The degree to which students accurately self-assess knowledge, competence, skills, and academic preparedness is related to students' ability to self-regulate learning. Prior research indicates that students who are more accurate in their self-assessments frequently perform at higher levels in the classroom than less accurate peers. In the present study, we examined developmental students' calibration accuracy and we also compared the calibration accuracy of developmental college students to that of regularly-admitted college students. Our findings indicate that developmental college students tend to overestimate their academic performance relative to actual performance. In addition, they are less accurate than nondevelopmental peers in self-evaluating their academic preparedness and competence. Implications for education are discussed.
Numerous studies have examined the relationship of self-regulated learning-the ability to organize, direct, and manage one's learning-to academic achievement (Bandura, 1986; Pintrich, 2000; Pintrich & DeGroot, 1990; Zimmerman & Bandura, 1994, 1989; Zimmerman & Martinez-Pons, 1990). Students who self-regulate their learning are actively engaged in their own learning processes, including analyzing the demands of academic assignments, planning for and organizing their resources to meet those demands, and monitoring their study tactics and achievement (Pintrich, 1995; Zimmerman & Martinez-Pons, 1990; Winne & Jamieson-Noel, 2002). In other words, they take responsibility for acquiring skills and knowledge rather than depending on external sources to regulate their learning. Students who successfully self-regulate their learning are able to monitor their study behavior and utilize a variety of study strategies. By definition, developmental college students possess academic deficiencies related to difficulty in managing and directing their own learning (The National Association for Developmental Education, 2001); therefore, examination of self-regulatory behaviors and cognitions is of interest to developmental educators.
Of particular interest to the self-regulation of learning is the degree to which students' self-evaluations of academic capability or performance actually represent competence. Self-evaluation is an essential component of self-regulated learning because accurate self-evaluations allow the learner to determine whether the learning strategies employed are effective and, if not, to make adjustments (Bandura, 1986; Zimmerman, 1989). In recent years, a growing body of research has investigated students' self-evaluation skill (Beyer, 1999; Hacker, BoI, Horgan, & Rakow, 2000; Koriat, Goldsmith, & Pansky, 2000; Maki, 1998; Pajares & Miller, 1997; Robins & Beer, 2001). Known as achievement calibration, this construct has been operationalized in many different ways, such as the degree to which students accurately predict test-item performance (Winne & Jamieson-Noel, 2001), test grades (Hacker et al., 2000), course grades (Jacobson, 1990), and capability to solve mathematics problems (Pajares & Miller, 1997). Some studies indicate that students tend to overestimate their test and course grades (Hacker, et al., 2000; Jacobson, 1990); however, this inaccuracy is not uniform across college students. Hacker et al. (2000) identified distinctions among college students based on achievement. Higher achieving students were more accurate and improved their accuracy over time, while the lowest achieving students grossly overestimated test performance and maintained this over-optimistic stance throughout the semester, regardless of contradictory feedback.
In general, the findings reveal a bias reflecting a positive illusion of performance or ability. This tendency toward self-enhancement is believed to be adaptive by some researchers (Taylor & Brown, 1988, 1994; Aspinwall & Taylor, 1992). However, others believe that "positive illusions may reflect a more general tendency to bolster self-esteem by denying information that threatens self-worth" (Robins & Beer, 2001, p. 340). In other words, students who are inaccurate calibrators may deny the validity of negative feedback (e.g. poor grades) and fail to take appropriate corrective action. Consistent with the latter view, recent studies indicate that more accurate calibration is related to more efficient study habits in college students (Horgan, 1990) and to well-developed self-assessment skills in medical students (Fitzgerald, Gruppen, White, & Davis, 1997). Similarly, Hacker et al. (2000) found that the lowest achieving college students displayed the lowest levels of achievement calibration and the highest degree of overconfidence.
Although approximately one-third of entering college freshmen are placed in at least one developmental course designed to overcome academic deficiencies (Boylan, 1995), few studies have examined the self-evaluation skills of these students to determine whether they differ from those of regular admission, nondevelopmental college students. Prior studies have identified a number of affective factors associated with developmental college students. For example, not only have these students experienced less academic success compared to higher achieving peers, but they often are the first in their families to attend college and may experience isolation, financial difficulties, passive learning style, external locus of control, and strong reliance on peers for assistance (Morrison, 1999; Terenzini, Springer, Yaeger, Pascarella, & Nora,, 1996; Larose & Roy, 1991; Higbee, et. al, 1991). Although they may indicate a desire to be successful in their college courses, many developmental students enter college without having developed the selfregulating skills necessary to achieve academic success (Ley & Young, 1998).
Research investigating low-achieving college students suggests that developmental students may differ from regular admission students in the way they plan, organize, monitor, and evaluate their study behavior, and also in the way they think about the learning process (Carr, Borkowski & Maxwell, 1991). Kruger and Dunning (1999) found that low-achieving students lack metacognitive awareness to distinguish accuracy from error and are unaware of their lack of skill. Results of a study comparing regular admission and developmental students (from a community college and a residential university) by Ley and Young (1998) indicated that self-regulation differs significantly between the regular admission and the developmental college students. Developmental students used fewer total strategies, and used them less consistently than did their regular admission counterparts. The researchers hypothesized that although the developmental students' motivational levels and beliefs about success are similar to the regular admission students', their lack of self-regulation may be the characteristic that distinguishes the two groups. These differences may suggest that developmental educators need to assist students in acquiring more accurate self-assessment skills to succeed in academically challenging college courses.
Another possible difference between developmental and regular admission college students is found in examining the relationship between college students' perceptions of their self-regulatory capabilities and achievement. For example, VanZile-Tamsen and Livingston (1999) found that lower achieving college students reported less self-regulated strategy use than higher achieving peers. Although the lower achieving students' beliefs paralleled the higher achieving students' beliefs about strategy use, their inability to self-regulate academic behavior negatively impacted their academic success.
The goal of developmental education is to prepare students to be successful in regular college courses. Because developmental college students are identified as having academic deficits and lack skills in self-regulation, the findings of the present study may provide developmental educators important information about how developmental students differ from nondevelopmental college students with regard to achievement calibration. The purposes of the present study were (1) to document the degree to which developmental college students are accurate in their achievement calibration and (2) to compare the achievement calibration accuracy of developmental college students to nondevelopmental university students. In the present study, we operationalized achievement calibration as students' accuracy in predicting final course grades or semester grade-point average. Given achievement differences between developmental students and nondevelopmental students, we expected differences in achievement calibration between the two groups.
Our sample was comprised of 473 students drawn from two post-secondary institutions. Two hundred eighty-six developmental reading students from a community college in the Midwest participated. One hundred ninety-nine were female and 87 were male. The average age was 26. Sixty-nine percent were African-American, 18% were Caucasian, 4% were Asian, 7% were Hispanic, and 10% indicated other or failed to respond. This sample was representative of the general student population in terms of age, gender and ethnicity. Participation in the study was voluntary and no incentives were provided to participants.
In addition, 187 nondevelopmental students from a research university in the Southeast participated. Ninety-five were female and 92 were male. The average age was 18. Sixty-four percent were Caucasian, 27% were African-American, and 9% indicated other or failed to respond.
Procedures and Measures
Students enrolled in developmental reading classes at the community college were asked to predict their grade in the class during the 10th week of the semester. The question was embedded in an instrument assessing students' use of learning strategies and was "What grade do you expect to earn in this course?" Students responded by circling a letter grade (A, A-, B+, B, B-, and so on). At the end of the semester, course grades were obtained from instructors and semester GPAs were obtained from students' college records.
At the research university, similar procedures were used to obtain expected and actual semester grade point average. ["Please state the grade point average (GPA) that you expect to earn this semester"]. These students were asked to predict their semester GPA because the course in which the data were collected was a university orientation course in which almost all freshmen earn As. Expected grades were obtained during the second week of the semester. At the end of the semester, GPAs were obtained from university records.
For our calculation of achievement calibration, we chose final course grade and semester GPA as indicators of achievement rather than individual test items or test scores. Gigerenzer (1991) indicates a prevalence of the over-confidence bias when expectations and probabilities of single events, such as the estimation of the correctness of single test items, are studied. He cites several studies from social psychology in which the over-confidence bias disappears when a more global indicator is used (e.g. score on a test as opposed to correctness of a single item).
Descriptive statistics for students' predicted and actual grades broken down by group (developmental versus regular) are presented in Table 1. Calibration refers to the difference between the predicted and actual grades. As can be seen, the means for the predicted grades in the two groups appear quite similar, whereas the actual grades of the developmental students appear to be lower that of the regular students.
To test the hypothesis that developmental college students overestimate their grades, we conducted a dependent measure t-test which compared the predicted and actual grades of the developmental students. The results confirmed our hypothesis, t (285) = -11.86; p < .001.
To test the hypothesis that developmental college students overestimate their grades more so than nondevelopmental university students, we conducted a repeated-measures ANOVA using the predicted and actual grades as dependent variables and the student group as the independent variable. The repeated measures procedure created a dependent variable that we called calibration, which is based on the difference between the actual and predicted grades. There was no main effect of student group, F (I, 469) = 3.28; p > .05. However, there was a main effect of calibration, F (1,469) = 155.95; p < .001, suggesting that students as a whole have higher predicted than actual grades. This was modified by an interaction between calibration and type of student, F (1, 469) = 17.95; p < .001. The interaction is depicted in Figure 1 and supports our hypothesis that developmental students overestimate their grades by a larger degree than do nondevelopmental university students.
Although many students in this study overestimated their grades, the range on the difference score in Table 1 suggests that some of these students were accurate and others actually underestimated their grades. Thus, to further explore the characteristics of accurate and inaccurate calibrators, we created three groups based on the difference between the predicted and actual grades: accurate estimators, whose predicted and actual grade did not vary by more than .5 on the 0.0 to 4.0 scale; over estimators, whose predicted grades exceeded the actual grades by a score of .51 or above; and under estimators, whose actual grades exceeded the predicted grades by .51 or more. As can be seen from the crosstabs analysis in Table 2, few students underestimated their grades and the percentage in this category was similar across nondevelopmental university and developmental students. However, there were a higher percentage of developmental students who overestimated their grades versus non developmental university students, and thus a lower percentage that were accurate in their self ratings. Table 2 also reports the percentage of students within each type (developmental versus nondevelopmental) in each estimation group.
We also examined the academic achievement of students who were over estimators, accurate estimators and under estimators. Based on previous research, we expected that, within both groups, overestimators would be students with the lowest grades. We conducted a 2 (type of student) by 3 (estimation group) ANOVA with grades as the dependent variable. There was a main effect of estimation group. Students who were over estimators had lower grades than did the under estimators or accurate estimators. There were no grade differences between accurate and under estimators. There was also a significant interaction of estimation group by type of student. As can be seen in Figure 1, the underestimating students had lower grades that the other two groups, and, when comparing nondevelopmental university to developmental students, we see that there are no differences in the average achievement of the accurate estimators and under estimators. However, among the over estimators, the grades of the developmental students were lower than those of the nondevelopmental university students. These findings suggest that the students most prone to overestimating their grades are low-achieving developmental students.
In this study, we explored the achievement calibration accuracy of developmental students in a community college setting. Calibration is considered an important aspect of students' self-regulation of their learning because students' estimates of their predicted achievement provide an important source of information to adjust their level of effort and to make decisions about employing strategies such as help seeking. It was therefore not surprising that developmental students as a group over predicted their competence, and that their level of over prediction was greater than that of nondevelopmental university students. Moreover, whereas there were no differences in the achievement of students who were accurate or under estimators, distinctions arose among students who overestimated their preparedness or competence. Although approximately the same percentage of students were under estimators in both groups, the majority of nondevelopmental university students (54%) were accurate estimators while the majority of developmental students were over estimators (53%). The interaction between calibration accuracy and type of student was another indicator of differences in self-evaluative skills between nondevelopmental university and developmental college students. Our hypotheses were supported indicating that developmental college students tend to overestimate their academic performance relative to actual performance. In addition, they are less accurate than nondevelopmental peers in self-evaluating their academic preparedness and competence.
Implications for Developmental Educators
Developmental educators may benefit from knowing that, in general, developmental students have a tendency to over estimate their academic preparedness and competence, more so than students in regular college classrooms. Other studies have found factors that influence students' overestimation of achievement include reading ability, metacognitive knowledge regarding task requirements and text difficulty (Lin, Moore, & Zabrucky, 2001), and motivational orientation (Kroll & Ford, 1992)-all potential deficits in developmental college students. Although Bandura (1986) argues that some overestimation of academic competence is useful because it increases effort and persistence, students who lack the metacognitive skills that enable them to accurately self-evaluate tend to maintain inflated views of their performance and abilities. Self-regulation is particularly important with regard to developmental college students because these cognitive and metacognitive strategies can be learned, as opposed to being stable personality traits or states of the learner (Pintrich & Garcia, 1994). Therefore, diagnosing students' self-evaluation skills may help them better understand what they know and do not know so that they may utilize strategies more effectively and increase their academic achievement.
Because some research suggests that overly optimistic self-evaluations are resistant to change (Hacker et al, 2000), a systematic approach is recommended for helping developmental students improve their calibration accuracy. Effective instructional strategies involve self-evaluation of academic activities and preparedness by the student with frequent evaluative feedback from the course instructor. More frequent feedback affords the student more opportunities to compare self-evaluations with instructor evaluations. However, given the resistant nature of overly positive self-appraisals, self-reflection and discussion with the instructor are also recommended to examine and discuss evaluative differences. Such an intervention may also involve self-monitoring of study time and other learning activities (Ormrod, 2003).
Limitations of the study include that the sample was drawn from postsecondary institutions in different parts of the country. Any differences between the community college and university students could be due to a variety of factors related to regional differences. In addition, the mean age of the community college students (26) was higher than the mean age of the university students (18); therefore, developmental issues might confound comparisons although this age differential might likely exist when comparing any groups of community college and research university students. Future research might seek to include non-traditional freshmen within the university sample. Further, students in the community college group were asked to predict expected grade in a course, while university students predicted semester grade point average. Future research might seek classroom settings that allow for the same benchmark to be used. Although, this does not appear to be an issue, given that differences were not uniform across settings (i.e. all community college students did not differ from all university students in any comparison).
Aspinwall, L. G., & Taylor, S. E. (1992). Modeling cognitive adaptation: A longitudinal investigation of the impact of individual differences and coping on college adjustment and performance, Journal of Personality and Social Psychology, 63, 989-1003.
Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice Hall.
Beyer, S. (1999). Gender differences in the accuracy of grade expectancies and evaluations. Sex Roles, 41, 279-296.
Boylan, H. R. (1995). Making the case for developmental education. Research in Developmental Education, J2(2), 1-4.
Carr, M., Borkowski, J., & Maxwell, S. (1991). Motivational components of underachievement. Developmental Psychology, 27(1), 108-118.
Fitzgerald, J. T., Gruppen, L., White, C, & Davis, W. K. (1997). Medical student self-assessment abilities: Accuracy and calibration. Paper presented at the annual meeting of the American Educational Research Association, Chicago.
Gigerenzer, G. (1991). How to make cognitive illusions disappear: Beyond "Heuristics and Biases." European Review of Social Psychology, 2, 83-115.
Hacker, D., BoI, L., Horgan, D., & Rakow, E.A. (2000). Test prediction and performance in a classroom context, Journal of Educational Psychology, 92 (1), 160-170.
Higbee, J. L., Dwinell, P. L., McAdams, C. R., Goldberg, B. E., & Tardola, M. E. (1991). Serving underprepared students in institutions of higher education. Journal of Humanistic Education and Development, 30(2), 73-80.
Horgan, D. (1990, April). Students' predictions of test grades: Calibration and metacognition. Paper presented at the annual meeting of the American Educational Research Association, Boston. (ERIC Document Reproduction Service No. ED319812)
Jacobson, J. M. (1990). Congruence of pretest predictions and posttest estimations with grades on short answer and essay tests. Educational Research Quarterly, 14(2), 41-47.
Koriat, A., Goldsmith, M., & Pansky, A. (2000). Toward a psychology of memory accuracy. Annual Review of Psychology, 51, 481-537.
Kroll, M. D., & Ford, M. L. (1992). The illusion of knowing, error detection, and motivational orientations. Contemporary Educational Psychology, 17, 371-378.
Kruger, J. & Dunning, D. (1999). Unskilled and unaware of it: How difficulties in recognizing one's own incompetence lead to inflated self-assessments. Journal of Personality and Social Psychology, 77, 1121-1134.
Larose, S., & Roy, R. (1991). The role of prior academic performance and nonacademicattributes in the prediction of the success of high-risk college students. Journal of College Student Development, 32(2), 171 -177.
Ley, K., & Young, D. B. (1998). Self-regulation behaviors in underprepared (developmental) and regular admission college students. Contemporary Educational Psychology, 23, 42-64.
Lin, L. M., Moore, D., 7 Zabrucky, K. M. (2001). An assessment of students' calibration of comprehension and calibration of performance using multiple measures. Reading Psychology, 22(2), 111-128.
Maki, R. (1998). Predicting performance on text: Delayed versus immediate predictions and tests. Memory & Cognition, 26(5), 959-964.
Morrison, B. H. (1999). Acknowledging student attributes associated with academic motivation. Journal of Developmental Education, 23(2), 10-16, 30-31.
Ormrod, J. (2003). Educational psychology: Developing learners (4th ed.). Upper Saddle River, NJ: Merrill/Prentice Hall.
Pajares, F. & Miller, D. M. (1997). Mathematics self-efficacy and mathematical problem solving: Implications of using different forms of assessment. The Journal of Experimental Education, 65(3), 213-238.
Pintrich, P. R. (1995). Understanding self-regulated learning. New Directions for Teaching and Learning, no. 63, 3-12.
Pintrich, P. R. (2000). Multiple goals, multiple pathways: The role of goal orientation in learning and achievement. Journal of Educational Psychology, 92(3), 544-555.
Pintrich, P. R., & De Groot, E. (1990). Motivational and self-regulated learning components of classroom academic performance. Journal oj Educational Psychology, 82, 33-40.
Pintrich, P. R. & Garcia, T. (1994). Self-regulated learning in college students: Knowledge, strategies, and motivation. In P. R. Pintrich, D. R. Brown, & C. E. Weinstein (Eds.), Student motivation, cognition and learning (pp. 113-133). Hillsdale, NJ: Laurence Erlbaum.
Robins, R. W., & Beer, J. S. (2001). Positive illusions about the self: Short-term benefits and long-term costs. Journal of Personality and Social Psychology, 80(2), 340-352.
Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: A social psychological perspective on mental health. Psychological Bulletin, 103(2), 193-210.
Taylor, S. E., & Brown, J. D. (1994). Positive illusions and well-being revisited: Separating fact from fiction. Psychological Bulletin, 116, 21-27.
Terenzini, P. T., Springer, L., Yaeger, P. M., Pascarella, E. T., & Nora, A. (1996). First-generation college students: Characteristics, experiences, and cognitive development. Research in Higher Education, 37, 1-22.
The National Association for Developmental Education (2001). Online. Available at www.nade.net
VanZile-Tamsen, C., & Livingston, J. A. (1999). The differential impact of motivation on the self-regulated strategy use of high - and low - achieving college students. Journal of College Student Development, 40, 54-60.
Winne, P. H., & Jamieson-Noel, D. (2002). Exploring students' calibration of self reports about study tactics and achievement. Contemporary Educational Psychology, 27(4), 551-572.
Zimmerman, B. J. (1989). A social cognitive view of self-regulated academic learning. Journal of Educational Psychology, 81, 329-339.
Zimmerman, B. J. & Martinez-Pons, M. (1990). Student differences in self-regulated learning: Relating grade, sex, and giftedness to self-efficacy and strategy use. Journal of Educational Psychology, 82, 51-59.
Zimmerman, B. J. & Bandura, A. (1994). Impact of self-regulatory influences on writing course attainment. American Educational Research Journal, 31, 845-862.
By Linda Garavalia, Marilyn Ray, Tamera Murdoch, University of Missouri-Kansas City, and Margaret Gredler, University of South Carolina
Linda Garavalia in an Assistant Professor of Psychology at the University of Missouri-Kansas City and specializes in self-regulation and assessment. Marilyn Ray recently completed her doctoral work at the University of Missouri-Kansas City and currently is an Instructor/Researcher at the University of Kansas Center for Research on Learning. Tamera Murdock is an Associate Professor of Psychology at the University of Missouri-Kansas City, whose research focuses on academic achievement motivation. Margaret Gredler is a Professor of Educational Psychology and Research at the University of South Carolina.…