The structure and function of academic self-concept in gifted and general education students.

Research suggests that academic self-concept helps to predict academic achievement. However, for academically gifted students, the relationship between academic self-concept and academic achievement may be less clear-cut. Our recent research has revealed several inconsistencies in the relationship between self-concept and achievement for gifted and nongifted students. In a recent study of gifted achievers and gifted underachievers, both gifted achievers and gifted underachievers exhibited high academic self-perceptions, and academic self-perceptions did not efficiently discriminate gifted achievers from gifted underachievers (McCoach & Siegle, in press, a). This finding stands in stark contrast to our previous results for the general population of high school students (McCoach & Siegle, 2001). In that study, we examined the relationship of academic self-perceptions to self-reported GPA with an entire population of high school students. Of all the subscales on the School Attitude Assessment Survey- Revised (SAAS-R; McCoach, 2000) academic self-perceptions most effectively separated students with self-reported high GPAs from those with low self-reported GPAs, and academic self-perceptions best predicted self-reported high school GPA (McCoach & Siegle, 2001).

This paper extends our exploration of academic self-perceptions in gifted and general populations of students by exploring the interplay between academic self-concept and giftedness and the influence that giftedness has on the relationship between academic self-concept and academic achievement. The research questions to be addressed in this paper are: 1) Is the construct of self-concept different for the gifted, and if so, how? 2) Ate there differences in the academic self-concepts of gifted and nongifted students? 3) Is the relationship between academic self-concept and academic achievement the same for gifted students as it is for a general population of students?

Review of the Literature

Students develop confidence in many ways, and those who are confident about their skills are more likely to engage in challenging activities. The perceptions students have about their skills influence the types of activities they select, how much they challenge themselves at those activities, and the persistence they exhibit once they are involved in the activities (Ames, 1990; Bandura, 1977, 1986; Schunk, 1981, 1984).

Academic self-concept involves a description and an evaluation of one's perceived academic abilities (Byrne, 1996; Hattie, 1992) and encompasses global beliefs of self-worth associated with one's perceived academic competence. Academic self-concept is a multidimensional construct that involves both internal and external comparisons. Students compare their own performance with that of their classmates (an external comparison), as well as with their own performance in other areas (an internal comparison) (Byrne; Marsh, 1986). This phenomenon is known as the internal/external frame of reference model of self-concept development. Social comparison theory suggests that when people compare favorably to those around them in a particular domain, they are more likely to maintain high self-concepts in that domain. Academic self-concept is a significant predictor of academic achievement (Lyon, 1993; Marsh, Chessor, Craven, & Roche, 1995; Wigfield & Karpathian, 1991); however, it is negatively related to the average achievement level within a school (Marsh, 1987). This implies that students' academic self-concepts are determined by their perceptions of their ability in an area as well as their assessment of their academic standing relative to their classmates. Therefore, academically gifted students should be able to maintain high academic self-concepts because their academic abilities compare favorably to those of their fellow students. There is some evidence to suggest that academically gifted students do possess higher academic self-concepts than nongifted students (Hoge & Renzulli, 1993; Pyryt & Mendaglio, 1994), especially when they are placed in mixed ability or heterogeneously grouped classes and/or schools (Marsh et al., 1995). In contrast, when students are placed in homogenous classes with other high ability students, their academic self-concept may fall, a phenomenon which is often referred to as the big-fish-little-pond effect (Marsh, 1987; Marsh, Koller & Baumert, 2001).

Research suggests that as much as one third of the variance in achievement can be accounted for by academic self-concept alone (Lyon, 1993). Our own research (McCoach & Siegle, in press, b) suggests that the academic self-perceptions subscale of the SAAS-R accounts for approximately 20% of the variance in students' self-reported GPA. Furthermore, positive self-concept appears to be linearly related to subsequent academic achievement (Marsh et al., 1995). Although a flow of causality for this relationship has not been conclusively established, some evidence (e.g., Marsh, 1990; Marsh & Yeung, 1997) suggests that academic achievement is significantly affected by prior academic self-concept. However, the impact of prior academic self-concept on subsequent academic achievement does not appear to be as strong as the impact of prior academic achievement on subsequent academic self-concept (Marsh & Yeung).

Several researchers within the area of self-concept literature have emphasized the importance of examining the structure of academic self-concept (Byrne, 1996; Hattie, 1992; Mui, Yeung, Low, & Jin, 2000). Previous studies using the SDQ-II examined the internal structure of academic self-concept using Marsh's (1986) internal/external frame of reference model. This model suggests that high achievement in one academic domain should positively influence self-concept in that domain, but it may exert a negative influence on academic self-concept in another academic domain (Marsh, 1986; Plucker & Stocking, 2001). Both studies found that the structure of academic self-concept was invariant across groups of gifted and average students (Mui et al.; Plucker & Stocking, 2001).

In a study of gifted achievers and gifted underachievers (McCoach & Siegle, in press, a), we found that both gifted achievers and gifted under-achievers possess high academic self-perceptions. The results of this study challenged the commonly reported notion that gifted underachievers suffer from low academic self-perceptions. In our study, most gifted students in both the achieving and underachieving groups had high mean scores on the academic self-perceptions scale. For example, over 50% of the gifted underachievers had means of at least 6.0 on the academic self-perceptions scale (1-7), and less than 11% of gifted underachievers had mean scores of less than 5.0. In comparison, almost 62% of gifted high achievers exhibited scores of 6.0 or higher on the academic self-perceptions factor, and less than 4% scored less than 5.0 on the scale. Therefore, although underachieving gifted students were more likely than gifted achievers to report low academic self-perceptions, those with low academic self-perceptions represented a small minority of the gifted underachievers.

Perhaps both gifted underachievers and gifted achievers have high academic self-perceptions because both groups know that they possess the cognitive skills and abilities to be successful in school. In addition, social comparison theory may help to explain why gifted underachievers are able to preserve such a high sense of academic self-concept in the face of academic failure (Marsh, 1987). Although gifted underachievers' achievement may lag behind that of their more successful peers, perhaps gifted underachievers compare their innate abilities with those of their classmates. Consequently, these students feel confident of their academic abilities, in spite of their low academic achievement.

Methodology

Sample

The sample consisted of two separate subsamples, drawn from different studies. The first subsample contained 210 high school gifted students in grades 9 through 12 from 28 school districts across the nation who had scored at or above the 92nd percentile on a standardized measure of aptitude or achievement administered within the last 3 years. Approximately 90% of these students scored at or above the 95th percentile on the standardized measure of aptitude or achievement. The majority of participants in the gifted sample (78%) identified themselves as White. Most of the students in this sample had been identified for gifted services within their school districts at some point during their academic careers. This was a convenient sample of school district volunteers, and it is not necessarily representative of high schools nationwide. A contact person at each of the 28 high schools coordinated the collection of the surveys and student information. The data from this sample was originally collected for use in a study that examined factors that distinguished gifted achievers from gifted underachievers (McCoach & Siegle, in press, a). The second subsample of 160 ninth-grade students came from a general school population of a high school in the Northeast. This high school was located on the urban fringe of a midsized city. Approximately 2,100 students from a variety of socioeconomic backgrounds attend this high school; however, most students in the school are from lower middle class or middle class backgrounds. Approximately 30% of the school population consists of African American students, 22% of the school population consists of Latino American students, 42% of the school population consists of White students, and 6% of the school population consists of Asian or Asian American students.

Instrumentation 

The academic self-perceptions (ASP) subscale used in this study is one of five subscales on the School Attitude Assessment Survey-Revised (SAAS-R) (McCoach, 2000). The SAAS-R instrument employed a 7-point Likert-type agreement scale. The scores on the academic self-perceptions subscale demonstrated a classical theory internal consistency reliability coefficient of .86 in the validation study of the SAAS-R (McCoach & Siegle, in press, b). Self-reported GPA was measured on a 10-point scale where 10= all A's (GPA of 4.0 or above), 9= mostly A's (3.75 to 4.0), to 1= mostly D's and F's.

Analyses

The sample of gifted students scored significantly higher than the general school sample on the ASP factor of the SAAS-R (p<.001). The sample of gifted students scored almost 1 point higher than the general school sample on the academic self-perceptions factor, and the effect size for this difference was greater than 1. In other words, the sample of gifted students scored more than 1 standard deviation unit higher than the general school sample on the academic self-perceptions factor.

Preliminary item-level analyses revealed that there were large differences between the two groups on the means of all the academic self-perceptions items. Gifted students consistently scored higher than students from the general school population, and the effect sizes of these differences were medium to large. Table 1 shows the means, standard deviations, and effect sizes for each of the items on the self-reported GPA variable and the ASP subscale of the SAAS-R.

We conducted a series of analyses to investigate the construct of the ASP factor of the SAAS-R. To determine whether the factor structure of the ASP factor of the SAAS-R is the same for the sample of gifted students and the general school sample, we conducted a multiple groups confirmatory factor analysis (CFA) and structural equation model (SEM) using EQS 5.7 (Bentler & Wu, 1995). As is customary in multiple groups CFA, we compared three different models: a model in which no cross-group constraints were imposed, a fully invariant measurement model, and a partially invariant measurement model (Kline, 1998). In the no-constraints model, the factor loadings for all of the items were allowed to differ as a function of group membership. In the fully invariant model, the factor loadings for each of the items were constrained to be equal across groups and the path from the academic self-perceptions factor to self-reported GPA was constrained to be equal across groups. In the partial invariance model, the factor loadings of items 2 and 37 were left free to vary between the two groups. However, the factor loadings of the other five items were constrained to be equal, and the path from the academic self-perceptions factor to self-reported GPA was constrained to be equal across groups.

We evaluated model fit using several common fit indices including chi-square ([chi square]), the ratio of chi-square to degrees of freedom ([chi square]/df), the Root-Mean Square Error of Approximation (RMSEA), the Comparative Fit Index (CFI), and the Tucker Lewis Index (TLI) (also known as the Bentler-Bonett Non-Normed Fit Index). These fit indices attempt to correct the problems that result from judging the fit of a model solely by examining the chi-square of the model. Most SEM programs provide at least a dozen fit indices in the output. The RMSEA is a function of the degrees of freedom in the model, the [chi square] of the model, and the sample size. The RMSEA has become one of the most popular fit indices because, unlike the [chi square], the value of the RMSEA should not be influenced by the sample size (Raykov & Marcoulides, 2000). The RMSEA approaches zero as the fit of the model nears perfection. The TLI and the CFI measure the proportionate amount of improvement in fit when the specified model is compared with a nested baseline model (Hu & Bentler, 1998). Generally speaking, TLI and CFI values at or above .95 indicate that there is a relatively good fit between the hypothesized model and the data (Hu & Bentler, 1995, 1999). TLI and CFI values below .90 generally indicate that the fit of the model to the data is less than satisfactory.

As suggested by Kline (1998), we also conducted [chi square] difference tests to assess which of the three models seemed to best fit the data. Table 2 reports the fit indices for each of the three models. The [chi square] difference tests revealed that the fully invariant model provided a less satisfactory fit to the data than the partially invariant model ([chi square] difference = 32.59 with 5 df, p<.05) or the no-constraints model ([chi square] difference = 35.68 with 7 df, p<.05). In addition, the partially invariant model and the no-constraints model provided a statistically equivalent fit to the data ([chi square] difference = 3.09 with 5 df, p>.05). Therefore, we concluded that the academic self-perceptions factor demonstrated partial measurement invariance. In other words, two of the factor loadings (V2 and V37) differed as a function of group membership, but the other five factor loadings did not (Kline). Table 3 compares the standardized regression weights of the gifted sample with those of the general school sample for the final, partially invariant measurement model.

Discussion

This exploratory study examined the structure of academic self-perceptions in a gifted and a general school population. Because of the limited and unrepresentative nature of the two samples, all of the results reported in this data should be regarded as preliminary hypotheses regarding differences in the structure of academic self-perceptions for gifted and general populations of students. Nevertheless, several interesting findings emerged.

In this study, the factor structure of the academic self-perceptions factor is partially invariant across the two samples of students. This suggests that the factor structure is somewhat different for the gifted and general school population samples. The two variables that do not exhibit the same factor structure across the two samples are item 2, "I am intelligent" and item 37, "I am capable of getting straight A's." The factor loadings for both of these items are significantly lower for the gifted sample than for the sample of general school population. Both questions demonstrate higher means and lower standard deviations for the gifted sample than for the sample of general secondary school students. In fact, 89% students in the gifted sample agreed or strongly agreed with the statement "I am intelligent," and 88% of the students in the gifted sample agreed or strongly agreed with the statement "I am capable of getting straight A's." Therefore, there is very little variability on these two items for the gifted sample. The lack of variability in the gifted sample attenuates the covariances between these items and the other items within the academic self-perceptions factor, thereby lowering their contribution to the academic self-perceptions factor. However, these items exhibit adequate variability within the general school sample; therefore, these items do help define the academic self-perceptions factor for a heterogeneous group of secondary students. Apparently, most gifted students in this sample feel that they are intelligent and that they are capable of getting straight A's in school. There could be several different reasons for these findings. Perhaps gifted students perceive intelligence and capacity to do well in school as more innate characteristics than the other items on the survey. Another explanation is that gifted students have repeatedly been told that they are intelligent and that they ate capable of getting straight A's in school; therefore, they have internalized these statements more than the other statements on the factor. Future research should further explore the reasons for these differences.

Clearly, the gifted sample and the general school sample exhibit large mean differences on the ASP factor. Based on this study, gifted students appear to be much more confident in their scholastic abilities than the general school population is. However, the relationship between academic self-perceptions and self-reported GPA is moderate and positive. In addition, for both groups, higher scores on the academic self-perceptions factor correspond to higher self-reported GPA, and the relationship between academic self-perception and GPA is similar across the two groups of students.

Limitations

This study suffers from several limitations. First, the use of self-reported GPA as a measure of academic achievement is less accurate than the use of actual GPA. In addition, self-reported GPA was measured on a 10-point ordinal scale, rather than a continuous interval scale. Future research should compare the results of the SAAS-R to actual achievement data (such as GPA). Furthermore, the definition of giftedness was restrictive and relied upon standardized aptitude or achievement test scores. Using standardized test scores as the sole criterion by which we identified students as "gifted" is restrictive and not necessarily indicative of many school districts' identification criteria.

In addition, the two samples of students were convenient samples and are not representative of the general population of high school students and the entire spectrum of gifted students. The sample of general education students contained more students from ethnically diverse backgrounds than the sample of gifted students did, and it is possible that some of the observed differences may reflect differences in socioeconomic background or cultural differences. Furthermore, because the general school sample consisted of ninth graders and the gifted sample contained ninth through twelfth grade students, grade level could have confounded the results of the study. The correlation between grade level and academic self perceptions was insignificant (r=.02), however, the correlation between grade level and GPA was small (r=.12). Therefore, these grade level inconsistencies could have some effect on the specified model. Additionally, there are likely to be gifted students within the sample of the general population of high school students. Therefore, these two samples do not represent two completely mutually exclusive groups. Future research should examine differences between gifted students and nongifted students. Finally, the results of this study were strictly correlational in nature; therefore, one cannot infer causality from these results.

Conclusion

This study investigated the differences in the factor structure of academic self-perceptions in a sample of gifted students and within a general population of students using confirmatory factor analytic and structural equation modeling techniques. The factor structure exhibited partial measurement invariance, suggesting that two of the items on the factor work differently in the gifted sample than in the sample from the general population. The gifted sample demonstrated higher academic self-perceptions than the general school sample and higher self-reported GPAs than the general school sample. However, the relationship between academic self-perceptions and self-reported GPA was invariant across groups. This study suggests that for both gifted students and the general population of high school students, GPA and academic Self-perception are moderately and positively related. Therefore, although there are large mean differences between gifted students and the general population of students on the academic self-perceptions scale, the structure of the relationship between academic self-perception and GPA was similar in this sample. It remains to be seen whether increasing students' academic self-perceptions results in achievement gains for low achieving students. Future longitudinal research should explore whether programs that seek to increase academic self-perceptions in underachieving students can effectively increase their academic achievement.

Conferences

Illinois Association for Gifted Children
February 2-4, 2003
Chicago, Illinois
847/963-1892
http://www.iagcgifted.org

Nebraska Association for the Gifted Annual Conference
February 20-21, 2003
Lincoln, Nebraska
email: [e-mail]
http://www.nebraskagifted.org/conference.html

Beyond Giftedness X
February 28, 2003
Arvada Center for the Arts, Colorado
303/444-7020 or 800/494-6178
email: [e-mail]
http://www.openspacecomm.com/conference.htm

New Jersey Association for Gifted Children
February 28-March 1, 2003
Princeton, New Jersey
856/273-7530
http://www.njagc.org/Conference/conference.htm

National Curriculum Network Conference
March 6-8, 2003
Center for Gifted Education
at the College of William and Mary
Williamsburg, Virginia
757/221-2362, [e-mail]
www.cfge.wm.edu

California Association for the Gifted
March 7-9, 2003
Santa Clara, California
562/789-9933
email: [e-mail]
http://www.cagifted.org/

North Carolina Association for the Gifted and Talented
March 13-15, 2003
Winston-Salem, NC
http://ncagt.org/conference/index.shtml

Pennsylvania Association for Gifted Education
Valley Forge, Pennsylvania
March 28-29, 2003
215/616-0470
http://www.penngifted.org

Association for the Education of Gifted Underachieving
Students
March 21-22, 2003
Baltimore/Washington area
http://www.aegus.org/

15th Biennial Conference of the World Council
for Gifted and Talented Children
August 1-5, 2003
Adelaide, South Australia
http://www.worldgifted.org/xconf.htm
U.S. contact: 818/368-7501
email: [e-mail]
Australia contact: 61-8-8201 3425
email: [e-mail]

Table 1
Descriptive Statistics for Each of the Variables Included
in the Multiple Group Confirmatory Factor Analysis

                                    Gifted      General Population
                                    n=203             n=149

Items                               M      SD     M      SD     d

Self-reported GPA (10 = all A's,
  1 = mostly D's & F's)            8.18   2.35   5.06   2.11   1.40
Academic self-perceptions mean     6.01   .777   5.05   1.10   1.02
  I am capable of getting
    straight A's.                  6.43   1.11   5.11   1.44   1.04
  I can grasp complex
    concepts in school.            6.00   1.03   4.74   1.44   1.02
  I can learn new ideas
    quickly in school.             6.07   .959   5.19   1.29    .78
  I am smart in school.            6.02   1.09   5.06   1.53    .75
  I am good at learning new
    things in school.              6.01   1.02   5.18   1.32    .71
  I am intelligent.                6.23   .673   5.46   1.57    .69
  School is easy for me.           5.29   1.39   4.62   1.54    .46

Note: ASP subscale 7 = strongly agree, 1 = strongly disagree

Table 2
Tests of Invariance of Academic Self-Perceptions (ASP)
Factor and Academic Achievement/ASP Path

Model             Chi-square   df   TLI   CFI   RMSEA   Description

Null                1104.75    5    --    --     --     Null Model
                               6

No constraints       120.29    4    .89   .92    .078   No constraints
                               0

Fully Invariant      155.97    4    .88   .90    .084   All constrained
Model                          7                        (All 7
                                                        constraints
                                                        imposed)

Final Model:         123.38    4    .91   .93    .073   5 constraints;
                                                        V2 and V37 free
Partial                        5
Invariance

Table 3
Comparison of the Standardized Regression Weights for
Each Item in the Gifted and General School Samples

                              Standardized   Standardized   Constraint
                                Regression     Regression      Imposed
                                  Weight         Weight
                                 (Gifted)       (General)

Item Stems: Academic
  Self-Perceptions

 2. I am intelligent.              .47            .68       No, p<.05
 3. I can learn new ideas
  quickly in school.               .82            .71          Yes
13. School is easy for me.         .56            .61          Yes
20. I can grasp complex
  concepts in school.              .84            .69          Yes
37. I am capable of getting
  straight A's.                    .56            .67       No, p<.05
40. I am good at learning
  new concepts in school.          .84            .79       Marker
                                                            variable
41. I am smart in school.          .75            .69          Yes
ASP Factor [right arrow]
  self-reported GPA                .24            .33          Yes

Manuscript submitted April, 2002.

Revision accepted July, 2002.

REFERENCES

Ames, C. A. (1990). Motivation: What teachers need to know. Teachers College Record, 91, 409-421.

Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84, 191-215.

Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall.

Bentler, P., & Wu, E. J. C. (1995). EQS 5 user's guide [computer software]. Encino, CA: Multivariate Software.

Byrne, B. M. (1996). Measuring self-concept across the lifespan: Issues and instrumentation. Washington, DC: American Psychological Association.

Hattie, J. (1992). Self-concept. Hillsdale, N J: Lawrence Erlbaum.

Hoge, R. D., & Renzulli, J. S. (1993). Exploring the link between giftedness and self-concept. Review of Educational Research, 63, 449-465.

Hu, L., & Bentler, P. M. (1995). Evaluating model fit. In R. H. Hoyle (Ed.), Structural equation modeling: Concepts, issues and applications (pp. 76-99). Thousand Oaks, CA: Sage.

Hu, L., & Bentler, P. M. (1998). Fit indices in covariance structure modeling: Sensitivity to unparameterized model misspecification. Psychological Methods, 3, 424-453.

Hu, L., & Bentler, P. M. (1999). Cut-off criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1-55.

Kline, R. B. (1998). Principles and practices of structural equation modeling. New York: Guilford Press.

Lyon, M. A. (1993). Academic self-concept and its relationship to achievement in a sample of junior high school students. Educational and Psychological Measurement, 53, 201-211.

Marsh, H. W. (1986). Verbal and math self-concepts: An internal/external frame of reference model. American Educational Research Journal, 23, 129-149.

Marsh, H. W. (1987). The big-fish-little-pond effect on academic self-concept. Journal of Educational Psychology, 79, 280-295.

Marsh, H. W. (1990). Causal ordering of academic self-concept and academic achievement: A multiwave longitudinal panel analysis. Journal of Educational Psychology, 82, 646-656.

Marsh, H. W., Chessor, D., Craven, R., & Roche, L. (1995). The effects of gifted and talented programs on academic self-concept: The big fish strikes again. American Educational Research Journal, 32, 285-319.

Marsh, H. W., Koller, O., & Baumert, J. (2001). Reunification of East and West German school systems: Longitudinal multilevel modeling study of the Big-Fish-Little-Pond Effect on academic self-concept. American Educational Research Journal, 28, 321-350.

Marsh, H. W., & Yeung, A. S. (1997). Causal effects of academic self-concept on academic achievement: Structural equation models of longitudinal data. Journal of Educational Psychology, 89, 41-54.

McCoach, D. B. (2000). The School Attitude Assessment Survey- Revised. Unpublished instrument.

McCoach, D. B., & Siegle, D. (2001). A comparison of high achievers' and low achievers' attitudes, perceptions, and motivations. Academic Exchange Quarterly, 5(2), 71-76.

McCoach, D. B., & Siegle, D. (in press, a). Factors that differentiate underachieving gifted students from high achieving gifted students. Gifted Child Quarterly.

McCoach, D. B., & Siegle, D. (in press, b). The SAAS-R: A new instrument to identify academically able students who underachieve. Educational and Psychological Measurement.

Mui, F. L. L., Yeung, A. S., Low, R., & Jin, P. (2000). Academic self-concept of talented students: Factor structure and applicability of the internal/external frame of reference model. Journal for the Education of the Gifted, 23, 343-367.

Plucker, J. A., & Stocking, V. B. (2001). Looking outside and inside: Self-concept development of gifted adolescents. Exceptional Children, 67, 535-548.

Pyryt, M. C., & Mendaglio, S. (1994). The multidimensional self-concept: A comparison of gifted and average ability adolescents. Journal for the Education of the Gifted, 17, 299-305.

Raykov, T., & Marcoulides, G. A. (2000). A first course in structural equation modeling. Mahwah, N J: Lawrence Erlbaum.

Schunk, D. H. (1981). Modeling and attributional effects on children's achievement: A self-efficacy analysis. Journal of Educational Psychology, 73, 93-105.

Schunk, D. H. (1984). Self-efficacy perspective on achievement behavior. Educational Psychologist, 19, 48-58.

Wigfield, A., & Karpathian, M. (1991). Who am I and what can I do? Children's self-concepts and motivation in achievement situations. Educational Psychologist, 26, 233-262.

D. Betsy McCoach is completing her Ph.D. in educational psychology with concentrations in gifted education, school psychology, and quantitative research methodology at the University of Connecticut's Neag School of Education. Previously, she was a teacher of the gifted and a secondary gifted specialist in Pennsylvania. She was a recipient of a Spencer Doctoral Fellowship from the American Educational Research Association for the 2001-2002 academic year. In addition, she received an AERA Dissertation Research Grant for the 2002-2003 academic year. Ms. McCoach's areas of research interest include the underachievement of gifted students, instrument design and analysis, quantitative research methodology, and issues related to the identification and assessment of high ability students

Del Siegle is an assistant professor of educational psychology in the Neag School of Education at the University of Connecticut. Del is currently on the Board of Directors of the National Association of Gifted Children and The Association for the Gifted. He serves as the webmaster for the CEC-TAG and AERA-Gifted SIG's websites. Del is the founder of Edufest, a summer gifted education institute, currently housed at Boise State University. Dr Siegle's research interests include web-based instruction, models of achievement orientation, the underachievement of gifted students, self-efficacy theory, and teacher bias in the identification of students for gifted programs.