Learning and Individual Differences 67 (2018) 41–52

Contents lists available at ScienceDirect

Learning and Individual Differences

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

Understanding the academic motivations of students with a history of T

reading difficulty: An expectancy-value-cost approach
⁎
Bradley W. Bergeya, , Rauno K. Parrilab, S. Hélène Deaconc
a
Queens College, City University of New York, Secondary Education and Youth Services Department, 65-30 Kissena Blvd., Flushing, NY 11367, United States
b
Macquarie University, Department of Educational Studies, NSW 2109, Australia
c
Dalhousie University, Psychology and Neuroscience Department, Life Science Centre, P.O. Box 15000, Halifax, Nova Scotia B3H 4R2, Canada

A R T I C LE I N FO A B S T R A C T

Keywords: We examined the academic self-efficacy, positive subjective task values, and perceived effort cost of first-year
Motivation undergraduates with (n = 168) and without (n = 314) a self-reported history of reading difficulty, and further
Academic achievement their relations with academic achievement and satisfaction. Students with a self-reported history of reading
Expectancy-value theory difficulty described lower academic self-efficacy, earned lower grades, and accrued fewer credits. The groups did
Cost
not differ significantly in their positive task values, effort cost, academic satisfaction, or institutional retention.
Reading difficulty
Path analyses indicated that for both groups, academic self-efficacy and effort cost were predictive of first-year
academic performance while intrinsic value was predictive of academic satisfaction and institutional retention.
Multi-group analyses indicated a significant group difference: academic self-efficacy explained unique variance
in academic satisfaction for students without a self-reported history of reading difficulty, but not for those who
reported such a history. We discuss implications of the relations between difficulties in reading acquisition and
motivations.

1. Introduction
The academic transition from high school to university is challen-
ging for many types of students, but particularly so for students with
learning difficulties (Eckes & Ochoa, 2005; Madaus, 2005). One group
that is academically vulnerable during this transition is university stu-
dents with a history of reading difficulty. While these students achieve
academic standing sufficient for admission to university, they tend to
have below-average reading abilities (Deacon, Cook, & Parrila, 2012).
As a result, they experience greater academic difficulty at university
and may face above-average risk of dropping out (Bergey, Deacon, &
Parrila, 2017; Chevalier, Parrila, Ritchie, & Deacon, 2017). Among
general university populations, motivational factors have been found to
have strong influences on academic outcomes (Schneider & Preckel,
2017), though little is known about the academic motivations of stu-
dents with a history of reading difficulty. In the current study, we ex-
amine these motivations through the lens of expectancy-value theory
(Eccles et al., 1983), which posits that motivation for academic tasks is
driven by expectations for success (e.g., Will I be successful?) and the
perceived value for the task (e.g., Do I care?). We investigate whether
students who self-report a history of reading difficulties differ from
students with no such history in academic self-efficacy and subjective
task values for obtaining a university degree. We also examine how
these motivational perceptions are associated with first-year academic
performance, academic satisfaction, and institutional retention for
students with and without a self-reported history of reading difficulty.
Understanding how motivational perceptions are linked to academic
outcomes for students with a history of reading difficulties illuminates
how universities can support the academic success of this at-risk po-
pulation.
1.1. Expectancy-value theory
Expectancy-value theory offers a broad theoretical framework for
understanding the development of motivation and its influence on
choices, persistence, and achievement in academic settings. Eccles and
her colleagues (Eccles et al., 1983; Eccles & Wigfield, 1995; Wigfield &
Eccles, 2000) developed the most prominent recent articulation of ex-
pectancy-value theory for academic contexts. According to Eccles et al.,
motivation is a function of one's expectation for success and values for a
task (Wigfield & Eccles, 2000). When individuals perceive themselves
to be capable of a task and expect to be successful, they are more likely
to choose to engage in it, persist at it longer, and achieve better out-
comes. Subjective task values refer to the extent to which an individual

⁎
Corresponding author.
E-mail addresses: bradley.bergey@qc.cuny.edu (B.W. Bergey), rauno.parrila@mq.edu.au (R.K. Parrila), helene.deacon@dal.ca (S.H. Deacon).

https://doi.org/10.1016/j.lindif.2018.06.008
Received 2 May 2017; Received in revised form 21 June 2018; Accepted 23 June 2018
1041-6080/ © 2018 Elsevier Inc. All rights reserved.
B.W. Bergey et al.
cares about or values the task at hand. Within the Eccles et al.'s model,
subjective task values are comprised of four interrelated values: in-
trinsic value, utility value, attainment value, and cost. Intrinsic value
refers to the enjoyment experienced while engaging in the task. Utility
value refers to how useful the task is for reaching immediate and future
goals. Attainment value refers to the extent to which the task is deemed
important to individual identity. Cost refers to what is suffered or given
up as a result of engaging in the task (Eccles, 2009; Wigfield & Eccles,
2000; Wigfield, Tonks, & Eccles, 2004). Intrinsic, utility and attainment
values are theorized to motivate task initiation and persistence while
costs exert the opposite influence.
Self-efficacy is the belief that one can organize and execute actions
to achieve designated results (Bandura, 1997). Applied to educational
contexts, academic self-efficacy refers to the belief that one can ac-
complish academic tasks at designated levels (Pajares, 1996; Schunk,
1991). Individuals who believe they are capable of success in a task are
more likely to initiate and persist at a challenging task. Academic self-
efficacy is a type of expectancy belief and is closely related to the
construct of expectancy for success, which Eccles et al. defined as the
belief in how well one will perform on future tasks (Wigfield & Eccles,
2000). The construct of academic self-efficacy is grounded in social-
cognitive theory (Bandura, 1977, 1986) rather than expectancy-value
theory; nevertheless, academic self-efficacy and expectancy for success
share substantial conceptual overlap and may be empirically indis-
tinguishable, especially when the task is operationalized broadly (Bong,
2001; Pajares, 1996; Wigfield & Eccles, 2000), as it is in the current
study.
1.2. Associations between academic self-efficacy, subjective task values,
academic performance and persistence
A compelling body of research has demonstrated that expectancy
beliefs, such as academic self-efficacy, and subjective task values are
correlates of academic performance and persistence in post-secondary
education settings. Students who report high levels of self-efficacy are
more willing to initiate and persist at challenging tasks, are more likely
to achieve higher academic performance, and express greater intentions
to persist in a program of study than are students who report low self-
efficacy (Bong, 2001; Devonport & Lane, 2006; Klassen & Usher, 2010;
Pajares, 1996). A recent review of meta-analytic studies examining
correlates of achievement in higher education found that academic self-
efficacy demonstrated consistently large effects on achievement
(Schneider & Preckel, 2017). In an earlier meta-analysis, Multon,
Brown, and Lent (1991) found that self-efficacy explained 14% of
variance in academic performance and 12% of variance in persistence;
ability level and age were significant moderators, with larger effects
found for high school and college students (vs. younger students) and
larger effects for students with low academic performance (vs. students
with normative academic performance).
The subjective task values held by university students have been
found to be positively associated with academic performance and the
intention to persist in programs of study and the intention to persist at
university (Bong, 2001; Neuville et al., 2007; Perez, Cromley, & Kaplan,
2014). When both expectancy beliefs and task values are examined
simultaneously, expectancy beliefs tend to be the strongest predictors of
performance while subjective task values tend to be the strongest pre-
dictors of task choice (Wigfield & Cambria, 2010). Although the con-
struct of effort cost has received less empirical attention, research
suggests that costs can be important negative predictors of academic
choices, such as the intention to choose or persist in a major (Battle &
Wigfield, 2003; Perez et al., 2014).
1.3. Associations between academic self-efficacy, subjective task values and
academic satisfaction
Academic satisfaction refers to the perceived enjoyment and
42
Learning and Individual Differences 67 (2018) 41–52
fulfillment in the role or experiences of being a student. There has been
a growing interest in academic satisfaction as an important outcome in
educational experiences (e.g., Garriott, Hudyma, Keene, & Santiago,
2015; Lent, 2004; Ojeda, Flores, & Navarro, 2011; Sheu, Mejia, Rigali-
Oiler, Primé, & Chong, 2016). This trend follows the long-standing at-
tention that job satisfaction has received in motivational research in
employment settings. Lent and colleagues (Lent et al., 2005; Lent,
Singley, Sheu, Schmidt, & Schmidt, 2007) have identified self-efficacy
and interest as positive predictors of academic satisfaction. Satisfaction
has also been found to be shaped by acting in alignment with personal
values and goals (Ryan & Deci, 2001) and therefore tasks that are
perceived to be personally relevant, aligned with life goals, and im-
portant to one's sense of self are likely to support satisfaction percep-
tions.
1.4. Associations between motivational beliefs and retention via academic
performance and satisfaction
In a seminal paper on student departure from higher education and
in subsequent work, Tinto (1975, 2007) developed a sociological model
of the decision to drop out of college. The model posits that the decision
to leave university prior to graduation is ultimately influenced by the
extent to which an individual is integrated into the academic and social
systems within an institution. The aspect of Tinto's theory that is most
relevant to questions taken up by the current study relate to academic
integration. According to Tinto's model, being and feeling academically
integrated is the product of a student's academic performance, such as
grades or credit accrual, and his or her intellectual development. The
model suggests that students whose academic performance and in-
tellectual development aligns with normative standards in the uni-
versity will feel more academically integrated and satisfied, and be less
likely to depart prematurely. Empirical studies have supported predic-
tions from Tinto's model. Low grades and failing courses are indicators
of the decision to leave university (Ishitani & DesJardins, 2002). Stu-
dents who feel satisfied with or integrated into academic life have been
found to be more likely to be engaged during learning (Wefald &
Downey, 2009) and more likely to be retained (Schertzer & Schertzer,
2004; Starr, Betz, & Menne, 1972; Tinto, 1993, 2007). A similar pattern
has been found for students with learning difficulties, with academic
integration explaining unique variance in the intention to persist at
university among first- and second-year university students with
learning disabilities (DeDeppo, 2009).
While Tinto's (1975) model does not explicitly address the role of
motivational beliefs, Bean and Eaton (2000, 2001) expanded on Tinto's
model to focus on the psychological processes that lead to institutional
integration. In their psychological model of retention, Bean and Eaton
posited that motivational beliefs such as self-efficacy were proximal
psychological predictors of academic performance and integration.
Rodgers and Summers (2008) have further expanded Bean and Eaton's
model to include subjective values. Taken together, these models sug-
gest that academic self-efficacy and subjective task values are asso-
ciated with academic satisfaction and performance (e.g., accruing
credits), and in turn, the decision to re-enroll in university, though to
our knowledge these indirect relations have not been empirically
tested.
1.5. Reading skills and academic achievement of students with a self-
reported history of reading difficulty
Children who struggle in the acquisition of reading skills typically
continue to experience reading difficulties into adulthood (Lefly &
Pennington, 1991). Some of these individuals learn to compensate for
their reading difficulties to the point of earning admission to university
(Parrila, Georgiou, & Corkett, 2007). Since many individuals who have
early reading or learning difficulty do not receive a diagnosis of a
learning disability (e.g., Brown, 2013; Kaplan & Shachter, 1991;
B.W. Bergey et al.
Orenstein, 2000), researchers have developed self-report measures of
reading history to identify university students with reading difficulties
(Lefly & Pennington, 2000; Parrila, Corkett, Kirby, & Hein, 2003). We
refer to individuals who report moderate to severe reading difficulties
on such measures as students who self-report a history of reading dif-
ficulty. There is now ample evidence that these self-reports of early
reading difficulties identify students who have poorer word reading and
timed reading comprehension skills than their peers who report no such
history (Deacon et al., 2012; McGonnell, Parrila, & Deacon, 2007;
Parrila et al., 2007). For example, Deacon et al. (2012) found that
university students who self-reported a history of reading difficulties
had performance on standardized measures of word reading and
reading comprehension equivalent to four grade levels below the per-
formance of students who reported no reading acquisition difficulties.
Critically, students who self-reported reading difficulty did not differ in
their scores on standardized measures of word reading and reading
comprehension compared to students with diagnosed learning dis-
abilities. Similar patterns were found in Parrila et al. (2007) and
McGonnell et al. (2007), providing further validity evidence towards a
claim that self-reports of early reading difficulty accurately reflect
challenges in this experience.
Research has suggested that the population of students who self-
report a history of reading difficulty in post-secondary institutions is
sizable. Deacon, Tucker, and Bergey (2015), summarizing research
from multiple cohorts across multiple universities in Canada, reported
that approximately 20% of in-coming first-year students reported at
least moderate reading acquisition difficulty. The authors noted that
approximately one-third of those who self-reported a history of reading
difficulty also reported a history of a diagnosis of a reading-related
learning disability. Therefore, the population of university students
who self-report a history of reading difficulty is likely to include both
students with and without a formal diagnosis of reading-related
learning disabilities.
Students who self-report early reading difficulty have been found to
have below-average academic performance. For students with low
reading skills, the reading-intensive academic tasks that are common at
university are likely to be particularly onerous. Students who self-re-
ported a history of reading difficulty have been found to earn, on
average, lower GPAs in each of their first four years at university
(Bergey et al., 2017; also see Chevalier et al., 2017). They also tend to
earn fewer of the credits they attempt. Bergey et al. (2017) found that
students who self-reported a history of reading difficulty earned a
median of three fewer credits each year compared to students with no
history of reading difficulty, despite attempting the same amount.
Academic difficulty may translate into reduced institutional retention.
In order to support this academically vulnerable group, it is important
to identify potential strengths and weakness as well as pinpoint corre-
lates of academic performance and retention.
1.6. Motivation of Students with a History of Reading Difficulty
Learning difficulties influence the psychological development of
children and adolescents and can shape academic motivation in a
variety of ways (Cohen, 1986; Sideridis, 2009; Wigfield & Eccles, 2000).
Some individuals can become highly motivated to achieve in the face of
their unique learning difficulties, especially when they frame their
difficulty as a personal challenge and receive support from others
(Cohen, 1986). An individual with reading difficulties, for example,
might use his personal motivation to achieve as a way of compensating
and overcoming cognitive weaknesses (Corkett, Hein, & Parrila, 2008;
Łockiewicz, Bogdanowicz, & Bogdanowicz, 2014). Yet for other stu-
dents, persistent learning difficulties can lead to self-beliefs and per-
ceptions that lower academic motivation, reflecting less enjoyment
with academic tasks and doubt about one's ability to succeed. For ex-
ample, Baird, Scott, Dearing, and Hamill (2009) found that secondary
students with diagnosed learning disabilities tended to report lower
43
Learning and Individual Differences 67 (2018) 41–52
academic self-efficacy compared to non-disabled peers. Similarly,
Klassen, Krawchuk, and Rajani (2008) found that university students
with diagnosed learning disabilities reported greater procrastination
and less confidence in their ability to self-regulate their studying be-
haviors.
In a review of motivation literature on students with diagnosed
learning disabilities, Sideridis (2009) emphasized the potential of using
an expectancy-value framework to guide research on achievement
motivations. Eccles et al. (1983) model (Wigfield & Eccles, 2000) posits
that expectancies and values for academic tasks are influenced by an
array of social-cognitive factors, including perceptions of individual
aptitudes, prior achievement experiences, and socializing influences.
From this perspective, a student with low reading abilities who reg-
ularly experiences academic tasks as arduous and frustrating may come
to perceive himself as less academically capable, may perceive aca-
demic endeavors as less enjoyable, or may question whether the effort
necessary to achieve in academic settings is worthwhile.
Two studies on the academic motivation of university students with
a history of reading difficulty point to unanswered questions about the
effects of such a history on motivational variables and their relation
with academic achievement. Corkett et al. (2008) interviewed 10 uni-
versity students or recent graduates who reported a history of reading
difficulty about what contributed to their academic success. High mo-
tivation and a belief in one's abilities were central themes in students'
reflections. These qualitative findings suggested that, at least among
academically successful students with reading difficulties, motivation in
general, and confidence in one's abilities in particular, were important
factors in supporting their achievement. It remains an open question
whether confidence beliefs are linked with academic achievement in a
larger sample of students with a history of reading difficulty that en-
compasses a broader range of levels of academic achievement.
Bergey et al. (2017) compared the self-reported academic motiva-
tion and academic achievement of first-year university students with
and without a self-reported history of reading difficulty. Students who
reported a history of reading difficulty described significantly lower
levels of academic motivation compared to students with no such his-
tory. Intriguingly, motivation was not significantly correlated with first-
year GPA among students with a history of reading difficulties, but was
for students with no history of reading difficulties. This pattern suggests
that motivational variables may relate to academic outcomes differ-
ently for students with versus without a history of reading difficulty.
Taken together, the results of these two studies raise questions about
how having a history of reading difficulty may impact specific moti-
vational perceptions, such as confidence in one's abilities and percep-
tions about the values of academic tasks, and further, whether and how
motivational perceptions support academic achievement and well-
being for students with a history of reading difficulty.
1.7. Current study
In the current study, we sought to identify motivational correlates of
academic performance, academic satisfaction, and institutional reten-
tion for students with a self-reported history of reading difficulty. To
assess whether and how these motivational perceptions operated dis-
tinctly for students with a self-reported history of reading difficulty, we
compared them with students with no such history. We pursued the
following two research aims:
1.7.1. Research aim 1
Our first aim was to examine whether students who self-reported a
history of reading difficulty differed from those who reported no such
history in their motivational self-perceptions, academic satisfaction,
and academic achievement. In terms of motivational self-perceptions,
we examined whether the groups differed in mean levels of academic
self-efficacy, positive subjective task values, and effort cost for
achieving a university degree. Within the expectancy-value framework,
B.W. Bergey et al.
we operationalized expectancy as academic self-efficacy. The concept of
psychological costs has received increased attention recently (Barron &
Hulleman, 2015) and several types of costs have been proposed (e.g.,
Flake, Barron, Hulleman, McCoach, & Welsh, 2015). In the current
study, we focused on effort expenditure (which we refer to as effort
cost) because low reading abilities of students who self-report a history
of reading difficulty may require greater level of effort expenditures for
success.
We first hypothesized (Hypothesis H1a) that students who self-re-
ported a history of reading difficulties would report lower academic
self-efficacy beliefs than students with no such history. This expectation
was based on extrapolating findings from studies on students with di-
agnosed learning disabilities (Baird et al., 2009; Klassen et al., 2008).
Second, we hypothesized (Hypothesis H1b) that students with a
history of reading difficulty would report lower intrinsic value, greater
effort cost, and lower academic satisfaction. This expectation was based
on the presumption that repeated experiences with academic difficul-
ties may lead students with a history of reading difficulty to view
academic tasks as less enjoyable or requiring more extensive effort
compared to how academic tasks are viewed by peers with no such
history. Further, prior research with students with a self-reported his-
tory of reading difficulty observed lower motivation, less positive atti-
tude, and higher anxiety compared to students with no history of
reading difficulty (Bergey et al., 2017). However, our explorations in
this area are novel as we are not aware of prior literature that has ex-
amined subjective task values and academic satisfaction among uni-
versity populations with learning difficulties or disabilities.
Third, we hypothesized (Hypothesis H1c) that students with a self-
reported history of reading difficulty would earn lower first-year GPA,
accrue fewer first-year credits, and be retained at the institution at a
lower rate than students with no such history. This expectation was
based on prior research with this population that has found below-
average first-year academic performance (Bergey et al., 2017; Chevalier
et al., 2017). We extrapolated from these findings that students with a
history of reading difficulty may be less likely to be retained in their
second year relative to students with no history of reading difficulty.
1.7.2. Research aim 2
Our second aim was to evaluate the extent to which academic self-
efficacy and subjective task values uniquely and collectively predicted
first-year academic performance, academic satisfaction and institu-
tional retention for each group. To examine these relations, we tested a
model shown in Fig. 1, and we examined whether relations differed for
the two groups. We examined direct effects of academic self-efficacy
and subjective task values on academic performance and satisfaction.
We also examined direct effects of academic self-efficacy and subjective
task values on institutional retention as well as their indirect relations
mediated by academic performance and satisfaction.
Fig. 1. Conceptual model of expectancy-value variables and academic outcomes.
Note. No additional controls were included in the model.
44
Learning and Individual Differences 67 (2018) 41–52
We hypothesized (Hypothesis H2a) that academic self-efficacy and
subjective task values would be positively associated with academic
performance, academic satisfaction, and institutional retention. We
further hypothesized (Hypothesis H2b) that subjective task values would
explain unique variance in institutional retention, beyond academic
performance and satisfaction. Hypotheses H2a and H2b were based on
prior research with general university populations that has suggested
that expectancy beliefs, such as self-efficacy, and task values predict
academic performance (Bong, 2001; Garriott et al., 2015; Sheu et al.,
2016). Among students with learning difficulties, expectancy beliefs
have been identified as important factors supporting academic
achievement (Margolis & McCabe, 2004); subjective task values, on the
other hand, have received little empirical attention.
We also hypothesized (Hypothesis H2c) that academic self-efficacy
would have an indirect effect on institutional retention via its effect on
academic performance. This expectation was based on the fact that self-
efficacy has been found to be a robust predictor of academic perfor-
mance, as noted above, and academic performance has been shown to
predict institutional retention (Allen, Robbins, Casillas, & Oh, 2008).
We did not make specific hypotheses regarding group differences.
Prior research has found the relations between general academic mo-
tivation and academic performance were less robust for university
students with a history of reading difficulty than for those with no such
history (Bergey et al., 2017). Yet, among academically successfully
individuals, self-efficacy and task values, such as interest, have been
identified as particularly potent motivators of performance and per-
sistence (Corkett et al., 2008; Fink, 1995).
2. Method
2.1. Participants
Participants were 482 university students at a research-intensive
Canadian university: 168 students self-reported a history of reading
difficulty (M age = 18 years 5 months; SD = 14 months; 117 females)
and 314 students self-reported no history of reading difficulty (M
age = 18 years 5 months; SD = 13 months; 248 females). Students had
a wide range of majors: 56% of students with a self-reported history of
reading difficulty and 57% of those with no such history were enrolled
in faculties of Science, Technology, Mathematics or Engineering; re-
spectively, 17% and 21% were enrolled in the Arts and Social Science
faculty, and 27% and 22% were enrolled in professional programs, such
as business management, journalism, and nursing. As a proxy for socio-
economic status, we report parental educational attainment. A bache-
lor's degree was the highest median degree earned by fathers (28% and
29%) and mothers (31% and 38%) of students with and without a self-
reported history of reading difficulty, respectively. Race and ethnicity
information were not available for the sample. Inclusion criteria were
B.W. Bergey et al.
(1) being in the first year of university, (2) reporting English as one's
first written and spoken language, and (3) consenting to tracking aca-
demic progress. Among students who reported a history of reading
difficulty, 18% also reported a current or prior diagnosis of a learning
disability; among students who reported no history of reading diffi-
culties, 5% reported a current or prior diagnosis of a learning disability.
2.2. Procedure
All first-year students at the university were invited by the registrar
to complete an online questionnaire about their reading history along
with other measures not reported on in the current study. Consenting
students provided a non-anonymous university identification number,
which was used to link survey responses to registrar data. Initial
questionnaires were completed by 2027 students (response rate: ap-
proximately 41%) within three weeks of the start of students' first se-
mester at university. In the middle of the second semester of students'
first year, students who completed the initial questionnaire, gave per-
mission to track academic progress, verified they were in their first year
of university, and reported English as their first language (n = 1229)
were invited to complete a follow-up questionnaire. The follow-up
questionnaire assessed academic self-efficacy beliefs, subjective task
values, and academic satisfaction, along with other measures of first-
year perceptions and experiences not analyzed here. The follow-up
questionnaire was completed by 546 students (response rate: 44%) with
all questionnaires completed prior to the final exam period of the
second semester. Following the academic year, GPA, credits accrued,
and institutional retention status were retrieved from the registrar. The
research was conducted with the approval of the University's
Institutional Review Board and in accordance with American
Psychological Association's guidelines for ethical research with human
subjects.
2.3. Measures
2.3.1. Reading history
Reading history was assessed with the Elementary scale of the Adult
Reading History Questionnaire-Revised (ARHQ-R; Parrila et al., 2003).
The ARHQ-R scale consisted of eight items (α = 0.90) assessing the
extent to which individuals reported having experienced reading ac-
quisition difficulty as children. An example item is How much difficulty
did you have learning to read in elementary school? Responses were re-
corded on a 5-point Likert scale with descriptors for each point on the
scale. Summed scores were transformed to a scale ranging from 0 (no
early reading difficulty) to 1 (widespread difficulty). Following cutoff
points described in prior research (Bergey et al., 2017; Deacon et al.,
2012), students with scores over 0.37 on the ARQH-R were identified as
having a history of moderate to severe difficulty. Those with scores
below 0.26 were identified as reporting no history of reading difficulty.
Following prior research (Chevalier et al., 2017; Parrila et al., 2007),
students whose scores fell between 0.26 and 0.37 (n = 64) were ex-
cluded from analyses to ensure clear separation between groups; sample
characteristics presented in Section 2.1 do not include these excluded
students.
2.3.2. Academic self-efficacy
Academic self-efficacy was measured with a five-item scale
(α = 0.85; Klassen et al., 2008). An example item is I'm confident I can
understand the most complex material presented by my instructors. Re-
sponses were recorded using a 5-point Likert scale with anchored de-
scriptors at 1 (Not at all true of me) and 5 (Very true of me).
2.3.3. Subjective task values
Positive subjective task values were measured with three scales
adapted from Conley (2012). Items were adapted to focus on the task of
achieving a university degree rather than math achievement. All scales
45
Learning and Individual Differences 67 (2018) 41–52
used a 5-point Likert response scale, anchored with descriptors at 1 (not
at all true of me) and 5 (very true of me). Intrinsic value was assessed
with five items (α = 0.92). A sample item is What I am learning in
university is exciting to me. Attainment value was assessed with five items
(α = 0.91). A sample item is Doing well in university is an important part
of who I am. Utility value was assessed with five items (α = 0.77). A
sample item is A university degree will be useful for me later in life. Effort
cost items were modified from Battle and Wigfield (2003), Conley
(2012), and Perez et al. (2014). Effort cost was assessed with four items
(α = 0.78). A sample item is I worry that I will waste a lot of time and
effort by going to university.
2.3.4. Academic satisfaction
Academic satisfaction was assessed with the Academic subscale of
the Extended Satisfaction with Life Scale (Alfonso, Allison, Rader, &
Gorman, 1996). The academic satisfaction scale consisted of five items
(α = 0.82), which assessed satisfaction with one's current education,
including satisfaction with instructors, classmates, curriculum, and
schooling in general. A sample item is The education I get at school is
great. Responses were recorded on a 7-point Likert scale with de-
scriptors at each point ranging from 1 (strongly disagree) to 7 (strongly
agree).
2.3.5. Academic performance and institutional retention
First-year academic performance was assessed with two variables:
cumulative first-year GPA and total credits accrued during the first
year. GPA was a continuous variable representing the cumulative
average grade using a scale from 0 to 4.3. Total credits accrued indicate
the total number of credits that were successfully completed (i.e.,
achieving a passing grade) during the first year. A typical full load for
two semesters was 30 credits. Institutional retention was oper-
ationalized as having attempted at least one course in the first semester
of the second year. All achievement data was retrieved from the
University Registrar.
2.4. Data analysis
2.4.1. Factor structure of academic self-efficacy and subjective task value
measures
We first established the factor structure of the expectancy-value
constructs. There is little prior research examining the factor structure
of academic self-efficacy and subjective task values when effort cost is
measured. Therefore, we conducted exploratory factor analysis in SPSS
(Vers. 24) using principle axis factoring with an Oblimin rotation on
academic self-efficacy and subjective task value items. Examination of
the scree plot and factor structure matrix suggested a four-factor solu-
tion with academic self-efficacy beliefs, intrinsic value, and costs items
loading on separate factors and with utility and attainment value items
loading together. To evaluate the adequacy of this four-factor model,
we conducted a Confirmatory Factor Analysis (CFA) in Mplus (Vers. 7,
Muthén & Muthén, 2012) using Maximum Likelihood estimator, spe-
cifying a model in which items from academic self-efficacy, intrinsic
value, attainment/utility value, and effort costs scales loaded on four
factors. Modification indices were used to improve the model fit by
correlating error terms among items within the same scale. In addition,
two utility value items were removed due to low factor loading (< 0.5).
The resulting model showed acceptable model fit (χ2 [182] = 387.367,
p < .001, CFI = 0.972, RMSEA = 0.048, SRMR = 0.055). Model fit
was evaluated against the following criteria suggested by Kline (2010):
chi-square p value ≥ .05; CFI ≥ 0.95; RMSEA ≤ 0.10; SRMR ≤ 0.08.
An alpha of p < .05 was adopted for all analyses.
2.4.2. Group differences in measurement and structural models
We evaluated group differences following the procedures outlined
in Kline (2010). First, we assessed configural invariance by fitting the
same four-factor model (described in Section 2.4.1) to both groups,
B.W. Bergey et al.
with no equality constraints (an unconstrained model). If the un-
constrained model had acceptable fit, we assessed construct-level me-
tric invariance by testing a model in which all parameters were con-
strained to be equal for the two groups (a fully constrained model). We
then tested the difference in chi-square between the constrained and
unconstrained models to determine whether the less parsimonious un-
constrained model resulted in significantly improved fit. Due to the
sample size of students who self-reported a history of reading difficul-
ties, we lacked the statistical power to include the measurement model
in the structural regression. Therefore, we saved factor scores from the
final CFA model and used these in subsequent analyses. Given that GPA
and credits accrued were relatively strongly correlated with each other
(see Table 3), we opted to include only the latter in the model, as shown
in Figs. 2 and 3. Analyses resulting from models with GPA instead of
credits accrued revealed similar results to the ones reported below.1
We assessed structural invariance by fitting the model illustrated in
Figs. 2 and 3 to both groups, with no equality constraints. If this model
had acceptable fit, we compared it to a nested model in which all model
parameters were constrained to be equal for the two groups (the fully
constrained model). If the constrained model resulted in significantly
poorer model fit, a series of models were tested in which individual
loadings were freed while all other were constrained to be equal to
determine which did not significantly improve model fit. In the final
model, we estimated indirect effects of expectancy-value constructs on
institutional retention via academic performance and satisfaction.
2.4.3. Additional analyses
We assessed mean group differences with a MANOVA in which self-
reported reading history status was the independent variable and aca-
demic self-efficacy, intrinsic value, attainment-utility value, effort cost,
academic satisfaction, first-year GPA, and credits accrued were de-
pendent variables. For significant effects, Cohen's ds were calculated to
evaluate effect size, with small, medium, and large effect sizes indicated
by value of 0.02, 0.05, and 0.08, respectively (Cohen, 1988). We used a
2 (history of reading difficulty, no difficulty) × 2 (retained, not re-
tained) chi-square test to assess whether groups differed in their in-
stitutional retention rate. Bivariate Pearson correlations were calcu-
lated for all variables to estimate zero-order correlations and identify
variables that might be multicollinear or redundant.
3. Results
3.1. Group differences in the factor structure of academic self-efficacy and
subjective task value scales
As shown in Table 1, both the fully constrained and fully un-
constrained models demonstrated adequate fit. The differences in fit
indices between these models were minimal: the change in chi-square
was non-significant and changes in CFI and RMSEA were slight (Chen,
2007). This suggested no significant differences in the factor structure
for the two groups. As a result, a model in which loadings were the
same for both group was more parsimonious and therefore preferable.
1
We examined whether gender and parental education merited being in-
cluded as statistical controls in the model. For both HRD and NRD samples,
bivariate correlations among gender and parental education with credits
earned, academic satisfaction, and retention were not statistically significant
(p > .10). We also tested the model shown in Figs. 2 and 3 controlling for
gender and parental education. Results indicated that gender and parental
education had non-significant effects on all endogenous variables in the model,
and all model estimates were similar to those of the reported model. Therefore,
in the interest of parsimony, our final model does not include gender and
parental education as control variables.
46
Learning and Individual Differences 67 (2018) 41–52
3.2. Mean group differences
Descriptive statistics for academic self-efficacy, subjective task va-
lues, academic satisfaction, GPA, credits accrued, and institutional re-
tention rates are shown in Table 2. Descriptively, students who self-
reported a history of reading difficulty reported lower academic self-
efficacy and subjective task values, higher costs, and lower academic
satisfaction; earned lower first-year GPA, accrued fewer first-year
credits, and had a lower institutional retention rate.
A significant MANOVA (F[8, 473] = 3.445, p = .001, partial
μ2 = 0.055) indicated that students who self-reported a history of
reading difficulty reported significantly lower academic self-efficacy (F
[1, 480] = 16.482, p < .001, d = 0.39), lower first-year GPA (F[1,
480] = 4.840, p = .028, d = 0.21), and fewer accrued credits (F[1,
480] = 4.512, p = .034, d = 0.20). Groups did not significantly differ
in attainment-utility value (F[1, 480] = 1.724, p = .190), effort cost (F
[1, 480] = 0.930, p = .335), intrinsic value (F[1, 480] = 2.912,
p = .089), or academic satisfaction (F[1, 480] = 0.627, p = .429). That
is, students with a self-reported history of reading difficulty had sig-
nificantly lower first-year performance and reported feeling less cap-
able of handling academic challenges than students with no such his-
tory. Yet, the two groups did not significantly differ in the perceived
value of attaining a university degree or their satisfaction with their
academic experience. A chi-square test indicated that the two groups
did not differ in their institutional retention rate (χ2 [1] = 0.596,
p = .440).
3.3. Bivariate correlations
Turning to our second aim, we examined the relations among aca-
demic self-efficacy beliefs, subjective task values, academic satisfaction,
academic performance, and institutional retention. Bivariate correla-
tion for students with and without a history of reading difficulties is
shown in Table 3. Both groups had the same pattern of relations in
terms of valence. All correlations were in the expected directions. For
both groups, GPA and credits accrued had significant positive correla-
tions with academic self-efficacy and intrinsic value, and significant
negative correlations with effort cost. For both groups, institutional
retention had significant positive correlations with intrinsic and at-
tainment/utility values and significant negative correlations with effort
cost. For students with a history of reading difficulties only, institu-
tional retention was significantly correlated with academic self-effi-
cacy.
3.4. Model modification and fitting
Invariance testing of the constrained and unconstrained structural
models is presented in Table 1. The unconstrained structural model
(shown in Figs. 2 and 3) demonstrated good fit, suggesting the ade-
quacy of the structural model for both groups. The fully constrained
model also demonstrated adequate fit. However, the change in chi-
square was significant, suggesting that one or more model parameters
were significantly different for the two groups. Invariance testing for
individual paths revealed that one structural coefficient differed sig-
nificantly for the two groups: academic self-efficacy had a significant
positive relation with academic satisfaction for students with no self-
reported history of reading difficulty (β = 0.121) while this relation-
ship was negative and non-significant for students with a self-reported
history of reading difficulty (β = −0.127). Academic self-efficacy did
not explain unique variance in academic satisfaction among students
who self-reported a history of reading difficulty, as it did among stu-
dents with no such history. Fit statistics and coefficients for the final
model are presented in Table 4 and illustrated in Figs. 2 and 3.
Indirect effects of academic self-efficacy and subjective task values
on institutional retention via credits accrued were then calculated. For
students with a self-reported history of reading difficulty, academic self-
B.W. Bergey et al. Learning and Individual Differences 67 (2018) 41–52

Fig. 2. Model showing significant standardized path coefficient for students with a self-reported history of reading difficulty.
Note. Solid lines denote significant paths; dashed lines denoted tested paths with non-significant coefficients.

Fig. 3. Model showing significant standardized path coefficient for students without a self-reported history of reading difficulty
Note. Solid lines denote significant paths; dashed lines denoted tested paths with non-significant coefficients.

Table 1 Table 2
Evaluation of group differences in measurement model. Descriptive statistics for academic self-efficacy, task values, satisfaction, and
academic achievement by reading history status.
χ2 df p CFI RMSEA SRMR
History of reading No history of reading
Measurement model difficulty difficulty
Fully constrained 662.828 421 < 0.001 0.967 0.049 0.078
Fully unconstrained 614.306 383 < 0.001 0.968 0.050 0.059 M SD M SD
Difference between 48.522 38 0.119 −0.001 −0.001 0.019
models Academic self-efficacy −0.13 0.58 0.07 0.50
Structural model Intrinsic value −0.08 0.68 0.04 0.74
Fully constrained 28.759 16 0.026 0.976 0.058 –a Utility-attainment value −0.05 0.56 0.02 0.53
Fully unconstrained 2.950 2 0.229 0.998 0.044 –a Effort cost 0.04 0.65 −0.02 0.72
Difference between 25.809 14 0.027 −0.022 0.014 –a Academic satisfaction 5.32 0.91 5.39 1.00
models GPA 3.05 0.76 3.21 0.80
Credits accrued 27.31 4.60 28.28 4.85
Mplus does not calculate SRMR for models involving categorical outcomes. Institutional retention 0.88 0.33 0.90 0.31

efficacy had a significant positive indirect effect on institutional re- Academic self-efficacy and subjective task values are standardized factor
loading scores from the measurement model. M = Mean; SD = Standard
tention (β = 0.060) and effort cost had a significant negative indirect
Deviation.
effect on institutional retention (β = −0.048). Similar indirect effects
were found for students with no self-reported history of reading diffi-
culty, as shown in Table 4.

47
B.W. Bergey et al. Learning and Individual Differences 67 (2018) 41–52

Table 3
Correlations among modelled variables for students with (above diagonal) and without (below diagonal) a self-reported history of reading difficulty.
1 2 3 4 5 6 7 8

1. Academic self-efficacy 0.744 0.252 −0.598 0.376 0.317 0.487 0.185

2. Intrinsic value 0.676 0.390 −0.726 0.567 0.303 0.386 0.342
3. Attainment-utility value 0.343 0.376 −0.420 0.279 0.128 0.108 0.164
4. Effort cost −0.622 −0.710 −0.567 −0.494 −0.344 −0.399 −0.294
5. Academic satisfaction 0.521 0.643 0.230 −0.444 0.242 0.260 0.155
6. Year 1 GPA 0.329 0.300 0.267 −0.324 0.226 0.603 0.356
7. Year 1 credits accrued 0.503 0.374 0.257 −0.389 0.306 0.611 0.213
8. Retained to year 2 0.069 0.213 0.148 −0.165 0.152 0.188 0.162

Correlations of 0.155 or greater were significant (p < .05) for students with a self-reported history of reading difficulty (above diagonal). Correlations of 0.148 or
greater were significant (p < .05) for students without a self-reported history of reading difficulty (below diagonal).

Table 4
Direct and indirect model estimates for students with and without a self-reported history of reading difficulties.
History of readings difficulty No history of reading difficulty

Unst SE Std Unst SE Std

Institutional retention on
Credits accrued 0.076 0.012 0.313⁎⁎⁎ 0.076 0.012 0.318⁎⁎⁎
Academic satisfaction 0.053 0.097 0.041 0.053 0.097 0.046
Academic self-efficacy −0.425 0.246 −0.220 −0.425 0.246 −0.184
Intrinsic value 0.530 0.163 0.320⁎⁎⁎ 0.530 0.163 0.337⁎⁎⁎
Utility-attainment 0.179 0.154 0.090 0.179 0.154 0.082
Effort cost −0.215 0.151 −0.124 −0.215 0.151 −0.134

Credits accrued on
Academic self-efficacy 1.530 0.548 0.191⁎⁎ 1.530 0.548 0.158⁎⁎
Intrinsic value 0.439 0.396 0.064 0.439 0.396 0.067
Utility-attainment 0.490 0.380 0.059 0.490 0.380 0.054
Effort cost −1.100 0.313 −0.153⁎⁎⁎ −1.100 0.313 −0.163⁎⁎⁎

Academic satisfaction on
Academic self-efficacy −0.114 0.099 −0.076 0.252 0.098 0.125⁎⁎
Intrinsic value 0.766 0.066 0.596⁎⁎⁎ 0.766 0.066 0.562⁎⁎⁎
Utility-attainment −0.010 0.074 −0.006 −0.010 0.074 −0.005
Effort cost −0.020 0.069 −0.015 −0.020 0.069 −0.015

Indirect effects on institutional retention

Via self-efficacy → credits accrued 0.116 0.046 0.060⁎ 0.116 0.046 0.050⁎
Via effort cost → credits accrued −0.083 0.028 −0.048⁎⁎ −0.083 0.028 −0.052⁎⁎

Unst = Unstandardized coefficient; SE = Standard error; Std = Standardized coefficient.

⁎
p < .05.
⁎⁎
p ≤ .01.
⁎⁎⁎
p ≤ .001.

4. Discussion
In our study, we applied an expectancy-value framework to un-
derstand the motivations and academic outcomes of university students
who reported a history of reading difficulty. Our results indicated that
academic self-efficacy beliefs and subjective task values were sig-
nificant correlates of academic satisfaction, academic performance, and
second-year institutional retention. We found several similarities and
differences between students with and without a self-reported history of
reading difficulty in motivational variables and their relations with
academic outcomes, which we discuss below. Together, these findings
underscore the importance of motivational variables in understanding
the academic satisfaction, achievement and persistence of students who
report a history of reading difficulty.
4.1. Group differences in motivational self-perceptions and academic
satisfaction, performance, and institutional retention
Our first aim was to examine whether students with and without a
self-reported history of reading difficulty differed in their motivational
self-perceptions, academic satisfaction, academic performance, and
institutional retention. A novel and important finding of the study is
that when compared to typically reading peers, students with a history
of reading difficulty reported lower academic self-efficacy, but similar
subjective task values and academic satisfaction, despite lower aca-
demic performance. These results confirm hypotheses H1a and H1c, but
not H1b. This pattern of motivational beliefs clarifies prior research on
this population. Bergey et al. (2017) found that students with a self-
reported history of reading difficulty reported below average levels of
motivation, less positive attitudes, and greater anxiety. Our findings
suggest that low motivation may be the result of lower academic self-
efficacy. Since students with a history of reading difficulties have been
found to have lower reading skills and have greater academic difficul-
ties (Chevalier et al., 2017; Deacon et al., 2012), it is understandable
that students who report a history of reading difficulty also report less
confidence in their academic abilities. Yet regardless of one's current
abilities, self-efficacy is an important motivational resource for per-
sisting in the face of challenge (Bandura, 1997; Bouffard-Bouchard,
Parent, & Larivee, 1991).
At the same time, comparable reported levels of subjective task
values and academic satisfaction suggest that a history of reading dif-
ficulties did not diminish the perceived enjoyment, utility and im-
portance of attaining a university degree or academic satisfaction with
the process. Also, students with a self-reported history of reading

48
B.W. Bergey et al.
difficulty did not perceive increased effort cost. These results are
somewhat surprising, since students with a history of reading difficulty
have likely experienced greater academic difficulty and report lower
confidence in their abilities. Nevertheless, these students have found
ways to experience typical levels of enjoyment and satisfaction in their
academic experience and perceive the endeavor of pursuing a uni-
versity degree as worth the effort. The results point to the fact that
many students with a self-reported history of reading difficulty held
motivational beliefs that are likely to support them through challenging
academic tasks (Wigfield & Cambria, 2010).
Our results regarding academic performance confirmed and ex-
tended prior research with this population. We found that students with
(vs. without) a self-reported history of reading difficulty had sig-
nificantly lower first-year GPAs and accrued fewer credits, consistent
with our hypothesis (H1c). While the pattern and magnitude of aca-
demic performance differences are consistent with prior research
(Bergey et al., 2017; Chevalier et al., 2017), results from our sample did
not bear out predictions that students who reported a history of reading
difficulty would be more likely to leave the institution. We note our
study was limited to second-year retention; assessing potential differ-
ences in institutional retention in subsequent years of study is an im-
portant direction for future research. These results are consistent with
prior research on university students with diagnosed learning dis-
abilities that has found similar institutional retention rates compared to
those of students with no disabilities (Vogel & Adelman, 1992; see also
Wessel, Jones, Markle, & Westfall, 2009).
4.2. Relations among academic self-efficacy, subjective task values and
academic outcomes
Our second aim was to evaluate the extent to which academic self-
efficacy and subjective task values uniquely and collectively predicted
first-year academic performance, academic satisfaction, and institu-
tional retention for each group. An important and novel finding is that
academic self-efficacy, effort cost, and interest value explained unique
variance in first-year academic performance or institutional retention,
consistent with our hypotheses. Multi-group analyses indicated that the
medium-to-large effect of academic self-efficacy and the small effect of
effort cost on academic performance were comparable in magnitude for
the two groups; the same is true for the medium effect of intrinsic value
on institutional retention.
Our findings extend prior research on the effects of academic self-
efficacy on academic performance of university students with learning
disabilities (Hampton & Mason, 2003; Hen & Goroshit, 2014) and of
general university populations (Wigfield & Cambria, 2010) to a new
population. Our study provides support for Corkett et al.'s (2008)
qualitative finding that academically successful students with a history
of reading difficulty credit their success to a belief in themselves. Our
results suggest that having a history of reading difficulty may diminish
university students' confidence in their academic abilities; such beliefs
may, in turn, exacerbate the effects of low reading ability on academic
performance and persistence, and indirectly on institutional retention.
Low self-efficacy beliefs may result in students avoiding initiating dif-
ficult tasks, exerting less effort during challenging tasks, using less ef-
fective study strategies, and giving up more easily in the face of diffi-
culties (Bandura, 1986; Pajares, 1996).
With regard to effort cost, our results extend the limited empirical
research on the relations among effort cost and academic outcomes.
Our factor analyses suggested effort cost was empirically distinct from
positive task values, consistent with prior research (Chiang, Byrd, &
Molin, 2011; Conley, 2012; Kosovich, Hulleman, Barron, & Getty, 2015;
Perez et al., 2014). While prior research has linked effort cost to the
intention to persist (e.g., Perez et al., 2014), our study extends this
research by demonstrating that—for both students with and without a
self-reported history of reading difficulty—effort cost was negatively
correlated with actual institutional retention and explained unique
49
Learning and Individual Differences 67 (2018) 41–52
variance in academic performance after controlling for academic self-
efficacy and positive subjective task values. Our results underscore the
importance of considering perceived drawbacks and costs alongside
positive task values when conceptualizing motivation through an ex-
pectancy-value framework (Barron & Hulleman, 2015).
An important finding of the study was that intrinsic value was the
strongest unique predictor of institutional retention for students with
and without a self-reported history of reading difficulty, explaining
more variance in institutional retention than even first-year academic
performance. Students who reported enjoying their first-year academic
experience were more likely to return the following year. Our results
extend prior research with general populations that show that values
positively predict college students' course enrollment intentions or ac-
tions (Bong, 2001; Perez et al., 2014; Wigfield & Cambria, 2010). We
extend these results by demonstrating the importance of intrinsic value
as it relates to actual institutional retention at university for an aca-
demically vulnerable group.
Finally, this study is novel in examining correlates of academic sa-
tisfaction from an expectancy-value framework. We observed an in-
teresting group difference in how academic satisfaction related to
subjective task values and academic self-efficacy. For both groups,
academic satisfaction was strongly and positively related to intrinsic
value. This is perhaps not surprising giving the conceptual similarity
between intrinsic enjoyment and satisfaction. Yet only for students who
reported no history of reading difficulty was academic satisfaction also
uniquely predicted by academic self-efficacy. That is, for students with
no self-reported history of reading difficulty, academic satisfaction was
dependent, in part, on confidence in their ability to achieve academic
success. By contrast, for students who reported a history of reading
difficulty, academic satisfaction was largely independent of such con-
fidence. One explanation is that students with a self-reported history of
reading difficulty may become accustomed to experiencing doubt about
their academic abilities, yet they have found ways to find satisfaction in
academic experiences even when they are less academically confident.
Future research might examine how students with a history of reading
difficulty report comparable levels of academic satisfaction and sub-
jective task values despite experiencing greater academic difficulties
and often facing academic endeavors with lower confidence in their
academic abilities.
Our results raise questions for future research using Tinto's (1975)
model of dropout from higher education. As Tinto's model would sug-
gest, academic satisfaction, which we used as a proxy for academic
integration, had significant and positive bivariate correlations with
retention for both students with and without a self-reported history of
reading difficulty. Yet model results indicated that, after controlling for
credits accrued, academic satisfaction did not explain a statistically
significant amount of unique variance in retention. This finding raises
the question of whether academic performance has a direct relationship
with student departure decisions, rather than a mediated relationship
through academic integration, as Tinto's model suggests. An alternative
interpretation is that perceptions of academic integration such as aca-
demic satisfaction are less predictive of student departure than are
actual, objective measures of academic performance, such as credit
accrual and grades. Future research should explore these issues in the
context of a model that more closely reflects the complexity of Tinto's
model.
4.3. Limitations
Our results should be considered alongside the following study
limitations. Our study design did not allow us to make causal claims
about the effect of academic self-efficacy and subjective task values on
academic performance or institutional retention. Academic self-efficacy
and subjective task values were assessed at a single time point and in
relatively close proximity to first-year academic performance outcomes.
While expectancy-value constructs may lead to higher achievement, the
B.W. Bergey et al.
reverse pattern of relations, that achievement affects expectancy-va-
lues, may also be true. In addition, we assessed academic self-efficacy
and subjective task values at a domain-general level (e.g., values for
achieving a university degree), rather than for domain-specific tasks,
such as achieving in science or humanities. While our approach allowed
us to examine relations across academic contexts, this may have masked
different relations that occur within specific academic domains (e.g.,
Selkirk, Bouchey, & Eccles, 2011; Watt, 2004). For example, ex-
pectancies and values may operate differently for students reporting a
history of reading difficulty within domains in which more or less
emphasis is placed on reading. Similarly, future research might examine
relations controlling for known predictors, such as prior achievement,
which were not included in the current analyses due to data avail-
ability. Further, the study did not include some information about
students with reading difficulties that may have affected their motiva-
tional beliefs, academic achievement, and their relations; these include
students' actual disability status, comorbidity of other diagnoses or
academic difficulties (e.g., ADHD), or the academic support or accom-
modations some may have received.
In addition, results on institutional retention should be considered
in light of the research design and operational definitions. Since we did
not follow-up with students who did not re-enroll in their second year,
we do not know their reason for leaving or whether they enrolled in
another institution; therefore, we caution against interpreting lack of
retention as a measure of dropout. That said, leaving the institution
certainly reflects a change in course of academic study. We also note
that the retention rates of the samples were likely lower than actual
institutional retention rates, since students who had already left the
institution prior to data collection in the second semester could not
participate in the study. In addition, institutional retention rates and
relations to motivational beliefs and academic performance and sa-
tisfaction may differ across contexts (by country, institution type, etc.),
highlighting the importance of replicating the patterns observed here.
Finally, we reported on indirect effects of academic self-efficacy and
subjective task values on institutional retention. While our model fol-
lows the temporal ordering of achievement variables (i.e., first-year
academic performance preceded the decision to re-enroll in year two),
conclusions about the indirect effects should be viewed as tentative, in
light of concerns that have been raised about using cross-sectional data
to evaluate indirect relations (Jose, 2013).
4.4. Implications
Our findings have practical implications. Prior research has in-
dicated that self-reports of reading difficulty identify university stu-
dents with actual reading deficits (Deacon et al., 2012; McGonnell
et al., 2007; Parrila et al., 2007) and below-average academic
achievement (Bergey et al., 2017; Chevalier et al., 2017). Given the
academic vulnerability of this group, institutions may wish to intervene
to support the academic success of students with a self-reported history
of reading difficulty (e.g., Deacon, Tucker, Bergey, Laroche, & Parrila,
2017). Results of the current study suggest that motivational percep-
tions might be directly addressed when attempting to support uni-
versity students with a self-reported history of reading difficulty. Given
that such students tend to report lower academic self-efficacy, they may
benefit from interventions that support confidence in their academic
abilities. For example, interventions that assist these students in de-
veloping effective study strategies, scaffold initial academic experi-
ences, and provide encouragement may bolster self-efficacy beliefs.
While support for academic self-efficacy is important, approaches to
motivate students with a self-reported history of reading difficulty that
address a range of motivational perceptions, beliefs, and goals are likely
to be the most successful (Hidi & Harackiewicz, 2000). The strong re-
lations among intrinsic value, academic satisfaction, and retention for
both students with and without a self-reported history of reading dif-
ficulty underscore the importance of designing learning activities and
50
Learning and Individual Differences 67 (2018) 41–52
environments that support student interest (e.g., Bergin, 1999). Stu-
dents with a self-reported history of reading difficulty might also ben-
efit from messages that make the case for why the effort it takes to
succeed and persist at university is worthwhile. In addition, it may be
important to help students who report a history of reading difficulty
identify personal goals and values that may make persistence at uni-
versity seem worth the effort. Recent work on self-reflective writing to
increase perceived relevance provides one promising way this might be
accomplished (Hartwell & Kaplan, 2018).
Students with a self-reported history of reading difficulty are not the
only students who are likely to benefit from motivational interventions.
We found similar relations among motivational perceptions and aca-
demic achievement, satisfaction, and retention for students with and
without a self-reported history of reading difficulty. Given that students
with no history of reading difficulty may also benefit from efforts to
support motivational beliefs, interventions might target any student
who experiences academic difficulty. Alternatively, educators could ask
students to self-report on their past and current reading difficultie-
s—using the ARHQ-R or a similar instrument—to identify and recruit
students who are academically vulnerable due to their reading abilities.
Acknowledgements
This research was funded by Employment and Social Development
Canada's Office of Literacy and Essential Skills (Agreement
#011684222) and by Social Sciences and Humanities Research Council
Partnership Development Grant 890-2011-0072. We thank Language
and Literacy Lab manager Annie Laroche for her help with study lo-
gistics.
References
Alfonso, V. C., Allison, D. B., Rader, D. E., & Gorman, B. S. (1996). The extended sa-
tisfaction with life scale: Development and psychometric properties. Social Indicators
Research, 38(3), 275–301. https://doi.org/10.1007/BF00292049.
Allen, J., Robbins, S. B., Casillas, A., & Oh, I. S. (2008). Third-year college retention and
transfer: Effects of academic performance, motivation, and social connectedness.
Research in Higher Education, 49(7), 647–664.
Baird, G. L., Scott, W. D., Dearing, E., & Hamill, S. K. (2009). Cognitive self-regulation in
youth with and without learning disabilities: Academic self-efficacy, theories of in-
telligence, learning vs. performance goal preferences, and effort attributions. Journal
of Social and Clinical Psychology, 28(7), 881–908. https://doi.org/10.1521/jscp.2009.
28.7.881.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change.
Psychological Review, 84(2), 191. https://doi.org/10.1037/0033-295X.84.2.191.
Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory.
Englewood Cliffs, NJ: Prentice Hall.
Bandura, A. (1997). Self-efficacy: The exercise of control. London: Macmillan.
Barron, K. E., & Hulleman, C. S. (2015). Expectancy-value-cost model of motivation.
Psychology, 84, 261–271. https://doi.org/10.1016/B978-0-08-097086-8.26099-6.
Battle, A., & Wigfield, A. (2003). College women's value orientations toward family, ca-
reer, and graduate school. Journal of Vocational Behavior, 62(1), 56–75. https://doi.
org/10.1016/S0001-8791(02)00037-4.
Bean, J., & Eaton, S. B. (2001). The psychology underlying successful retention practices.
Journal of College Student Retention, 3(1), 73–89.
Bean, J. P., & Eaton, S. B. (2000). A psychological model of college student retention. In J.
M. Braxton (Ed.). Reworking the student departure puzzle (pp. 48–61). Nashville, TN:
Vanderbilt University Press.
Bergey, B. W., Deacon, S. H., & Parrila, R. K. (2017). Metacognitive reading and study
strategies and academic achievement of university students with and without a his-
tory of reading difficulties. Journal of Learning Disabilities, 50(1), 81–94. https://doi.
org/10.1177/0022219415597020.
Bergin, D. A. (1999). Influences on classroom interest. Educational Psychologist, 34(2),
87–98.
Bong, M. (2001). Role of self-efficacy and task-value in predicting college students' course
performance and future enrollment intentions. Contemporary Educational Psychology,
26(4), 553–570. https://doi.org/10.1006/ceps.2000.1048.
Bouffard-Bouchard, T., Parent, S., & Larivee, S. (1991). Influence of self-efficacy on self-
regulation and performance among junior and senior high-school age students.
International Journal of Behavioral Development, 14(2), 153–164. https://doi.org/10.
1177/016502549101400203.
Brown, K. J. (2013). The experiences of individuals who are diagnosed with learning dis-
abilities in adulthood: A qualitative study (Doctoral dissertation)New York University.
Chen, F. F. (2007). Sensitivity of goodness of fit indexes to lack of measurement in-
variance. Structural Equation Modeling, 14(3), 464–504. https://doi.org/10.1080/
10705510701301834.
B.W. Bergey et al.
Chevalier, T. M., Parrila, R., Ritchie, K. C., & Deacon, S. H. (2017). The role of meta-
cognitive reading strategies, metacognitive study and learning strategies, and beha-
vioral study and learning strategies in predicting academic success in students with
and without a history of reading difficulties. Journal of Learning Disabilities, 50(1),
34–48. https://doi.org/10.1177/0022219415588850.
Chiang, E. S., Byrd, S. P., & Molin, A. J. (2011). Children's perceived cost for exercise:
Application of an expectancy-value paradigm. Health Education & Behavior, 38(2),
143–149. https://doi.org/10.1177/1090198110376350.
Cohen, J. (1986). Learning disabilities and psychological development in childhood and
adolescence. Annals of Dyslexia, 36(1), 287–300. https://doi.org/10.1007/
BF02648035.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale,
NJ: Lawrence Earlbaum Associates.
Conley, A. M. (2012). Patterns of motivation beliefs: Combining achievement goal and
expectancy-value perspectives. Journal of Educational Psychology, 104(1), 32–47.
https://doi.org/10.1037/a0026042.
Corkett, J. K., Hein, S. F., & Parrila, R. (2008). Compensating for reading difficulties: A
qualitative investigation of university students' experiences of influential personal
characteristics. Exceptionality Education International, 18(2), 51–68.
Davenport, T. J., & Lane, A. M. (2006). Relationships between self-efficacy, coping and
student retention. Social Behavior and Personality, 34, 127–138.
Deacon, S. H., Cook, K., & Parrila, R. (2012). Identifying high-functioning dyslexics: Is
self-report of early reading problems enough? Annals of Dyslexia, 62(2), 120–134.
https://doi.org/10.1007/s11881-012-0068-2.
Deacon, S. H., Tucker, R., & Bergey, B. W. (2015). The role of metalinguistic and socio-
cognitive factors in reading skill. In E. Segers, & P. van den Broek (Eds.).
Developmental perspectives in written language and literacy. Amsterdam: John
Benjamins.
Deacon, S. H., Tucker, R., Bergey, B. W., Laroche, A., & Parrila, R. (2017). Personalized
outreach to university students with a history of reading difficulties: Early screening
and outreach to support academically at-risk students. Journal of College Student
Development, 58(3), 432–450. https://doi.org/10.1353/csd.2017.0032.
DeDeppo, L. M. (2009). Integration factors related to the academic success and intent to
persist of college students with learning disabilities. Learning Disabilities Research and
Practice, 24(3), 122–131. https://doi.org/10.1111/j.1540-5826.2009.00286.x.
Eccles, J. (2009). Who am I and what am I going to do with my life? Personal and col-
lective identities as motivators of action. Educational Psychologist, 44(2), 78–89.
https://doi.org/10.1080/00461520902832368.
Eccles, J. S., Adler, T. F., Futterman, R., Goff, S. B., Kaczala, C. M., Meece, J. L., &
Midgley, C. (1983). Expectancies, values, and academic behaviors. In J. T. Spence
(Ed.). Achievement and achievement motivation (pp. 75–146). San Francisco, CA: W. H.
Freeman.
Eccles, J. S., & Wigfield, A. (1995). In the mind of the actor: The structure of adolescents'
achievement task values and expectancy-related beliefs. Personality and Social
Psychology Bulletin, 21(3), 215–225. https://doi.org/10.1177/0146167295213003.
Eckes, S. E., & Ochoa, T. A. (2005). Students with disabilities: Transitioning from high
school to higher education. American Secondary Education, 33(3), 6–20.
Fink, R. P. (1995). Successful dyslexics: A constructivist study of passionate interest
reading. Journal of Adolescent & Adult Literacy: A Journal from the International Reading
Association, 39(4), 268–280. https://doi.org/10.1007/s11881-998-0014-5.
Flake, J. K., Barron, K. E., Hulleman, C., McCoach, B. D., & Welsh, M. E. (2015).
Measuring cost: The forgotten component of expectancy-value theory. Contemporary
Educational Psychology, 41, 232–244. https://doi.org/10.1016/j.cedpsych.2015.03.
002.
Garriott, P. O., Hudyma, A., Keene, C., & Santiago, D. (2015). Social cognitive predictors
of first- and non-first-generation college students' academic and life satisfaction.
Journal of Counseling Psychology, 62(2), 253–263. https://doi.org/10.1037/
cou0000066.
Hampton, N. Z., & Mason, E. (2003). Learning disabilities, gender, sources of efficacy,
self-efficacy beliefs, and academic achievement in high school students. Journal of
School Psychology, 41(2), 101–112. https://doi.org/10.1016/S0022-4405(03)
00028-1.
Hartwell, M., & Kaplan, A. (2018). Students' personal connection with science:
Investigating the multidimensional phenomenological structure of self-relevance. The
Journal of Experimental Education, 86(1), 86–104.
Hen, M., & Goroshit, M. (2014). Academic procrastination, emotional intelligence, aca-
demic self-efficacy, and GPA: A comparison between students with and without
learning disabilities. Journal of Learning Disabilities, 47(2), 116–124. https://doi.org/
10.1177/0022219412439325.
Hidi, S., & Harackiewicz, J. M. (2000). Motivating the academically unmotivated: A
critical issue for the 21st century. Review of Educational Research, 70(2), 151–179.
https://doi.org/10.2307/1170660.
Ishitani, T. T., & DesJardins, S. L. (2002). A longitudinal investigation of dropout from
college in the United States. Journal of College Student Retention, 4(2), 173–201.
Jose, P. E. (2013). Doing statistical mediation and moderation. New York: Guilford Press.
Kaplan, C. P., & Shachter, E. (1991). Adults with undiagnosed learning disabilities:
Practice considerations. Families in Society, 72(4), 195–201.
Klassen, R. M., Krawchuk, L. L., & Rajani, S. (2008). Academic procrastination of un-
dergraduates: Low self-efficacy to self-regulate predicts higher levels of procrastina-
tion. Contemporary Educational Psychology, 33(4), 915–931. https://doi.org/10.1016/
j.cedpsych.2007.07.001.
Klassen, R. M., & Usher, E. L. (2010). Self-efficacy in educational settings: Recent research
and emerging directions. In S. A. Karabenick, & T. C. Urdan (Eds.). The decade ahead:
Theoretical perspectives on motivation and achievement (pp. 1–33). Bingley, England:
Emerald Group Publishing Limited.
Kline, R. B. (2010). Principles and practice of structural equation modeling (3rd ed.). New
51
Learning and Individual Differences 67 (2018) 41–52
York: Guilford Press.
Kosovich, J. J., Hulleman, C. S., Barron, K. E., & Getty, S. (2015). A practical measure of
student motivation: Establishing validity evidence for the expectancy-value-cost scale
in middle school. Journal of Early Adolescence, 35(5–6), 790–816. https://doi.org/10.
1111/j.1467-8624.2007.00995.x.
Lefly, D. L., & Pennington, B. F. (1991). Spelling errors and reading fluency in compen-
sated adult dyslexics. Annals of Dyslexia, 41(1), 141–162. https://doi.org/10.1007/
BF02648083.
Lefly, D. L., & Pennington, B. F. (2000). Reliability and validity of the adult reading
history questionnaire. Journal of Learning Disabilities, 33(3), 286–296. https://doi.
org/10.1177/002221940003300306.
Lent, R. W. (2004). Toward a unifying theoretical and practical perspective on well-being
and psychosocial adjustment. Journal of Counseling Psychology, 51(4), 482–509.
https://doi.org/10.1037/0022-0167.51.4.482.
Lent, R. W., Singley, D., Sheu, H.-B., Gainor, K. A., Brenner, B. R., Treistman, D., & Ades,
L. (2005). Social cognitive predictors of domain and life satisfaction: Exploring the
theoretical precursors of subjective well-being. Journal of Counseling Psychology,
52(3), 429–442. https://doi.org/10.1037/0022-0167.52.3.429.
Lent, R. W., Singley, D., Sheu, H.-B., Schmidt, J. A., & Schmidt, L. C. (2007). Relation of
social-cognitive factors to academic satisfaction in engineering students. Journal of
Career Assessment, 15(1), 87–97. https://doi.org/10.1177/1069072706294518.
Łockiewicz, M., Bogdanowicz, K. M., & Bogdanowicz, M. (2014). Psychological resources
of adults with developmental dyslexia. Journal of Learning Disabilities, 47(6),
543–555. https://doi.org/10.1177/0022219413478663.
Madaus, J. W. (2005). Navigating the college transition maze: A guide for students with
learning disabilities. Teaching Exceptional Children, 37(3), 32–37. https://doi.org/10.
1177/004005990503700305.
Margolis, H., & McCabe, P. P. (2004). Self-efficacy: A key to improving the motivation of
struggling learners. The Clearing House: A Journal of Educational Strategies, Issues and
Ideas, 77(6), 241–249. https://doi.org/10.1080/10459880309603362.
McGonnell, M., Parrila, R., & Deacon, S. H. (2007). The recruitment and description of
university students who self-report difficulty acquiring early reading skills.
Exceptionality Education Canada, 17(1/2), 155–174.
Multon, K. D., Brown, S. D., & Lent, R. W. (1991). Relation of self-efficacy beliefs to
academic outcomes: A meta-analytic investigation. Journal of Counseling Psychology,
38(10), 30–38. https://doi.org/10.1037/0022-0167.38.1.30.
Muthén, L. K., & Muthén, B. O. (2012). Mplus user's guide (Seventh Edition). Los Angeles,
CA: Muthén & Muthén.
Neuville, S., Frenay, M., Schmitz, J., Boudrenghien, G., Noël, B., & Wertz, V. (2007).
Tinto's theoretical perspective and expectancy-value paradigm: A confrontation to
explain freshmen's academic achievement. Psychologica Belgica, 47(1), 31–50.
https://doi.org/10.5334/pb-47-1-31.
Ojeda, L., Flores, L. Y., & Navarro, R. L. (2011). Social cognitive predictors of Mexican
American college students' academic and life satisfaction. Journal of Counseling
Psychology, 58(1), 61–71. https://doi.org/10.1037/a0021687.
Orenstein, M. (2000). Picking up the clues: Understanding undiagnosed learning dis-
abilities, shame, and imprisoned intelligence. Journal of College Student Psychotherapy,
15(2), 35–46.
Pajares, F. (1996). Self-efficacy beliefs in academic settings. Review of Educational
Research, 66(4), 543–578. https://doi.org/10.3102/00346543066004543.
Parrila, R., Corkett, J., Kirby, J., & Hein, S. (2003). Adult reading history questionnaire-
Revised. Unpublished questionnaireUniversity of Alberta.
Parrila, R., Georgiou, G., & Corkett, J. (2007). University students with a significant
history of reading difficulty: What is and is not compensated? Exceptionality Education
Canada, 17, 195–220. https://doi.org/10.1037/t52486-000.
Perez, T., Cromley, J. G., & Kaplan, A. (2014). The role of identity development, values,
and costs in college STEM retention. Journal of Educational Psychology, 106(1),
315–329. https://doi.org/10.1037/a0034027.
Rodgers, K. A., & Summers, J. J. (2008). African American students at predominantly
white institutions: A motivational and self-systems approach to understanding re-
tention. Educational Psychology Review, 20(2), 171–190. https://doi.org/10.1007/
s10648-008-9072-9.
Ryan, R. M., & Deci, E. L. (2001). On happiness and human potentials: A review of re-
search on hedonic and eudaimonic well-being. Annual Review of Psychology, 52(1),
141–166. https://doi.org/10.1146/annurev.psych.52.1.141.
Schertzer, C. B., & Schertzer, S. M. B. (2004). Student satisfaction and retention: A con-
ceptual model. Journal of Marketing for Higher Education, 14(1), 79–91. https://doi.
org/10.1300/J050v14n01_05.
Schneider, M., & Preckel, F. (2017). Variables associated with achievement in higher
education: A systematic review of meta-analyses. Psychological Bulletin, 143(6),
565–600.
Schunk, D. H. (1991). Self-efficacy and academic motivation. Educational Psychologist,
26(3–4), 207–231. https://doi.org/10.1080/00461520.1991.9653133.
Selkirk, L. C., Bouchey, H. A., & Eccles, J. S. (2011). Interactions among domain-specific
expectancies, values, and gender: Predictors of test anxiety during early adolescence.
Journal of Early Adolescence, 31(3), 361–389. https://doi.org/10.1177/
0272431610363156.
Sheu, H. B., Mejia, A., Rigali-Oiler, M., Primé, D. R., & Chong, S. S. (2016). Social cog-
nitive predictors of academic and life satisfaction: Measurement and structural
equivalence across three racial/ethnic groups. Journal of Counseling Psychology, 63(4),
460–474. https://doi.org/10.1037/cou0000158.
Sideridis, G. D. (2009). Motivation and learning disabilities: Past, present, and future. In
K. R. Wenzel, A. Wigfield, K. R. Wenzel, & A. Wigfield (Eds.). Handbook of motivation
at school (pp. 605–625). New York, NY, US: Routledge/Taylor & Francis Group.
Starr, A., Betz, E. L., & Menne, J. (1972). Differences in college student satisfaction:
Academic dropouts, nonacademic dropouts and nondropouts. Journal of Counseling
B.W. Bergey et al.
Psychology, 19(4), 318–322. https://doi.org/10.1037/h0033083.
Tinto, V. (1975). Dropout from higher education: A theoretical synthesis of recent re-
search. Review of Educational Research, 45(1), 89–125.
Tinto, V. (1993). Leaving college: Rethinking the causes and cures of student attrition (2nd
ed.). Chicago: The University of Chicago Press.
Tinto, V. (2007). Taking student retention seriously. Syracuse, NY: Syracuse University.
Vogel, S. A., & Adelman, P. B. (1992). The success of college students with learning
disabilities: Factors related to educational attainment. Journal of Learning Disabilities,
25(7), 430–441. https://doi.org/10.1177/002221949202500703.
Watt, H. M. (2004). Development of adolescents' self-perceptions, values, and task per-
ceptions according to gender and domain in 7th-through 11th-grade Australian stu-
dents. Child Development, 75(5), 1556–1574. https://doi.org/10.1111/j.1467-8624.
2004.00757.x.
Wefald, A. J., & Downey, R. G. (2009). Construct dimensionality of engagement and its
relation with satisfaction. The Journal of Psychology, 143(1), 91–112. https://doi.org/
52
Learning and Individual Differences 67 (2018) 41–52
10.3200/JRLP.143.1.91-112.
Wessel, R. D., Jones, J. A., Markle, L., & Westfall, C. (2009). Retention and graduation of
students with disabilities: Facilitating student success. Journal of Postsecondary
Education and Disability, 21(3), 116–125.
Wigfield, A., & Cambria, J. (2010). Students' achievement values, goal orientations, and
interest: Definitions, development, and relations to achievement outcomes.
Developmental Review, 30(1), 1–35. https://doi.org/10.1016/j.dr.2009.12.001.
Wigfield, A., & Eccles, J. S. (2000). Expectancy–value theory of achievement motivation.
Contemporary Educational Psychology, 25(1), 68–81. https://doi.org/10.1006/ceps.
1999.1015.
Wigfield, A., Tonks, S., & Eccles, J. S. (2004). Expectancy-value theory in cross-cultural
perspective. In D. M. McInerney, & S. Van Etten (Vol. Eds.), Big theories revisited:
Research on sociocultural influences on motivation and learning. Vol. 4. Big theories re-
visited: Research on sociocultural influences on motivation and learning (pp. 165–198).
Greenwich, CT: Information Age Publishing.