 = Strongly Disagree, 2 = Dis-
agree, 3 = Agree, 4 = Strongly Agree), enjoy their current math course
(1 = Strongly Disagree…, 4 = Strongly Agree), were taking math
courses for the challenge (1 = Yes, 0 = No), and were taking math
courses for enjoyment (1 = Yes, 0 = No). Similar variables have been
used to construct a motivation variable and predict achievement in pre-
vious longitudinal studies (e.g., Froiland & Oros, 2014). See Table 1 for
descriptive statistics for the four variables that composed intrinsic mo-
tivation, as well as the descriptive statistics for the other variables in
this study. In an exploratory factoring, the first factor accounted for
over half the variance of the items, had the only eigenvalue N1.0, and
accounted for more than twice the variance of any other factor. In a con-
firmatory factoring, omega was 0.70, indicating adequate reliability.
parents. A similar single item measure of student expectations has pre-
dicted achievement in previous studies with youth (e.g., Froiland,
Peterson, & Davison, 2013).
10.2.6. Covariates: ethnicity, gender and family SES
Race/ethnicity was analyzed in separate variables for African Amer-
ican (Black or African American = 1; Other = 0); Asian American
(Asian American = 1; Other = 0); and Latino American (Latino Amer-
ican = 1; Other = 0). As in other studies controlling for race/ethnicity,
European American served as the reference group (e.g., Powell et al.,
2012). Gender was coded as follows: female = 1 and male = 0. Family
SES was measured as a z-score composite of family income, highest level
of parent education, and occupation prestige.

10.2.2. Peer interest in school 10.2.7. Mathematics achievement in 9th and 11th grade
This variable was measured with a self-report question that asked In both ninth and eleventh grade, youth were given online assess-
students in ninth grade whether or not their best friend was interested ments of algebraic reasoning administered by the NCES and RTI Interna-
in school (1 = Yes, 0 = No). Thirty-three percent of students across the tional. Because these tests were based on item response theory (IRT),
U.S. reported that their best friend was not interested in high school. students only attempted items within and just outside of their optimal
challenge range (see Ingels et al., 2014, p. 154 for a full description of
10.2.3. Highest level math course IRT procedures used by NCES). Algebraic reasoning domains included
For each student, the dataset contains a variable naming the stu- the following: “proportional relationships and change; linear equations,
dents highest math course taken by Spring 2012, which is 2.5 years inequalities, and functions; nonlinear equations; systems of equations;
after the study started. The highest math course was then scaled with sequences and recursive relationships; and performing algebraic rea-
respect to rigor as follows: 0 = business or other math; 1 = pre-alge- soning” (Ingels et al., 2014, p. 18).
bra; 2 = algebra or integrated math 1; 3 = geometry or integrated
math 2; 4 = algebra 2 or integrated math 3; 5 = algebra 3, analytic ge- 10.3. Data analysis plan
ometry, trigonometry, or pre-calculus; and 6 = calculus. Since it is un-
clear where statistics fits into this math sequence, it was coded as Structural equation modeling (SEM) in AMOS 23 enabled an exami-
missing data. Prior research has shown that the rigor of a students' nation of the effects of peer interest, parent expectations, intrinsic moti-
highest high school math course is related to end of high school mathe- vation, and highest level math course taken on 11th grade math
matics achievement (Davenport et al., 2013) and various college out- achievement, while controlling for gender, race/ethnicity, family SES
comes (Adelman, 2003). and 9th grade achievement (see Fig. 1). An alpha level of 0.01 was
used to test the significance of effects. In assessing model fit, the follow-
10.2.4. Parent expectations ing criteria were utilized: Comparative Fit Index (CFI) and Tucker-Lewis
In the fall of ninth grade, HSLS interviewers asked parents what level Index (TLI) of 0.95 or higher; root mean square error of approximation
of education they expected of their children, with the following options: (RMSEA) b 0.06 (Hu & Bentler, 1999). A large n makes it unlikely to have
1 = less than a high school diploma; 2 = graduate from high school or a non-significant chi-square (Kenny, 2011); therefore, the Hoelter Index
GED; 3 = start an Associate's degree; 4 = complete Associate's; 5 = provided an estimate of how small the sample size would need to be in
start a Bachelor's degree; 6 = finish a Bachelor's; 7 = start a Master's; order for the chi-square to become non-significant (Kenny, 2011). Be-
8 = complete a Master's; 9 = start PhD/MD/JD; 10 = complete PhD/ cause measures of a construct taken at the same time period are consid-
MD/JD. A similar single item measure of parent expectations has pre- ered to covary, errors of the observed motivation variables were
dicted achievement in numerous studies with youth (e.g., Englund, allowed to correlate (Byrne, 2001; Shin et al., 2013; Suárez-Álvarez et
Luckner, Whaley, & Egelund, 2004; Froiland, Peterson, & Davison, 2013). al., 2014).
Use of the normalized population weight (W2W1stu_norm) makes
10.2.5. Student expectations the data representative of high school students across the U.S. because
In the fall of ninth grade, students reported the level of education it accounts for the fact that participants were a probability sample of
they expect to attain, which was on the same 1 to 10 scale used with high school students in the U.S. The normalized weight was divided
by the design effect (4.4) appropriate for the longitudinal study, which
enhances the accuracy of standard errors and significance tests for
Table 1
parameters. Analyses were treated as if conducted with 4232 cases
Descriptive statistics for variables involving math development.
(i.e., 18,623 actual cases with positive weights that were treated by
Range Mean Std. deviation SPSS as 4232 after dividing by the design effect). To thoroughly address
Math person 1.00–4.00 2.49 0.95 missing data, we used SPSS 23 multiple (15) imputations with all other
Enjoy course 1.00–4.00 2.75 0.84 variables as predictors in the regression imputation model to create 15
Take challenge 0.00–1.00 0.15 0.37 correlation matrices, averaged the 15 correlation matrices, and input
Take enjoy 0.00–1.00 0.17 0.38
the average correlation matrix into the SEM analysis. In order to address
Friends' interest 0.00–1.00 0.67 0.47
Parent expect 1.00–10.00 6.61 2.45 indirect relations between peer interest and 11th grade math achieve-
Student expect 1.00–10.00 6.49 2.81 ment (hypothesis 5), as well as intrinsic motivation and 11th grade
SES −1.93–2.88 −0.07 0.74 math achievement (hypothesis 1), the bootstrapping test was used to
Highest level 0.00–6.00 3.85 1.26
examine the significance of the indirect effects (Shrout & Bolger, 2002).
Math 9 15.86–69.93 38.75 11.84
Math 11 25.01–115.10 64.44 18.95
11. Results
Note. Take challenge = take math for the challenge. Take enjoy = take math for enjoy-
ment. Parent expect = parent expectations. Student expect = student expectations.
Highest level = highest level math course taken. Math 9 = math IRT score in 9th grade. Table 2 displays the bivariate correlations between all continuous
Math 11 = math IRT score in 11th grade. variables in the study. The correlation between IRT Math scores in 9th
256 J.M. Froiland, M.L. Davison / Learning and Individual Differences 50 (2016) 252–259

Fig. 1. Predictors of math achievement development in high school. Path coefficients refer to standardized variables (mean = 0 and variance = 1.0). All coefficients depicted are significant
at p b 0.01. The following non-significant relationships are not depicted in the figure, but were included in the model: African American and Girl to Highest Math Course; Latino to Highest
Math Course; Friend Interested in School to Student Expectation; Student Expectation to Highest Math Course; Student Expectation to Math Score in 11th grade; and Latino to Math Score
in 11th grade. These non-significant paths were included in the model for testing purposes, but are not included in this figure, for the sake of visual clarity. Exhaustive covariances between
exogenous variables were included in the model, but are not depicted here for the sake of aesthetics and clarity. Disturbance terms for endogenous variables and error terms for observed
variables contributing to the latent variable, Motivation, were included in the model but are not depicted here for visual clarity. The following statistically significant relationships were
modeled but are not included in the figure: SES to Parent Expectation; Asian American to Parent Expectation; African American to Parent Expectation; Latino to Parent Expectation; Girl to
Parent Expectation; African American to Math Score in 9th grade; African American to Math Score in 11th grade; Asian American to Highest Math Course; Asian American to Motivation,
Asian American to Math Score in 11th grade, Latino to Math Score in 9th grade, and Asian American to Math Score in 9th grade. See Table 3 for all coefficients and significance levels. CFI =
0.99, TLI = 0.96, and RMSEA = 0.03. R-squared for the model predicting 11th grade math scores = 0.59.

and 11th grade was strong, providing further evidence that prior 0.96, and RMSEA = 0.03. The chi-square was (χ2(40) = 202.03,
achievement is a crucial control variable not usually present in prior p b 0.01) and the Hoelter Index indicated that 1334 cases or less
studies. would lead to a non-significant chi-square, whereas this study involved
4232 students. Overall, the model fit the data well.
11.1. Primary findings The first hypothesis, that intrinsic motivation for math would pre-
dict math achievement in 9th grade, as well as in 11th grade, was sup-
The structural equation model (see Fig. 1) provided a good fit with ported (see Fig. 1). The combined indirect effect of intrinsic
the data, according to the following fit statistics: CFI = 0.99, TLI = motivation for math on 11th grade math achievement was significant
(standardized indirect effect = 0.20, p b 0.01) as was the standardized
Table 2 direct effect of intrinsic motivation on 11th grade achievement (0.10,
Pearson correlations for continuous variables relevant to math development. p b 0.01). The standardized total effect of intrinsic motivation on 11th
1 2 3 4 5 6 7
grade achievement was 0.31, p b 0.01. In accordance with hypothesis
2, intrinsic motivation predicted highest level of math course taken by
1. Student expect –
11th grade, which also predicted math achievement in 11th grade.
2. Parent expect 0.43⁎ –
3. Math person 0.19⁎ 0.16⁎ – The results pertaining to hypotheses 1 and 2 indicate that intrinsic mo-
4. Enjoy course 0.13⁎ 0.11⁎ 0.47⁎ – tivation in 9th grade affects 11th grade achievement in three ways: di-
5. SES 0.29⁎ 0.25⁎ 0.09⁎ 0.01 – rectly, via 9th grade achievement, and via highest level course taken
6. Math 9 0.32⁎ 0.34⁎ 0.33⁎ 0.14⁎ 0.43⁎ – by 11th grade.
7. Math 11 0.31⁎ 0.34⁎ 0.32⁎ 0.14⁎ 0.42⁎ 0.74⁎ –
8. Highest level 0.26⁎ 0.27⁎ 0.21⁎ 0.10⁎ 0.27⁎ 0.43⁎ 0.45⁎
As expected based on the third hypothesis, parent expectations and
peer interest both predicted higher intrinsic motivation, with the stan-
Note. Parent expect = parent expectations. Student expect = student expectations.
dardized coefficients being comparable (0.16 and 0.15, respectively,
Highest level = highest level math course taken. Math 9 = math IRT score in 9th grade.
Math 11 = math IRT score in 11th grade. p b 0.01). The fourth hypothesis was supported in that parent expecta-
⁎ p b 0.01. tions had positive direct effects on 9th grade achievement, highest
 J.M. Froiland, M.L. Davison / Learning and Individual Differences 50 (2016) 252–259 257

course taken, and 11th grade achievement. The standardized indirect ef- Table 3
fect of parent expectations in 9th grade on 11th grade achievement was Regression weights: standardized and unstandardized with criterion variables in bold and
predictors underneath.
0.19 (p b 0.01), which is via intrinsic motivation, 9th grade achievement,
and highest math course taken. The standardized total effect of parent Standardized Unstandardized Standard error
expectations on 11th grade achievement was 0.25 (p b 0.01). Parent expect
As expected based on hypothesis 5, peer interest had positive indi- SES 0.303 1.003⁎⁎⁎ 0.052
rect effects on 9th grade achievement (standardized indirect effect = Girl 0.144 0.707⁎⁎⁎ 0.071
Asian 0.066 0.896⁎⁎⁎ 0.198
0.05, p b 0.01) and 11th grade achievement (standardized indirect ef-
Black 0.128 0.916⁎⁎⁎ 0.109
fect = 0.05, p b 0.01) primarily via intrinsic motivation. There was Latino 0.137 0.807⁎⁎⁎ 0.094
also a positive indirect effect of peer interest on highest course taken Student expect
by 11th grade via motivation (standardized indirect effect = 0.03, Parent expect 0.207 0.263⁎⁎⁎ 0.019
p b 0.01). These small but significant indirect effects suggest that peer Friend interest 0.026 0.170 0.098
SES 0.148 0.622⁎⁎⁎ 0.064
interest contributes to math development via intrinsic motivation. Girl 0.058 0.359⁎⁎⁎ 0.093
As predicted in hypothesis 6, parent expectations were moderately Motivation
positively associated with student expectations. Disconfirming hypoth- SES 0.075 0.038⁎⁎⁎ 0.009
esis 7, student expectations in 9th grade did not contribute to the pre- Girl 0.104 −0.077⁎⁎⁎ 0.014
Asian 0.061 0.125⁎⁎⁎ 0.035
diction of the highest level course taken. Confirming hypothesis 8,
Friend interest 0.148 0.117⁎⁎⁎ 0.015
student expectations were positively associated with intrinsic motiva- Parent expect 0.163 0.025⁎⁎⁎ 0.003
tion. Hypothesis 9 was only partially confirmed in that student expecta- Student expect 0.051 0.006⁎⁎ 0.002
tions did not directly predict 11th grade achievement; however, the Math 9
indirect effect of student expectations on 11th grade achievement, via SES 0.291 4.638⁎⁎⁎ 0.245
Motivation 0.315 10.038⁎⁎⁎ 2.175
intrinsic motivation and 9th grade achievement was positive and signif-
Parent expect 0.196 0.943⁎⁎⁎ 0.080
icant (standardized indirect effect = 0.04, p b 0.01). The overall R- Latino −0.037 −1.039⁎⁎ 0.391
squared for 11th grade math achievement was 0.59, indicating that Black −0.176 −6.050⁎⁎⁎ 0.453
59% of the variance in 11th grade math achievement was explained by Asian 0.057 3.694⁎⁎⁎ 0.869
Student expect 0.035 0.132⁎ 0.051
the variables studied. The standardized total effect of intrinsic motiva-
Highest level
tion on 11th grade achievement (0.31, p b 0.01) was fairly comparable Parent expect 0.108 0.055⁎⁎⁎ 0.008
to the total standardized effect of SES (0.39, p b 0.01) and to the stan- Motivation 0.119 0.406⁎⁎⁎ 0.115
dardized total effect of parent expectations (0.25, p b 0.01). See the Sup- SES 0.094 0.160⁎⁎⁎ 0.028
plementary file for information about differential item functioning and Girl 0.018 0.045 0.035
Black 0.019 0.068 0.055
the results of a second model with congruent findings. See Table 3 for
Asian 0.051 0.355⁎⁎⁎ 0.096
all direct effects in the main model, including those involving control Latino 0.002 0.005 0.046
variables. Math 9 0.297 0.032⁎⁎⁎ 0.002
Student expect 0.025 0.010 0.006
Take challenge
12. Discussion
Motivation 0.575 0.573⁎⁎⁎ 0.047
Enjoy math
The current study found that parent expectations for students' long- Motivation 0.443 1.00
term educational attainment, student expectations, friends' interest in Math person
high school, students' intrinsic motivation for math, and taking intellec- Motivation 0.813 2.077⁎⁎⁎ 0.134
Take enjoy
tually rigorous math courses all contributed positively to the develop-
Motivation 0.411 0.418⁎⁎⁎ 0.037
ment of math achievement between 9th and 11th grade. Using Math 11
nationally representative longitudinal data and rigorous control vari- Math 9 0.561 0.898⁎⁎⁎ 0.025
ables, this study confirmed and extended the findings of previous stud- Motivation 0.104 5.298⁎⁎⁎ 1.444
SES 0.099 2.535⁎⁎⁎ 0.306
ies that have found that intrinsic motivation for math predicts math
Parent expect 0.063 0.482⁎⁎⁎ 0.086
achievement (e.g., Areepattamannil et al., 2011). In accordance with Latino −0.012 −0.568 0.501
previous studies (e.g., Froiland, Peterson, & Davison, 2013), parent ex- Asian 0.028 2.918⁎⁎ 1.055
pectations had a direct positive effect on achievement above and be- Black −0.055 −3.002⁎⁎⁎ 0.602
yond the indirect effect via student expectations. The total Highest level 0.124 1.855⁎⁎⁎ 0.175
Student expect 0.002 0.013 0.063
standardized positive effect of parent expectations was 0.25, suggesting
that positive parent expectations combined with intrinsic motivation Note. Parent expect = parent expectations. Student expect = student expectations.
Highest level = highest level math course taken. Math 9 = math IRT score in 9th grade.
(0.31) contributed substantially to the development of math achieve-
Math 11 = math IRT score in 11th grade.
ment in high school. Student expectations were positively associated ⁎ p b 0.05.
with intrinsic motivation, which in turn predicted 11th grade achieve- ⁎⁎ p b 0.01.
ment. However, student expectations did not significantly affect math ⁎⁎⁎ p b 0.001.
course-taking, whereas parent expectations did. This suggests that ex-
pectancy-value theory may need to be updated to account for direct ef- levels of intrinsic motivation subsequently predict taking more chal-
fects of parent expectations on achievement related choices (Wigfield & lenging courses and achieving more in math. Whereas expectancy-
Eccles, 2000). Furthermore, the stronger positive effect of parent expec- value theory posits that parent expectations and value for academic
tations than student expectations on intrinsic motivation for math is tasks are indirectly positively related (Wigfield & Eccles, 2000), this is
also worthy of consideration for expectancy-value theorists and self-de- the second study with rigorous control variables to find a direct positive
termination theorists alike. As suggested by Vansteenkiste et al. (2005) relation. Self-determination theory posits that structure (including par-
and Brophy (2013), we found further evidence that self-determination ent expectations) predicts children's motivation (Farkas & Grolnick,
theory and expectancy-value theory can work in concert to predict im- 2010), but previous studies have focused on more immediate expecta-
portant educational outcomes (Froiland & Worrell, 2016b). tions surrounding homework. Further understanding how long-term
This study corroborated Fan and Williams (2010) finding that parent parent expectations promote intrinsic motivation is worthy of further
expectations predict intrinsic motivation, while adding that higher research.
258 J.M. Froiland, M.L. Davison / Learning and Individual Differences 50 (2016) 252–259
Students' perceptions of their friends' interest in school had a posi-
tive indirect effect on 11th grade achievement via intrinsic motivation.
This supports the findings of Radel et al. (2010), wherein they found
that physical education students could be influenced by their peers' en-
joyment in learning. Although their experimental study had high levels
of internal validity, the current study complements it with generaliz-
ability to high school students across a large nation, while adding that
peer interest not only influences short-term intrinsic motivation in
physical education, it predicts rigorous course-taking and the develop-
ment of math achievement over two years.
Intrinsic motivation is a key contributor to math development in
high school; thus, interventions that elevate intrinsic motivation, such
as teaching teachers (Stipek et al., 1998; Su & Reeve, 2011) and parents
to be more autonomy supportive (Froiland, 2011, 2015) should be fur-
ther developed and applied in the realm of high school math. Likewise,
interventions that seek to elevate expectations and value should be in-
vestigated. Hulleman, Godes, Hendricks, and Harackiewicz (2010)
have approached this by finding that college students who are trained
to focus on the relevance of math and psychology to their lives see the
subjects as more interesting, especially if they started with lower expec-
tations or lower grades. However, there is the potential to simulta-
neously elevate both intrinsic motivation and expectations, especially
because there is ample evidence that interventions based on both self-
determination theory and expectancy-value theory are effective
(Lazowski & Hulleman, 2015). Harackiewicz et al. (2012) found evi-
dence that parents can help elevate students' science and math-course
taking by helping students see the relevance of math and science in ev-
eryday life, as well as the way in which math and science prepare stu-
dents for college and STEM careers. Perhaps this intervention could be
studied again while examining the effects on intrinsic motivation, stu-
dent and parent expectations for college graduation, math-course tak-
ing and math achievement. Although the current study suggests that
parent expectations have a stronger positive effect than student expec-
tations on high school students' intrinsic motivation, rigorous course-
taking, and math achievement, both parent expectations and student
expectations could be elevated in harmony to promote better educa-
tional outcomes (Marcenaro-Gutierrez & Lopez-Agudo, 2016).
12.1. Future research and limitations
Because students' perceptions of their friends' intrinsic motivation
are more highly correlated with students' intrinsic motivation than
friends' self-reports of their intrinsic motivation (Bissell-Havran &
Loken, 2009), future studies could also seek to ask peers directly about
their motivation. Also, asking questions pertaining to friends' interest
in learning math specifically could enhance the friends' interest mea-
sure, by making the measure specifically focused on math. Despite the
rigorous control variables in this two and a half-year longitudinal
study, intervention studies with random assignment to the treatment
and control group are needed in order to see if parent expectations, stu-
dent expectations, peer interest, student intrinsic motivation, and
course-taking can be elevated to synergistically improve math achieve-
ment scores among high school students.
13. Conclusions
This study indicated that parent expectations and intrinsic motiva-
tion contributed significantly to the development of math achievement,
in part by predicting that students will take more challenging courses.
We echo previous calls for the field to study interventions designed to
enhance parent expectations (Froiland & Davison, 2014; Jeynes, 2012).
We also encourage researchers to develop interventions that include
strategically affecting the motivation of peers as part of a package to en-
hance intrinsic motivation. Without high levels of motivation and
strong expectations, students may not enroll in challenging courses
and, if they enroll, they may not have the motivation needed to meet
the challenge. Because most students lose intrinsic motivation for
math in high school (Lee & Kim, 2014), it is important to develop inter-
ventions that promote this important motivational resource that pre-
dicts rigorous course-taking and stronger achievement.
Appendix A. Supplementary information
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.lindif.2016.07.012.
References
Adelman, C. (2003). Postsecondary attainment, attendance, curriculum and performance
(NCES 2003-394). Washington, D. C.: National Center for Education Statistics, U. S. De-
partment of Education.
Alvarado, S. E., & An, B. P. (2015). Race, friends, and college readiness: Evidence from the
High School Longitudinal Study. Race and Social Problems, 7, 150–167.
Andersen, L., & Cross, T. L. (2014). Are students with high ability in math more motivated
in math and science than other students? Roeper Review, 36, 221–234.
Andersen, L., & Ward, T. J. (2014). Expectancy-value models for the STEM persistence
plans of ninth-grade, high-ability students: A comparison between Black, Hispanic,
and White students. Science Education, 98(2), 216–242.
Areepattamannil, S., Freeman, J. G., & Klinger, D. A. (2011). Intrinsic motivation, extrinsic
motivation, and academic achievement among Indian adolescents in Canada and
India. Social Psychology of Education, 14, 427–439.
Attewell, P., & Domina, T. (2008). Raising the bar: Curricular intensity and academic per-
formance. Educational Evaluation and Policy Analysis, 30, 51–71.
Bandura, A., Barbaranelli, C., Caprara, G. V., & Pastorelli, C. (2001). Self-efficacy beliefs as
shapers of children's aspirations and career trajectories. Child Development, 72,
187–206. http://dx.doi.org/10.1111/1467-8624.00273.
Bissell-Havran, J. M., & Loken, E. (2009). The role of friends in early adolescents' academic
self-competence and intrinsic value for math and English. Journal of Youth and
Adolescence, 38(1), 41–50. http://dx.doi.org/10.1007/s10964-007-9266-3.
Brophy, J. E. (2013). Motivating students to learn. Oxford, UK: Routledge.
Byrne, B. M. (2001). Structural equation modeling with AMOS. Mahwah, NJ: Lawrence
Erlbaum Associates.
Chatzistamatiou, M., Dermitzaki, I., Efklides, A., & Leondari, A. (2013). Motivational and af-
fective determinants of self-regulatory strategy use in elementary school mathemat-
ics. Educational Psychology, 1–16 ahead-of-print.
Chen, C., & Stevenson, H. W. (1995). Motivation and mathematics achievement: A com-
parative study of Asian-American, Caucasian-American, and East Asian high school
students. Child Development, 66, 1215–1234.
Davenport, E. C., Davison, M. L., Kuang, H., Ding, S., Kim, S. K., & Kwak, N. (1998). High
school mathematics course-taking by gender and ethnicity. American Educational
Research Journal, 35(3), 497–514.
Davenport, E. C., Davison, M. L., Wu, Y. -C., Kim, S. -K., Kuang, H., Kwak, N., ... Ayodele, A.
(2013). Number of courses, content of courses, and prior achievement related to eth-
nic achievement gaps in mathematics. Journal of Educational Leadership in Action, 2
http://www.lindenwood.edu/ela/issue03/davenport-lite.html.
Davison, M. L., & Davenport, E. C., Jr. (2002). Identifying criterion-related patterns of pre-
dictor scores using multiple regression. Psychological Methods, 7, 468–484.
Doren, B., Gau, J. M., & Lindstrom, L. E. (2012). The relationship between parent expecta-
tions and postschool outcomes of adolescents with disabilities. Exceptional Children,
79(1), 7–23.
Eccles, J. S., & Wigfield, A. (2002). Motivational beliefs, values, and goals. Annual Review of
Psychology, 53, 109–132. http://dx.doi.org/10.1146/annurev.psych.53.100901.
135153.
Else-Quest, N. M., Hyde, J., & Linn, M. C. (2010). Cross-national patterns of gender differ-
ences in mathematics: A meta-analysis. Psychological Bulletin, 136(1), 103–127.
http://dx.doi.org/10.1037/a0018053.
Englund, M. M., Luckner, A. E., Whaley, G. J. L., & Egelund, B. (2004). Children's achieve-
ment in early elementary school: Longitudinal effects of parental involvement, ex-
pectations, and quality of assistance. Journal of Educational Psychology, 96, 723–730.
http://dx.doi.org/10.1037/0022-0663.96.4.723.
Fan, W., & Williams, C. M. (2010). The effects of parental involvement on students' aca-
demic self-efficacy, engagement and intrinsic motivation. Educational Psychology,
30(1), 53–74.
Farkas, M. S., & Grolnick, W. S. (2010). Examining the components and concomitants of
parental structure in the academic domain. Motivation and Emotion, 34, 266–279.
Froiland, J. M. (2011). Parental autonomy support and student learning goals: A prelimi-
nary examination of an intrinsic motivation intervention. Child & Youth Care Forum,
40, 135–149. http://dx.doi.org/10.1007/s10566-010-9126-2.
Froiland, J. M. (2015). Parents' weekly descriptions of autonomy supportive communica-
tion: Promoting children's motivation to learn and positive emotions. Journal of Child
and Family Studies, 24, 117–126. http://dx.doi.org/10.1007/s10826-013-9819-x.
Froiland, J. M., & Davison, M. L. (2014). Parental expectations and school relationships as
contributors to adolescents' positive outcomes. Social Psychology of Education, 17,
1–17. http://dx.doi.org/10.1007/s11218-013-9237-3.
Froiland, J. M., & Oros, E. (2014). Intrinsic motivation, perceived competence and class-
room engagement as longitudinal predictors of adolescent reading achievement.
Educational Psychology, 34, 119–132. http://dx.doi.org/10.1080/01443410.2013.
822964.
 J.M. Froiland, M.L. Davison / Learning and Individual Differences 50 (2016) 252–259
Froiland, J. M., & Worrell, F. C. (2016a). Intrinsic motivation, learning goals, engagement,
and achievement in a diverse high school. Psychology in the Schools, 53, 321–336.
http://dx.doi.org/10.1002/pits.21901.
Froiland, J. M., & Worrell, F. C. (2016b). Parental autonomy support, community feeling,
and student expectations as contributors to later achievement among adolescents.
Educational Psychology. http://dx.doi.org/10.1080/01443410.2016.1214687 Advance
online publication.
Froiland, J. M., Mayor, P., & Herlevi, M. (2015). Motives emanating from personality asso-
ciated with achievement in a Finnish senior high school: Physical activity, curiosity,
and family motives. School Psychology International, 36, 207–221.
Froiland, J. M., Oros, E., Smith, L., & Hirchert, T. (2012). Intrinsic motivation to learn: The
nexus between psychological health and academic success. Contemporary School
Psychology, 16, 91–100. http://dx.doi.org/10.1007/BF03340978.
Froiland, J. M., Peterson, A., & Davison, M. L. (2013a). The long-term effects of early parent
involvement and parent expectation in the USA. School Psychology International, 34,
33–50. http://dx.doi.org/10.1177/0143034312454361.
Froiland, J. M., Powell, D. R., & Diamond, K. E. (2014). Relations among neighborhood so-
cial networks, home literacy environments, and children's expressive vocabulary in
suburban at-risk families. School Psychology International, 35, 429–444. http://dx.
doi.org/10.1177/0143034313500415.
Froiland, J. M., Powell, D. R., Diamond, K. E., & Son, S. -H. (2013b). Neighborhood socioeco-
nomic well-being, home literacy, and early literacy skills of at-risk preschoolers.
Psychology in the Schools, 50, 755–769. http://dx.doi.org/10.1002/pits.21711.
Gottfried, A. E., Fleming, J. S., & Gottfried, A. W. (2001). Continuity of academic intrinsic
motivation from childhood through late adolescence: A longitudinal study. Journal
of Educational Psychology, 93, 3–13.
Gottfried, A. E., Marcoulides, G. A., Gottfried, A. W., & Oliver, P. H. (2013). Longitudinal
pathways from math intrinsic motivation and achievement to math course accom-
plishments and educational attainment. Journal of Research on Educational
Effectiveness, 6, 68–92.
Harackiewicz, J. M., Rozek, C. S., Hulleman, C. S., & Hyde, J. S. (2012). Helping parents to
motivate adolescents in mathematics and science an experimental test of a utility-
value intervention. Psychological Science, 23, 899–906.
Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure anal-
ysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A
Multidisciplinary Journal, 6, 1–55.
Hulleman, C. S., Godes, O., Hendricks, B. L., & Harackiewicz, J. M. (2010). Enhancing inter-
est and performance with a utility value intervention. Journal of Educational
Psychology, 102(4), 880–895.
Hyde, J. S., Lindberg, S. M., Linn, M. C., Ellis, A. B., & Williams, C. C. (2008). Gender similar-
ities characterize math performance. Science, 321, 494–495.
Ingels, S. J., Pratt, D. J., Herget, D. R., Dever, J. A., Fritch, L. B., Ottem, R., ... Leinwand, S.
(2014). High School Longitudinal Study of 2009 (HSLS:09) base year to first follow-up
data file documentation. Washington, DC: National Center for Education Statistics, In-
stitute of Education Sciences, U.S. Department of Education (http://nces.ed.gov/
pubs2014/2014361.pdf).
Jang, H., Kim, E. J., & Reeve, J. (2012). Longitudinal test of self-determination theory's mo-
tivation mediation model in a naturally occurring classroom context. Journal of
Educational Psychology, 104, 1175.
Jeynes, W. (2012). A meta-analysis of the efficacy of different types of parental involve-
ment programs for urban students. Urban Education, 47(4), 706–742.
Kenny, D. (2011). Measuring model fit. Retrieved March 3, 2015, from http://www.
davidakenny.net/cm/fit.htm
Lazowski, R. A., & Hulleman, C. S. (2015). Motivation interventions in education a meta-
analytic review. Review of Educational Research. http://dx.doi.org/10.3102/
0034654315617832 Advance online publication.
Lee, H., & Kim, Y. (2014). Korean adolescents' longitudinal change of intrinsic motivation
in learning English and mathematics during secondary school years: Focusing on
gender difference and school characteristics. Learning and Individual Differences, 36,
131–139.
Lepper, M. R., Corpus, J. H., & Iyengar, S. S. (2005). Intrinsic and extrinsic motivational ori-
entations in the classroom: Age differences and academic correlates. Journal of
Educational Psychology, 97, 184.
259
Ma, X., & Wilkins, J. L. (2007). Mathematics coursework regulates growth in mathematics
achievement. Journal for Research in Mathematics Education, 38, 230–257.
Marcenaro-Gutierrez, O. D., & Lopez-Agudo, L. A. (2016). The influence of the gap be-
tween parental and their children's expectations on children's academic attainment.
Child Indicators Research. http://dx.doi.org/10.1007/s12187-015-9361-z Advance on-
line publication.
Maruyama, G. (2012). Assessing college readiness: Should we be satisfied with ACT or
other threshold scores? Educational Researcher, 41, 252–261.
Mau, W. C., & Lynn, R. (1999). Racial and ethnic differences in motivation for educational
achievement in the United States. Personality and Individual Differences, 27,
1091–1096. http://dx.doi.org/10.1016/S0191-8869(99)00051-3.
Middleton, J. A. (2013). More than motivation: The combined effects of critical motiva-
tional variables on middle school mathematics achievement. Middle Grades
Research Journal, 8, 77–95.
Nagengast, B., Marsh, H. W., Scalas, L. F., Xu, M. K., Hau, K. T., & Trautwein, U. (2011). Who
took the “×” out of expectancy-value theory? A psychological mystery, a substantive-
methodological synergy, and a cross-national generalization. Psychological Science, 22,
1058–1066.
Nagy, G., Trautwein, U., Baumert, J., Köller, O., & Garrett, J. (2006). Gender and course se-
lection in upper secondary education: Effects of academic self-concept and intrinsic
value. Educational Research and Evaluation, 12, 323–345.
Powell, D. R., Son, S. -H., File, N., & Froiland, J. M. (2012). Changes in parent involvement
across the transition from public school prekindergarten to first grade and children's
academic outcomes. The Elementary School Journal, 113(2), 276–300. http://dx.doi.
org/10.1086/667726.
Radel, R., Sarrazin, P., Legrain, P., & Wild, T. C. (2010). Social contagion of motivation be-
tween teacher and student: Analyzing underlying processes. Journal of Educational
Psychology, 102, 577–587. http://dx.doi.org/10.1037/a0019051.
Reiss, S. (2009). Six motivational reasons for low school achievement. Child & Youth Care
Forum, 38, 219–225. http://dx.doi.org/10.1007/s10566-009-9075-9.
Riegl-Crumb, C. (2006). The path through math: Course sequence and academic perfor-
mance. American Journal of Education, 113, 101–122.
Ryan, R. M., & Deci, E. L. (2000). Intrinsic and extrinsic motivations: Classic definitions and
new directions. Contemporary Educational Psychology, 25, 54–67.
Shin, T., Davison, M. L., Long, J. D., Chan, C. K., & Heistad, D. (2013). Exploring gains in
reading and mathematics achievement among regular and exceptional students
using growth curve modeling. Learning and Individual Differences, 23, 92–100.
Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimental stud-
ies: New procedures and recommendations. Psychological Methods, 7, 422–445.
Skaalvik, E. M., & Rankin, R. J. (1994). Gender differences in mathematics and verbal
achievement, self-perception and motivation. British Journal of Educational
Psychology, 64, 419–428.
Stipek, D., Salmon, J. M., Givvin, K. B., Kazemi, E., Saxe, G., & MacGyvers, V. L. (1998). The
value (and convergence) of practices suggested by motivation research and promot-
ed by mathematics education reformers. Journal for Research in Mathematics
Education, 465–488.
Su, Y. L., & Reeve, J. (2011). A meta-analysis of the effectiveness of intervention programs
designed to support autonomy. Educational Psychology Review, 23, 159–188. http://
dx.doi.org/10.1007/s10648-010-9142-7.
Suárez-Álvarez, J., Fernández-Alonso, R., & Muñiz, J. (2014). Self-concept, motivation, ex-
pectations, and socioeconomic level as predictors of academic performance in math-
ematics. Learning and Individual Differences, 30, 118–123.
Vansteenkiste, V., Lens, W., Witte, H., & Feather, N. T. (2005). Understanding unemployed
people's job search behaviour, unemployment experience and well-being: A compar-
ison of expectancy-value theory and self-determination theory. British Journal of
Social Psychology, 44, 269–287.
Wigfield, A., & Eccles, J. S. (2000). Expectancy–value theory of achievement motivation.
Contemporary Educational Psychology, 25, 68–81.
Yamamoto, Y., & Holloway, S. D. (2010). Parental expectations and children's academic
performance in sociocultural context. Educational Psychology Review, 22, 189–214.
Zhu, Y., & Leung, F. K. (2011). Motivation and achievement: Is there an East Asian model?
International Journal of Science and Mathematics Education, 9(5), 1189–1212.