Journal of Educational Psychology
2001, Vol. 93, No. 1, 55-64
Copyright 2001 by the American Psychological Association, Inc.
0O22-O663/01/S5.OO DOI: 10.1037//0022-0663.93.1.55
Academic Self-Efficacy and First-Year College Student
Performance and Adjustment
Martin M. Chemers, Li-tze Hu, and Ben F. Garcia
University of California, Santa Cruz
A longitudinal study of lst-year university student adjustment examined the effects of academic self-efficacy and optimism on students' academic performance, stress, health, and commitment to remain in school. Predictor variables (high school grade-point average, academic self-efficacy, and optimism) and moderator variables (academic expectations and self-perceived coping ability) were measured at the end of the first academic quarter and were related to classroom performance, personal adjustment, stress, and health, measured at the end of the school year. Academic self-efficacy and optimism were strongly related to performance and adjustment, both directly on academic performance and indirectly through expectations and coping perceptions (challenge-threat evaluations) on classroom performance, stress, health, and overall satisfaction and commitment to remain in school. Observed relationships corresponded closely to the hypothesized model.
Change can be unsettling. The transition from high school to college can place significant demands on young adults (Tinto,
1982,1993). College life can be demanding and stressful for a new student (Noel, Levitz, & Saluri, 1985) and requires higher levels of independence, initiative, and self-regulation (Bryde & Milburn,
1990). It is the thesis of this article that confidence in one's relevant abilities (i.e., self-efficacy) and optimism play a major role in an individual's successful negotiation of challenging life transitions. We present evidence to support a theoretical model of the effects, both direct and indirect, of self-efficacy and optimism on academic performance and personal adjustment of lst-year college students. We first argue for, and then test, a model in which moderator variables such as academic expectations and selfperceptions of coping ability mediate the effects of efficacy and optimism beliefs. Tinto (1993) argued that the key determinant of persistence and success at college is commitment. This emphasis on intention or commitment is shared by many researchers in this field (Anderson, 1985; Bean, 1990; Pascarella & Terenzini, 1980), and researchers agree that both personal and environmental characteristics interact in the prediction of persistence (Anderson,
1985; Bean, 1985, 1990; Pascarella & Terenzini, 1980; Rocha-
Singh, 1990; Russell & Petrie, 1992). Both Tinto (1993) and Bean
(1990) have talked about aptitudes and capabilities as contributing to a sense of academic confidence or efficacy that helps to determine goal commitment. There are some very good reasons for focusing attention more closely on academic self-efficacy as a
Martin M. Chemers, Li-tze Hu, and Ben F. Garcia, Division of Social
Sciences, University of California, Santa Cruz.
The authors would like to thank the Division of Student Affairs at the
University of California, Santa Cruz, for providing financial support for this research.
Correspondence concerning this article should be addressed to Martin
M. Chemers, Division of Social Sciences, University of California, Santa
Cruz, Santa Cruz, California 95064. Electronic mail may be sent to mchemers@zzyx.ucsc.edu. central determinant of the success of high school to university transitions. Self-Efficacy, Academic Persistence, and Success
Bandura (1997) described self-efficacy as "the belief in one's capabilities to organize and execute courses of action required to produce given attainments" (p. 3). Efficacy beliefs influence the particular courses of action a person chooses to pursue, the amount of effort that will be expended, perseverance in the face of challenges and failures, resilience, and the ability to cope with the demands associated with the chosen course.
Self-efficacy has been related to persistence, tenacity, and achievement in educational settings (Bandura, 1986; Schunk,
1981; Zimmerman, 1989). A meta-analysis of research in educational settings (Multon, Brown, & Lent, 1991) found that selfefficacy was related both to academic performance (r = .38) and to persistence (r = .34). The contribution of self-efficacy to educational achievement is based both on the increased use of specific cognitive activities and strategies and on the positive impact of efficacy beliefs on the broader, more general classes of metacognitive skills and coping abilities.
The evidence that self-efficacy is able to improve performance in specific cognitive areas is well developed, and it is also very clear that self-efficacy is much more than the reflection of contentspecific ability. In studies of mathematics problem solving, children with high efficacy were found to persist longer (Bouffard-
Bouchard, Parent, & Larivee, 1991) and to use more efficient problem-solving strategies (Collins, 1982) than low-efficacy learners. Even more impressive support for the independent contribution of efficacy beyond ability is provided by studies that manipulate, rather than measure, existing levels of efficacy. Bouffard-
Bouchard (1990) and Cervone and Peake (1986) manipulated efficacy beliefs of students by providing fictitious performance norms during feedback. Students in the positive feedback (i.e., high self-efficacy) condition set higher aspirations, showed greater
55
56 CHEMERS, HU, AND GARCIA strategic flexibility in the search for solutions, achieved higher performance, and were more accurate in evaluating the level of their performance than were students of equal ability who received less positive feedback.
Mediating Processes of Self-Efficacy Effects
Cognitive Processes
Confidence in one's ability to complete a task or solve a problem contributes to a more calm and thoughtful approach. For example, making decisions in complex environments requires integrating large amounts of diverse information, interpreting feedback, testing and revising knowledge, and implementing selection options. In an extensive series of studies of decision making,
Bandura and his associates (Bandura & Jourden, 1991; Bandura &
Wood, 1989; Wood & Bandura, 1989; Wood, Bandura, & Bailey,
1990) found that participants (usually MB.A. students) who were higher in decision-making self-efficacy, either through selfappraisal or through manipulated performance feedback, used more thoughtful and skillful analytic strategies for improving performance than did less efficacious participants.
Self-efficacy acts on a broader level through the more effective use of metacognitive strategies, which involve planning and selfregulation— skills that become increasingly important as an individual progresses through educational levels to environments that are less ordered and constrained (e.g., college or university life).
Metacognition involves the appraisal and control of one's cognitive activity (i.e., thinking about thinking) and making use of all the resources available in the task and social environment to achieve goal attainment (Zimmerman, 1995; Zimmerman &
Martinez-Pons, 1988). Students high in academic self-efficacy make greater use of effective cognitive strategies in learning, manage their time and learning environments more effectively, and are better at monitoring and regulating their own effort. Academic self-efficacy is related to students' confidence in mastering academic subjects, which in turn predicted grades in school.
Motivational Processes
Bandura (1997) argued that self-efficacy has its most powerful motivational effects through the process of cognized goals. Goals provide the basis for self-regulation of effort by providing a standard for judging the adequacy and effectiveness of goalrelevant effort and strategy (Bandura & Cervone, 1983). Specific and difficult (but not impossible) goals are strongly related to performance in a wide variety of tasks and settings (Locke &
Latham, 1990). Self-efficacy leads to higher goals being set (Wood et al., 1990; Zimmerman, Bandura, & Martinez-Pons, 1992), and high goals increase the positive effects of self-efficacy by providing an evaluative context to aid self-regulation (Cervone, Jiwani,
& Wood, 1991). When goals provide a standard, highly efficacious persons show a stronger relationship among self-evaluation, selfdirection, and performance (Bandura & Schunk, 1981). Goals, and the broader category of positive expectations, are one type of vehicle by which efficacy effects are manifested.
Affective Processes
Anxiety and negative emotions can be debilitating. Self-efficacy has an impact on affect through its effects on attention and construal of environmental demands, by the choice of actions taken, and through its effect on the ability to control and manage negative or potentially negative emotions.
Lazarus and Folkman (1984) argued that the way in which an individual construes the demands placed by the environment can have dramatic impact on his or her ability to cope with that environment. They made a distinction between regarding demands as "threats" versus "challenges." Bandura (1997) also argued that a high sense of coping efficacy encourages individuals to adopt courses of action designed to change hazardous environments.
People with high perceived efficacy are less likely to be immobilized by anxiety (Betz & Hackett, 1983; Krampen, 1988). For example, Meece, Wigfield, and Eccles (1990) found that effects of past performance on math anxiety and math performance were mediated by personal efficacy beliefs. Pintrich and DeGroot
(1990) also reported that it was the efficacy beliefs, rather than anxiety, that were predictive of academic achievement.
Self-efficacy beliefs are related to an enhanced ability to use effective problem-solving and decision-making strategies, to plan and manage one's personal resources more efficiently, to entertain more positive expectations, and to set higher goals. Very central to these self-efficacy effects seems to be the ability to manage the stressors created in demanding situations by means of a more positive analysis of extant risks and available coping resources, which results in the tendency to see demanding situations as challenges rather than threats.
Optimism and Positive Expectations
Optimism may also be a good candidate for a characteristic affecting challenge-threat evaluations. Leadership self-efficacy and general optimism have been shown to be related to superiors' judgments of effective performance in the leadership role
(Chemers, Watson, & May, 2000). Optimism is related to psychological well-being in a broad variety of demanding situations
(Scheier, Carver, & Bridges, 2000). Optimism is a dispositional tendency to hold generalized positive expectancies even ". . .when people confront adversity or difficulty in their lives" (Scheier et al., 2000, p. 3). More positive emotional reactions and expectations—from excitement to eagerness—are associated with the greater use of active, problem-focused coping rather than avoidance or withdrawal.
Positive expectancies also predict better reactions during transitions to new academic environments. Optimistic students report lower levels of psychological stress and loneliness and higher levels of social support and psychological and physical well-being
(Aspinwall & Taylor, 1992; Scheier & Carver, 1992). Optimistic medical students suffer less depression and anxiety (Stewart et al.,
1997), and optimistic lst-year law school students evidence less mood disturbance compared with less optimistic students (Segerstrom,
Taylor, Kemeny, & Fahey, 1998). Effects of optimism appear to be mediated by coping style. Both active (e.g., planning and problem solving) and passive (e.g., control of negative emotions) coping mediate the effects of optimism on the outcome variables (Segerstrom et al., 1998; Stewart et al., 1997).
Thus, optimism should play an important role in maintaining positive expectations and reactions to lst-year college life. The effects of optimism should be indirect; that is, they should operate through the impact of optimism on judgments and expectations. In
SELF-EFFICACY IN UNIVERSITY ADJUSTMENT 57 particular, we expected generalized positive expectations of optimists to be related to the more specific academic expectations, which we hypothesized to underlie classroom performance. Optimism should also have considerable influence on the perceptions of extant demands and available resources that are at the core of challenge and threat evaluations.
Challenge-Threat Evaluations as Moderators of the
Self-Efficacy-Adjustment Relationship
The most extensive operationalization, elaboration, and extension of the challenge-threat concept has been offered by Blascovich and his associates (Blascovich & Tomaka, 1996; Tomaka,
Blascovich, Kelsey, & Leitten, 1993). Their work is particularly relevant to the present study because of their focus on "motivated performance situations," which are situations that are "goal relevant for performers and require instrumental cognitive-behavioral responses by them" (Blascovich, Mendes, Lickel, & Hunter, in press, p. 3). Transitions to university life do indeed involve goalrelevant performance requiring effective response across an extended time period.
The perception of a situation as challenging or threatening depends on the individual's phenomenological experience of the relationship between situational demands and coping resources.
Threat occurs when the individual experiences resources as insufficient to meet demands, and challenge occurs when resources are felt to be adequate to demands. A considerable body of empirical work has shown that the challenge-threat evaluation is strongly related to affective responses and physiological reactivity, with threat evaluations associated with potentially pernicious patterns of cardiovascular responsivity.
Recently, Blascovich and Mendes (in press) more finely elucidated the factors that influence demand and resource evaluations.
Demand evaluations are based on perceptions of the amount of required effort, danger, and uncertainty involved in the particular performance situation. The 1st year of college might contain elements of all three of these factors in varying amounts. Clearly, estimates of required effort are very prominent in the new student's appraisal of the situation. Uncertainty about new friends, living conditions, and finances is also likely to be a part of many students' worries. The potential for academic failure, social embarrassment, or even physical threats (e.g., from sexual attack, drug or alcohol availability, etc.) may also enter into a new student's evaluations of situational demand.
Blascovich and Mendes (in press) described resource evaluations to include perceptions of one's knowledge and abilities relevant to the task, availability of external support, and dispositional characteristics. Students who come from scholastically excellent high schools with a strong college-prepatory curriculum may see themselves as possessing many resources. Available support networks of family, peers, faculty, or institutional structure also enhance resource evaluations. In particular, the present research hypothesizes that dispositional characteristics related to efficacy and optimism should have a very strong impact on resource perceptions. Students high in academic self-efficacy should see themselves as more able to meet the demands of the situation and should therefore be more likely to regard the 1st year of college as a challenge rather than a threat. Optimism, associated with generalized positive expectations and more active coping styles, should be related both to higher expectations for success in academic pursuits and to more positive expectations and reactions to adversity or emotional demand (i.e., stress).
The Hypothesized Model
In the present study, we surveyed the lst-year class at a state university. Measures of academic self-efficacy, optimism, and challenge-threat evaluation were taken immediately after the completion of the first quarter of university work but before students had received formal evaluation feedback. Measures of self-rated academic performance, future academic expectations, stress, health, and satisfaction were taken at the same time. In a follow-up survey conducted near the end of the first academic year, measures of academic performance (self-rated), academic expectations, stress, health, and adjustment were repeated. Faculty ratings of academic performance for all courses taken during the 1st year were also collected.
On the basis of our interpretation of the efficacy literature, we expected academic self-efficacy to have a profound impact on the academic performance and personal adjustment of lst-year college students as they navigate the demanding environment of university life. Self-efficacy was expected to exert direct and mediated effects on performance, health, and adjustment. Figure 1 displays the path diagram for the hypothesized model.
Previous findings on the relationship of self-efficacy to academic and cognitive performance (e.g., Zimmerman, 1989) suggest that the calm and thoughtful demeanor encouraged by confidence should result in direct effects of academic self-efficacy on classroom performance. Highly efficacious students should make better use of analytic strategies and metacognitive skills for managing the learning environment, resulting in better learning and better evaluations.
Given the importance of goals on perseverance and the effective use of self-regulatory strategies, we also expected academic selfefficacy to affect performance through its impact on academic expectations. However, we believe that expectations arise out of broad evaluations of the situation and its demands, so we expected an indirect path to expectations through challenge-threat evaluations.
We predicted that self-efficacy and optimism would influence challenge-threat evaluations, which would in turn have an effect on academic expectations, which would predict academic performance. The impact of self-efficacy on personal adjustment and health was presumed to be moderated by challenge-threat evaluations.
The challenge-threat evaluation seems most closely related to affective coping and stress management. We predicted that the effects of dispositional characteristics, academic self-efficacy, and optimism on perceived stress, health, and adjustment would be moderated by challenge-threat evaluations. Note that high school grade-point average (GPA) was included in the model as a covariate. Method
Overview
The participants in this study were members of the lst-year class at the
University of California, Santa Cruz. All lst-year students received a questionnaire packet during the 1st week of the winter quarter. The
58 CHEMERS, HU, AND GARCIA
GPA
OPT
ADJUST
Figure 1. The path diagram for the hypothesized model. Squares are used to represent measured variables, and the circle is used to represent latent variables (i.e., factors and errors). GPA = high school grade-point average;
ACAPERM = academic performance; ASR = self-rated academic performance; ACAEXP = academic expectations; EVAL = faculty ratings of academic performance; ASE = academic self-efficacy; CHALL = challenge-threat evaluations; OPT = optimism; HEALTH = health problems; ADJUST = adjustment.
University of California, Santa Cruz, campus has an unconventional grading policy involving written evaluations in place of formal grades, and at this point in the school year, the students had experienced one quarter of university-level work but had not yet received a detailed evaluation of their performance. Participants received a second packet of questionnaires in the last weeks of the spring quarter, close to the end of the academic year.
The questionnaires focused on academic and social adjustment during the 1st year. The first questionnaire elicited student self-reports regarding their perceived academic self-efficacy, social self-efficacy, and general optimism as well as ratings of current and expected academic performance, ratings of current and expected social adjustment and social support, assessments of the level of stress-eliciting demand in their academic and social environments and their abilities to cope with that demand (i.e., challenge-threat assessments), and ratings of stress and illness. The second questionnaire repeated the measures that were included in the first administration, with the exception of self-efficacy and optimism. (The present article reports the results of this study with respect to academic achievement and does not include results related to social adjustment. The effects for social adjustment were based on a different model that was tested separately.) Questionnaires were mailed to 1,600 lst-year students at their campus addresses in the eight residential colleges of the university. Each questionnaire was preceded by a letter from the college provost (the chief academic officer of the college) encouraging participation in the study. These letters were identical for all colleges. The provost's letter was followed by a letter from the principal investigator, which explained the purposes of the study
(i.e., "an attempt to understand the psychological and social factors that influence student adjustment to college") and included a plea for participation, release forms, and information regarding informed consent. As an added incentive for participation in each administration of the survey, all participants would become eligible for a lottery drawing for gift certificates at the campus bookstore, including one $250 prize and five $50 prizes.
Approximately 1 week after each mailing, research assistants attempted to call every student to encourage return of the questionnaire. At each contact, students were informed that participation was completely voluntary and that all responses would be kept anonymous and confidential.
Participants
The first questionnaire administration elicited complete responses from
373 (23%) students. The second questionnaire was mailed only to those participating students, and of those, 256 (69%) responded again. Table 1 provides a breakdown of the samples by demographic characteristics.
Responders and nonresponders to the second administration were compared on data from the first administration, and very slight and nonsignificant differences between the two groups were found.
Measures
High school GPA. Participants' secondary school GPAs were obtained from their university records (with permission).
Academic self-efficacy. An eight-item measure was developed for the present study. Participants were asked to rate on 7-point Likert scales their agreement with statements reflecting their confidence in their ability to
Table 1
Participants' Characteristics
Participant characteristics
Gender
Males
Females
Age
17
18
19
20+
Ethnicity
White
Hispanic
Asian
Other
Wave 1 (n = 373)
78(21%)
295 (79%)
3 (0.8%)
217 (60%)
142 (38%)
7 (0.3%)
210 (56%)
54 (15%)
58 (16%)
51 (14%)
Wave 2 (n = 256)
46 (18%)
210 (82%)
1 (0.4%)
147 (58%)
102 (40%)
5 (0.4%)
148 (58%)
38(15%)
42 (16%)
28(11%)
SELF-EFFICACY IN UNIVERSITY ADJUSTMENT 59 perform well academically. Following Bandura (1997), the measure was designed to reflect a variety of specific skills pertinent to academic achievement, including scheduling of tasks, note taking, test taking, and researching and writing papers, and included general statements regarding scholarly ability. Because the intent was to predict overall college performance, the self-efficacy measure did not focus on highly specific subjects such as math, science, languages, and so on. Coefficient alpha was .81.
Optimism. The Life Orientations Test (Scheier & Carver, 1985) was used as a measure of generalized optimism. Eleven items were rated on
5-point scales of agreement ranging from 1 (strongly agree) to 5 (strongly disagree). High scores on the measure indicate that respondents "believe that things usually work out well for them," "expect things to go their way," "are always optimistic," and so on. Scheier and Carver (1985) reported a coefficient alpha of .76 and test-retest reliability of .79 (at 4 weeks). For the present sample, the coefficient alpha was .80.
Challenge—threat. Using a procedure similar to that used by Blascovich and his colleagues (Blascovich & Tomaka, 1996), we asked participants to rate the "level of pressure and demand expected in your academic work during the next year" on a 7-point scale ranging from 1 (not much at all) to 7 (a very great deal). Following that response, we asked participants to rate their ability to cope with the rated level of academic pressure and demand (these ratings were made on the same 7-point scale). The challenge-threat score was derived by subtracting the rating of demand from the level of coping ability such that numbers larger than zero reflect the participants' belief that coping ability is adequate (challenge situation), and numbers below zero indicate that the participants perceive demand as exceeding coping ability (threat situation).
Academic self-rating. Participants rated their recent academic performance on a 5-point scale ranging from 1 (not good) to 5 (excellent). This measure was created for the present study.
Academic evaluations. The university has a grade-optional evaluation policy. For every class, all students receive a narrative evaluation written by the instructor. These evaluations are filed with the registrar and sent to the students approximately 3 to 5 weeks after the end of the academic quarter. Students may also request a grade in the course, and approximately
20% of the students choose that option. However, this percentage varies by major. Because grades were not available for all participants, we chose to use the narrative evaluations as the measure of classroom performance.
Raters were trained to convert the narrative evaluations to a quantitative index. It is very common for narrative evaluations to follow a customary form and language. The use of common keywords (e.g., outstanding, excellent, satisfactory) makes them roughly comparable with levels of performance indicated by grades. Raters were trained to use the keywords to convert the evaluations to a quantitative score. The index ranged from 1 to 5. Interrater reliability for the evaluations averaged .85.
Academic expectations. Participants expressed their expectations for future academic performance by responding, on 5-point agreement scales
(1 = strongly agree, 5 = strongly disagree), to four items concerning performance in courses, getting good evaluations, meeting academic goals, and generally performing well academically. This measure was created for the present study. Unlike the measure of academic self-efficacy, which was focused on abilities in the present, the academic expectations measure was focused on outcomes (e.g., performance, continued enrollment) in the future. The coefficient alpha for this measure was .77 and .76 for the first and second administrations, respectively.
Stress. Experienced stress was measured with a scale developed for this study. Participants responded on a 5-point scale (1 = never, 2 = almost never, 3 = sometimes, 4 = fairly often, 5 = very often) eliciting the frequency with which they experienced feeling of stress, loss of control, or nervousness and irritability during the previous month. Examples include
"How often have you been upset because of something that happened unexpectedly," "How often have you felt nervous or 'stressed'," "How often have you felt you were unable to control the important things in your life," and so on. Coefficient alpha for this measure was .87 for both administrations. Health. A 20-item scale (created for this study) asked participants to rate on a 4-point scale (1 = rarely/none, 2 = some/little, 3 = occasionally,
4 = most/all) how often they had experienced a range of physical and mental health symptoms or problems, including sleeping problems; headaches or dizziness; stomach problems; eating problems; feelings of shame, loneliness, insecurity, panic, worthlessness, and depression; drug or alcohol overuse; and so on. Coefficient alphas for this measure were .90 for both administrations.
Adjustment. Pascarella and Terenzini's (1980) college social support scale includes subscales measuring satisfaction with academic progress
(Academic and Intellectual Development) and intention to persist at the university (Institutional and Goal Commitment).
Items from these subscales were combined to create a measure of satisfaction with academic progress and intention to continue at the university, which we labeled "adjustment." The resultant scale included eight items rated on 5-point scales of agreement (1 = strongly agree, 5 = strongly disagree). Examples of specific items include "I am satisfied with the extent of my intellectual development since enrolling at the university,"
"I am satisfied with my academic experience at this university," "I am confident that I made the right decision in choosing to attend this university," and "It is important for me to graduate from this college." Coefficient alphas of .78 were found for both administrations of the measure.
Results
A structural equation modeling (SEM) approach was used to test the adequacy of the hypothesized model (e.g., Bentler, 1980;
Joreskog & Sorbom, 1993). The EQS program was used for the analyses. The maximum-likelihood (ML)-based Comparative Fit
Index (CFI) and the Standardized Root Mean Square Residual
(SRMR) were used to evaluate the extent to which the hypothesized model fit the observed data (Hu & Bentler, 1998, 1999). In addition, the robust standard errors were used to evaluate the significance of parameter estimates (Bentler & Dudgeon, 1996;
Hu, Bentler, & Kano, 1992). The quality of solution for the model reported here was excellent (i.e., there was no convergence problem and no out-of-range parameter estimate).
Table 2 presents the means, standard deviations, and zero-order correlations for all of the observed variables. Because female students were overrepresented in our sample, we analyzed for gender effects. Participant gender was dummy coded and correlated with each of the variables. There were no substantial relationships
(rs ranged from .01 to .12), so gender was not included as a variable in the model, and male and female students were pooled for subsequent analyses.
For the hypothesized model, the SEM technique was used to assess (a) the direct effect of self-efficacy on challenge-threat evaluations, academic expectations, and academic performance
(measured by two indicators: self-rated academic performance and faculty ratings of academic performance); (b) the direct effect of optimism on self-efficacy, challenge-threat evaluations, and academic expectations; (c) the direct effect of challenge-threat evaluations on academic expectations and stress; (d) the direct effect of academic expectations on academic performance; (e) the direct effect of stress on health and adjustment; (f) the mediated effect of self-efficacy on academic expectations, academic performance, stress, health, and adjustment; and (g) the mediated effect of optimism on challenge-threat evaluations, academic expectations, academic performance, stress, health, and adjustment. The direct
60 CHEMERS, HU, AND GARCIA
Table 2
Design Variables
Variables
1. GPA
2. ASE
3. OPT
4. CHALL
5. ACAEXP
6. STRESS
7. ASR
8. EVAL
9. HEALTH
10. SATADJ
M
SD
Range
1
_
.29
.21
.12
.12
-.09
.24
.43
-.10
.04
3.57
0.36
2.57-4.40
2
—
.35
.32
.43
-.19
.51
.35
-.20
.18
5.40
0.82
2.57-7.00
3
—
.25
.32
-.17
.32
.15
-.18
-.00
3.45
0.69
1.67-5.00
4
—
.43
-.31
.21
.19
-.30
.12
6.39
1.77
1.00-12.00
5
—
-.29
.36
.23
-.22
.20
4.08
0.55
2.00-5.00
6
—
-.18
-.13
.66
-.25
2.71
0.67
1.10-4.90
7
—
.49
-.19
.13
3.28
1.04
1.00-5.00
8
—
-.06
.13
3.82
0.50
1.98-5.00
9
—
-.26
1.82
0.51
1.00-3.70
10
—
3.82
0.65
1.13-5.00
Note. GPA = grade-point average; ASE = academic self-efficacy; OPT = optimism; CHALL = challenge-threat evaluations; ACAEXP = academic expectations; ASR = self-rated academic performance; EVAL = faculty ratings of academic performance; HEALTH = health problems; SATADJ = satisfaction and adjustment. effect of high school GPA on self-efficacy and academic performance and the direct effect of academic performance on levels of stress and adjustment were also examined. Note that the inclusion of GPA allowed us to examine the effects of self-efficacy and optimism on various outcome variables after partialing out the effect of GPA. (Ideally, we would want to use latent variables in our structural equation model. This was not possible in the present study because of sample size limitations and the lack of multiple indicators for some of our variables. Therefore, some of the relationships reported may be slightly inflated or deflated.)
Figure 2 displays the path diagram and the standardized parameter estimates for the hypothesized model. The ML-based CFI was
.97, and the ML-based SRMR was .057. These fit measures reveal that there was a relatively good fit between the hypothesized model and the observed data.
The factor loading of faculty ratings of academic performance on the construct of academic performance (standardized factor loading = .65) was substantial in magnitude and was significant at p < .001. Note that the unstandardized factor loading for self-rated academic performance was fixed to 1 for identification purposes
(see Figure 2).
There were significant and substantial direct effects of selfefficacy on challenge-threat evaluations (standardized coefficient
= .27, p < .001), academic expectations (standardized coefficient = .28, p < .001), and academic performance (standardized coefficient = .34, p < .001). Highly efficacious students had higher challenge-threat evaluations (i.e., they perceived academic work demand to be more of a challenge than a threat), greater academic expectations, and better academic performance.
(An analysis of the zero-order correlations between academic self-efficacy and challenge-threat reveals that efficacy affected evaluations of coping resources [r = .39, p < .001] rather than evaluations of demands [r = —.01, ns].) Optimism had a moderate effect on self-efficacy (standardized coefficient = .31, p < .001) and a small effect on challenge-threat evaluations (standardized coefficient = .16, p < .01) and academic expectation (standardized coefficient = . 15, p < .01). Highly optimistic students tended to be more efficacious. They had more positive challenge-threat evaluations and higher academic expectations. (Again, the effects of optimism on challenge-threat evaluations, similar to academic efficacy effects, were on perceptions of resources [r = .28, p <
.001] rather than on perceptions of demands [r = -.08, ns].) The direct effects of challenge-threat evaluations on academic expectations
(standardized coefficient = .30, p < .001) and stress
(standardized coefficient = —.26, p < .001) were significant, indicating that students with higher challenge-threat evaluations had higher academic expectations and experienced less stress. The significant standardized path coefficient (.22, p < .001) between academic expectation and academic performance indicates that students with higher academic expectations had better academic performance. Students who experienced more stress tended to have greater health problems (standardized coefficient = .68, p < .001) and worse adjustment (standardized coefficient = — .20, p < .001).
Students with higher high school GPA were more efficacious
(standardized coefficient = .23, p < .001) and had better academic performance (standardized coefficient = .45, p < .001). The direct effects of academic performance on stress (standardized coefficient
= —.19,p < .01) and adjustment (standardized coefficient =
.19, p < .01) were significant but small in magnitude. These results suggest that students with better academic performance experienced less stress and were better adjusted.
There were significant mediated effects of self-efficacy on academic expectations (standardized coefficient = .08, p < .001), academic performance (standardized coefficient = .08, p < .01), stress (standardized coefficient = — .16, p < .001), health (standardized coefficient = -.15,p