Effect of Social Setting, Self-Concept, and Relative Age on the Social Status of Moderately and Highly Gifted Students.

An ongoing debate continues regarding the relationship between giftedness and social adjustment. On one side are researchers who believe that students who are highly gifted tend not to be popular or well accepted by peers. This side of the debate dates back to 1893 in a work by Lombroso (1893) who reported that gifted children were unpopular. Likewise, Hollingworth (1942) stated that highly gifted children, with IQs above 180, had difficulties with educational adjustments as well as social adjustments, and that these bright children were likely to have conduct problems. On the other hand, in another early study of children with IQs of 140 and above, Terman (1925, 1959) reported that these children were better adjusted than less gifted children.

More recent research finds the relationship between ability and social adjustment still unclear. Studies comparing self-concept and self-esteem of intellectually gifted and non-gifted students generally report gifted students to be at least as well adjusted as regular students (Coleman & Fults, 1983, 1985; Janos & Robinson, 1985; Karnes & Wherry, 1981). Yet, adjustment problems for some gifted students have been documented in autobiographies, case studies, and research studies of highly gifted or talented individuals (Gallagher, 1958; Gross, 1993; Maas, 1992; Montour, 1976, 1977; Tannenbaum, 1983).

In an earlier study, Norman, Ramsay, Martray, and Roberts (1999) argued that mixed findings regarding giftedness might be related to the failure to distinguish between moderately and highly gifted students. A review of current literature suggested that many highly gifted students (those at least two standard deviations above the mean on an IQ or standardized achievement test) may be likely to suffer from some psychological and social adjustment problems (Dauber & Benbow, 1990; Gross, 1993; Janos, Fung, & Robinson, 1985; Janos & Robinson, 1985; Tannenbaum, 1983). Yet other studies of highly gifted students found no difficulties in social or psychological adjustment (Grossberg & Cornell, 1988; Haier & Denham, 1976; Richardson & Benbow, 1990). In contrast to the mixed findings regarding highly gifted students, studies of moderately gifted students generally have reported positive psychological and social adjustment (Cornell, Delcourt, Bland, Goldberg, & Oram, 1994; Holliday, Koller, & Kunce, 1996; Karnes & Wherry, 1981; Lehman & Erdwins, 1981; Olszewski-Kubilius, Kulieke, & Krasney, 1988; Sayler & Brookshire, 1996;). Only a handful of studies have suggested that some moderately gifted students may have adjustment problems (Coleman & Cross, 1988; Cornell, 1990).

To investigate potential differences between moderately and highly gifted students, our earlier study (Norman et al, 1999) divided students attending summer programs for the gifted into moderately and highly gifted groups based on scores on a school ability test. These students were administered a battery of self-concept and adjustment measures. Contrary to our hypothesis that differences would emerge between these groups, no significant differences were found on any of the measures. However, a major weakness of this study, as well as most studies of the psychological and social adjustment of the gifted, was the reliance upon self-report measures.

The current study investigates potential differences in the social status and social adjustment of highly and moderately gifted students as measured by peer ratings and peer nominations. In their study using peer ratings to create five peer status groups--popular, rejected, neglected, controversial, and average--Coie, Dodge, and Kupersmidt (1990) found that popular students were also more likely to be nominated as cooperative and good leaders than other children. Popular students also were less likely to be nominated for starting fights and being disruptive. On the other hand, children classified as rejected were nominated as starting fights and being disruptive. Other studies report similar findings (Carlson, Lahey, & Neeper, 1984; Coie & Dodge, 1983; Lemerise, 1997; Rogosch & Newcomb, 1989).

Studies of the social status of gifted students generally find favorable results (Austin & Draper, 1981; Cohen, Duncan, & Cohen, 1994; Luftig & Nichols, 1990; Schneider et al, 1989). Gifted students usually enjoy high social status, especially in the elementary years. However, none of these studies differentiate between moderately and highly gifted students. Thus, the question remains open that perhaps a point of diminishing returns may exist related to giftedness and social acceptance.

Furthermore, following the lead of Cornell et al (1990), this study investigates the consistency of social status across academic (the classroom) and non-academic (the dormitory) settings, as well as the relationship of social status to several facets of self-concept. Cornell et al found social status to be consistent across settings. They also found social status to be related to varied types of self-concept. For boys, social status was most strongly related to social, athletic and physical appearance self-concept. For girls, social self-concept was the major factor related to social status.

Finally, because the gifted groups for this study were in mixed-age settings, the relationship between relative age and social status is examined. Findings from several studies provide evidence of differences between mixed-age and same-age group interactions (Brody, Graziano, & Musser, 1983; French, Waas, Stright, & Baker, 1986; Graziano et al, 1976). Furthermore, studies have consistently found that children typically prefer older classmates and that younger children are less popular (Ahlbrand & Reynolds, 1972; Allen, 1989; Lemerise, 1997; Lemerise, Harper, & Howes, 1998).

Thus, the goals of the present study were to examine differences in social status between moderately and highly gifted students, examine potential differences in social status across classroom and dorm settings, examine the relationships between social status, peer nominations, and self-report measures of self-concept, and, examine the relationship between relative age and social status.

Method

Subjects

Subjects consisted of attendees of a summer program for gifted students held at a university in south central United States. Students attending the program were part of the Duke Talent Identification Program in which participants typically take the ACT or SAT during their seventh grade school year. Qualifying ACT scores for math are 18 or above out of a possible 36, and for English, 25 out of 36; qualifying math or verbal SAT scores are 500 or above. Of the approximately 175 students, 154 (88%) completed sufficient data to participate in the study.

The average age of participants was 14.61 and the range was 12 to 16 years of age. Ninety-five were males, 59 were females. One hundred and twenty (78%) participants classified themselves as white, 18 (12%) as Asian, 5 (3%) as black, 3 (2%) as Hispanic, and 8 (3%) did not classify themselves in any of the above racial groups.

Participants' socioeconomic status was scored using Hollingshead's (1975) standard scales for parental education and occupation. Scores for both education and occupation can range from 1 to 9 and were inverted so that higher scores relate to higher education and occupational status. The mean educational level was 6.1 for fathers and 5.9 for mothers. A rating of 5 indicates some college training and 6 indicates a college degree. The mean occupational level was 7.5 for fathers and 6.9 for mothers. A rating of 6 refers to occupations such as technicians and secretaries, and a rating of 7 includes occupations such as managers and computer programers.

To differentiate between moderately and highly gifted students, the Otis-Lennon School Ability Test, Sixth Edition, Forms F and G was administered to all attendees. According to the OLSAT's Multilevel Norms Booklet (Otis & Lennon, 1988), only 2% of the national student population obtains School Ability Index scores above 132 (over two standard deviations above the mean). Those students in this study who scored above 132 were designated the highly gifted group (N=42), all others were designated the moderately gifted group (N=109). Three students failed to complete the OLSAT but are included in analyses not related to gifted classification. A t-test analysis revealed the highly gifted group (Mean OLSAT score=139.31, SD=6.31) scored significantly higher than the moderately gifted group (Mean OLSAT score=122.25, SD=7.99), t(149)= 12.42, p [is less than] .000).

Instruments

Demographics. Data such as sex, age, race, socioeconomic status, and family structure (parent marital status, siblings, etc.) were collected via parent and student questionnaires.

Academic ability. The Otis-Lennon School Ability Test, Sixth Edition, Forms F and G (OLSAT), was administered to all students. This test yields three scores: the School Ability Index Nonverbal (SAI Nonverbal), a measure of pictorial, figural, and mathematical ability, the School Ability Index Verbal (SAI Verbal), and an overall score based on the average of the first two, the School Ability Index (SAI). These are normalized standard scores with a mean of 100 and a standard deviation of 16. Internal consistency reliability estimates on all three scores are in the high .80's and .90's range. Scores' content, criterion-related, and construct related evidences of validity are also high (Anastasi, 1992; Swerdlik, 1992).

Social status. Social status was determined by the use of peer ratings and peer nominations from each student's classmates (a mixed sex group) and dorm group (a same sex group). First, students were asked to rate, using a Likert type scale of 1 "like least" to 5 "like most", how much they liked to spend time with each student in their class or group. Second, students were asked to nominate other class or group members for the following descriptions: "Students you enjoy being with most," "Students who are sometimes disruptive," "Students who act shy or are hard to get to know," and "Students who are easy going, helpful, and cooperative." Standardized scores for each nomination category were calculated to create like most (LM), fights, shy, and gets along scores. Because the director of the summer program was uncomfortable with a negative nomination question (i.e., "Students you enjoy being with the least"), social status categories were derived using the method suggested by Asher and Dodge (1986). With this method, the number of like least ratings (1) from the peer ratings is substituted for the number of negative nominations to create a standardized like least (LL) score. Leading researchers in the area of sociometrics have found this substitution useful (Asher & Dodge, 1986; Coie et al, 1982).

The standardized LM and LL scores were used to define social status. Social preference (SP) was determined by subtracting LL from LM. Social impact (SI) was calculated by adding LL and LM. Students were classified as rejected if SP [is less than or equal to] -1.00, LL [is greater than] 0, and LM [is less than] 0. Popular status was defined as SP [is greater than or equal to] + 1.00, LM [is greater than] 0, and LL [is less than] 0. Participants were classified as neglected if SI [is less than or equal to] -1.00, and absolute LM=0. Controversial status was defined as SI [is greater than or equal to] +1.00, LL [is greater than] 0, and LM [is greater than] 0. Average status was defined as SP [is greater than or equal to] -0.5 and [is less than or equal to] +0.5. Those participants who did not meet the above criteria were deemed unclassifiable.

Relative age. To obtain a measure of a student's age relative to his/her dorm or class group, each participant's exact age calculated in years, months, and days was standardized to create z-scores within groups. The following age groups were then created: "younger", those with a relative age [is less than or equal to] -0.5; "intermediate", those with a relative age [is greater than] -0.5 and [is less than] +0.5; and "older", those with a relative age [is greater than or equal to] +0.5 (Lemerise, 1997; Lemerise, Harper, & Howes, 1998). A [chi square] test of independence revealed no significant relationship between relative age and sex in the mixed-sex classroom setting.

Self-concept. Two measures of self-concept were collected, Harter's (1995) Self-Perception Profile for Children (SPPC) and Marsh's (1990) Self-Description Questionnaire-II (SDQ-II). A four-point Likert scale, the SPPC contains six measures of self-concept: physical appearance, athletic competence, behavioral conduct, scholastic competence, social acceptance, and global self-worth. The validity of the SPPC is well established. Although developed to use with children through eighth grade, it has been successfully used with children through age sixteen (Cornell et al, 1990). Furthermore, concerning reliability, Cornell et al found internal consistency correlations for all scales to be .75 and above.

A six-point Likert scale, the SDQ-II provides several measures of academic and non-academic self-concept. This measure was specifically included because of its breakdown of academic self-concept into three areas: math, verbal, and general-school. Non-academic scales include physical appearance, emotional stability, honesty-trustworthiness, opposite-sex relations, physical abilities, same-sex relations, parent relations, and general-self. The parent relations scale was not used in this study. Ample research establishing the validity and reliability of the SDQ-II can be found in Marsh (1990) and others (Marsh, Parker, & Barnes, 1985; Marsh & Peart, 1988). Coefficient alphas for the scales range from .83 to .94; test-retest reliability coefficients range from .72 to .88. Factor analysis for construct validity revealed target loadings ranging from .48 to .80, nontarget loadings from -.12 to .27, and correlations among factors ranging from -.03 to .39.

Procedures

As part of the application to attend the summer program, parents were asked to complete a demographic questionnaire and a consent form granting permission for further testing of their child. All testing took place during two evening sessions about midway through the summer program specifically scheduled for that purpose.

Results

Before examining relationships among variables, social status in both class and dorm settings was obtained. In the class setting, of the 154 students, 49 were popular, 33 were rejected, 38 were average, 8 were neglected, 1 was controversial, and 25 were unclassifiable. In the dorm setting, 44 were popular, 32 were rejected, 39 were average, 8 were neglected, 1 was controversial, and 30 were unclassifiable. Because of the small numbers representing the neglected and controversial groups and the lack of overlap across settings, these groups were dropped from further analyses.

Regarding demographic variables, social status in either setting was not significantly related to parent education, parent occupation, or participants' race (white versus minority). Although previous research has warned of possible sex bias in peer ratings (Asher & Hymel, 1981; Cornell et al, 1990), statistical analysis revealed none in the mixed sex classroom setting.

To ascertain differences in social status between moderately and highly gifted students, the [chi square] statistic was used. No significant differences in social status emerged in either the dorm or classroom setting. Furthermore, classroom peer nominations for fights, gets along, and shy and dorm peer nominations for fights and gets along were not significantly different between moderately and highly gifted students. Only on the dorm shy nomination were highly gifted students (Mean z score=.32, SD=I. 11) significantly more nominated than moderately gifted students (Mean z score=-1.32, SD=.88), t(67.38)=2.32, p [is less than] .05.

Stability of social status across settings (popular, rejected, average, and unclassifiable) was ascertained through [chi square] and K analyses (see Table 1). Social status was significantly stable across settings ([chi square] =54.52, p [is less than] .0001, K=.27, p [is less than] .0001). Of the various statuses, popular and rejected were the most stable with 28 (49%) of the 57 popular students being so designated across both settings and with 17 (40%) of the 43 rejected students being so designated across settings (see Table 1). Analysis of the likelihood of a popular or rejected person in one setting being in the opposite category in the other revealed these two statuses to be significantly stable ([Kappa] =.87, p [is less than] .0001). Only 3 students were rejected in one setting and popular in the other.

Table 1. Cross Comparisons of Dorm and Classroom Social Status

                               Dorm Social Status
Classroom
Social Status         Popular         Rejected     Average

Popular                 28                1           12
Rejected                 2               17            5
Average                  7                6           11
Unclassifiable           4                6            8
Total                   41               30           36

                 Dorm Social Status
Classroom
Social Status      Unclassifiable      TOTAL

Popular                  5               46
Rejected                 6               30
Average                 12               36
Unclassifiable           6               24
Total                   29              136

Further analyses focused on differences between popular and rejected students. Because of the stability of these two groups and in order to be conservative in our analysis, two groups of students were created, stably popular and stably rejected (Lemerise, Harper, & Howes, 1998). As the group names suggest, students were so designated only if they received the same social status across both settings. Twenty-eight students were stably popular, 17 were stably rejected.

A repeated measures design was used to analyze peer nomination differences between stably popular and stably rejected students because students could be nominated for the gets along, fights, and shy categories in the dorm and/or classroom settings. This allowed for the exploration of the stability of these nominations and any possible interactions. For the gets along nomination, stably popular students were significantly more likely to be nominated than stably rejected students, F(1,43)=64.20, p [is less than] .001. A significant interaction (p [is less than] .05) was observed with differences in gets along scores between popular and rejected students being more extreme in the classroom setting than in the dorm settings. For the fights and shy nominations, stably rejected students were significantly more likely to be nominated than the stably popular students, fights F(1,43)=22.90, p [is less than] .001 and shy F(1,43)=5.21, p [is less than] .05. No significant interactions were observed.

Because of the small numbers of students in these two groups and the large number of scales on the SPPC and SDQ-II self-concept measures, principal components analysis was used to reduce the number of contrasts needed to explore the relationship between social status and self-concept. The scores of all summer program students on the SPPC and the SDQ-II's scales (excluding parent relations) were used to create factors. An initial scree plot analysis (Catell, 1966) suggested three factors, with the third factor marginally passing. Furthermore, the rotated factor matrix showed multiple loadings of several scales on the third factor and one of the other factors. Thus, a two factor solution seemed to offer a better simple structure.

As Table 2 reveals, the SPPC and SDQ-II scales that measure physical appearance, athletic ability, and social adeptness loaded highest on the first factor. Scales related to good behavior and school ability loaded highest on the second factor. It should be noted that several scales loaded equally high on both factors, the SPPC global self worth scale and the SDQ-II emotional stability and general self scales. Because these scales emphasize general self-concept whereas our interest was in types of self-concept and because our goal was to maximize independence of these two factors, a second two factor solution was derived without these scales. As Table 2 reveals, the same scales as before load highly on the two factors. The first factor was deemed social self-concept, the second scholastic self-concept.

Table 2. Varimax Rotated Loadings of SPPC and SDQ-II Scales on Factors 1 and 2 With All Scales Entered (First Solution) and With Double Loading Scales at Time 1 Removed (Second Solution)

                          First Solution       Second Solution

Self-Concept Scales    Factor 1   Factor 2   Factor 1   Factor 2

SPPC
 Physical Appearance    .63449     .28919     .63328     .27654
 Athletic Competence    .70818     .04551     .70731     .04492
 Behavioral Conduct    -.06983     .82989     .07121     .82136
 Global
  Self-Worth(*)         .50343     .68195
 Scholastic
  Competence            .25230     .70208     .27934     .72925
 Social Acceptance      .59942     .35880     .62706     .37370
SDQ-II
 Physical Appearance    .82118     .11005     .81769     .11206
 Emotional
  Stability(*)          .41989     .35933
 General Self(*)        .58217     .50951
 Honesty-
  Trustworthiness       .06221     .79075     .06819     .79017
 Math                   .09444     .52993     .09216     .51989
 Opposite-Sex
  Relations             .84696     .03071     .85144     .02920
 Physical Abilities     .74986     .12600     .73895     .12174
 General School         .22964     .83350     .23474     .82964
 Same-Sex Relations     .77212     .18964     .77361     .18132
 Verbal                 .20938     .47930     .23236     .51158

(*) Scales loading equally on both factors on First Solution. Bold = Scales loading high(>.50) on each factor on Second Solution.

To compare stably popular and stably rejected students on levels of social and scholastic self-concept, standardized composite scores were formed by adding together each student's z-scores on only those scales that loaded high on each factor and then by dividing by the number of scales. Thus the social self-concept score was formed by adding together z-scores on the SPPC's physical appearance, athletic competence, and social acceptance scales, and the SDQ-II's physical appearance, opposite-sex relations, physical abilities, and same-sex relations scales and then dividing by 7. The scholastic self-concept score was formed by adding z-scores on the SPPC's behavioral conduct and scholastic competence scales and the SDQ-II's honesty-trustworthiness, math, general school, and verbal scales and dividing by 6. Inter-item reliability analyses of these two new scales based on all summer program students' scores revealed estimated coefficient alphas of .88 for social self-concept and .82 for scholastic self-concept.

T-test analyses comparing stably popular and stably rejected students on these scores revealed significant differences. Popular students (Mean z score=. 17, SD=.54) scored significantly higher on the new social self-concept measure than rejected students (Mean z score=-.39, SD=.76), t(38)=2.70, p [is less than] .05. Likewise, popular students (Mean z score=.18, SD=.46) scored significantly higher on the new scholastic self-concept measure than rejected students (Mean z score=-.41, SD=.95), t(40)=2.68, p [is less than] .05.

A repeated measures design also was used to examine the relationship between social status and relative age because a student's relative age z-score could vary across the dorm and classroom settings. Stably popular students were significantly more likely to be older than stably rejected students, F(43,1)=4.63, p [is less than] .05. No significant interaction in relative age across settings was observed.

To further explore the effect of relative age on social status, the standardized peer nomination scores (fights, shy, gets along), social impact scores, and social preference scores of students categorized as older across the dorm and class settings (stably older) were compared to the scores of students categorized as younger across both settings (stably younger). The stably older group consisted of 46 students, whereas the stably younger group consisted of 37 students. Statistical analyses ([chi square]) revealed no significant association between these relative age groups and sex nor between relative age groups and gifted status.

Because these stable age groups received standardized social status scores in both settings, a repeated measures design was again used. For the fights nomination, no difference was found between the groups. For the shy nomination, stably younger students were more likely to be named than stably older students, F(81,1)=6.36, p [is less than] .05, with no interaction observed. For the gets along nomination, stably older students were more likely to be named than stably younger students, F(81,1)=7.54, p [is less than] .01, with no interaction observed. On the social impact score, created by adding together standardized like most (LM) and like least (LL) scores, no difference was found. Last, on the social preference score, determined by subtracting LM scores from LL scores, stably older students were significantly more preferred than stably younger students, F(81,1)=54.35, p [is less than] .001, with no significant interaction present.

Discussion

The finding of no difference in social status between moderately and highly gifted students provides additional support for our earlier conclusions that few personality and social adjustment differences exist between moderately and highly gifted students (Norman et al, in press). It also reinforces in general those studies that have found gifted students as a whole to be no more or less adjusted than other students (Coleman & Fults, 1983, 1985; Janos & Robinson, 1985; Karnes & Wherry, 1981) and in particular those studies that have found highly gifted students to be normally adjusted (Grossberg & Cornell, 1988; Richardson & Benbow, 1990). Furthermore, it lends credence to the work of Janos, Fung, and Robinson (1985) and Cornell (1990) who report that factors unrelated to giftedness often contribute to adjustment problems.

The finding of significant stability in social status, especially in popular and rejected status, across dorm and classroom settings is consistent with other studies of gifted students (Cornell et al, 1990). Likewise, it supports the work of Lemerise and others with kindergarten and ungraded primary students (Lemerise, Harper, & Howes, 1998; Lemerise, Harper, Caverly, & Howes, 1997). They found social status to be longitudinally stable.

The finding of a significant relationship between social status and self-concept is in partial agreement with Cornell' s (1990) study of unpopular gifted students. He found unpopular students to have poorer social self-concepts than popular students, but did not find lower academic self-concepts among unpopular students. Interestingly enough, however, when Cornell asked teachers to rate the academic self-esteem of unpopular students, they rated these students significantly lower than popular students.

Perhaps the most important finding of our study is the significant relationship between relative age and social status. Although studies of mixed age relationships are numerous (e.g., Ahbrand & Reynolds, 1972; Allen, 1989; Lemerise, 1997; Lemerise, Harper, & Howes, 1998), most have focused on children under high school age, and few have focused on gifted students (e.g., Cornell et al, 1990). In their study, Cornell et al did not find age to be correlated to peer status. However, they did not take into account the relative age of students within class and dorm groups. Thus, our finding provides additional support for the hypothesis that relative age among a group of peers is a key factor in social status.

Although our findings seem promising, there are reasons to interpret them with caution. First, this study looked at social status of highly and moderately gifted students within a gifted group. It is still possible that highly gifted students are not as well accepted by more average students, as some research suggests (e.g., Gross, 1993; Maas, 1992). Second, although we did find significant differences in self-concept between stably popular and stably rejected students, a comparison of the raw scores on the SDQ scales among these two groups and Marsh's (1990) norm group reveal both groups to be scoring at or above average on nearly all scales. Thus, even stably rejected gifted students do not suffer from below average self-esteem. Finally, regarding generalizability, the sample from this study was affected by self-selection. Because this summer gifted program is away from home for all students and involves extensive interaction with others, it is possible that gifted students who are independent and socially confident would be more likely to attend, whereas gifted students experiencing difficulties in these areas would choose not to attend. The high SDQ self-concept scores seem to bear this out.

In conclusion, our findings suggest that one's level of giftedness does not play a major role in social status, at least among other gifted students. However, other factors do seem related to one's position among peers, especially when it comes to being popular or rejected. For instance, one's social reputation as being either friendly, disruptive, or shy affects peer status. Likewise, self-concept seems related to social standing. Furthermore, relative age is a big contributor to social position.

REFERENCES

Ahlbrand, W. P., & Reynolds. J. A. (1972). Some social effects of cross-age grouping. Elementary School Journal, 72, 327-332.

Allen, J. P. (1989). Social impact of age mixing and age segregation in school: A context-sensitive investigation. Journal of Educational Psychology, 81, 408-416.

Anastasi, A. (1992). Review of the Otis-Lennon School Ability Test, Sixth Edition [CD-ROM]. 11 Mental Measurements Yearbook.

Asher, S. R., & Dodge, K. A. (1986). Identifying children who are rejected by their peers. Developmental Psychology, 22, 444-449.

Asher, S. R. & Hymel, S. (1981). Children's social competence in peer relations: Sociometric and behavioral assessment. In J. D. Wine & M. D. Smye (Eds.), Social Competence (pp. 125-157), New York: Guilford.

Austin, A. M. B., & Draper, D. C. (1981). Peer relationships of the academically gifted: A review. Gifted Child Quarterly, 25, 129-133.

Austin, A. M. B., & Draper, D. C. (1984). The relationship among peer acceptance, social impact, and academic achievement in middle childhood. American Educational Research Journal, 21(3), 597-604.

Brody, G. H., Graziano, W. G., & Musser, L. M. (1983). Familiarity and children's behavior in same-age and mixed-age peer groups. Developmental Psychology, 19, 568-576.

Carlson, C.L., Lahey, B.B., & Neeper, R. (1984). Peer assessment of the social behavior of accepted, rejected, and neglected children. Journal of Abnormal Child Psychology, 12, 189-198.

Catell, R. B. (1966). The scree test for the number of factors. Multivariate Behavioral Research, 1, 245-276.

Cohen, R., Duncan, M., & Cohen, S. (1994). Classroom peer relations of children participating in a pull-out enrichment program. Gifted Child Quarterly, 38(1), 33-37.

Coie, J., Dodge, K., & Kupersmidt, J. B. (1990). Peer group behavior and social status. In S. R. Asher & J. D. Coie (Eds.), Peer Rejection in Childhood (pp. 17-59). New York, NY: Cambridge University Press.

Coie, J. D. & Dodge, K. A. (1983). Continuities and changes in children's social status: A five year longitudinal study. Merrill-Palmer Quarterly, 29, 261-281.

Coleman, L. J. & Cross, T. L. (1988). Is being gifted a social handicap? Journal for the Education of the Gifted, 11, 41-56.

Coleman, J. M., & Fults, B. A. (1983, May). Self-concept and the gifted child. Roeper Review, 44-47.

Coleman, J. M., & Fults, B. A. (1985). Special class placement, level of intelligence, and the self-concept of gifted children: A social comparison perspective. Remedial and Special Education, 6, 7-11.

Cornell, D. G. (1990). High ability students who are unpopular with their peers. Gifted Child Quarterly, 34, 155-160.

Cornell, D. G., Delcourt, M. A. B., Bland, L. C., Goldberg, M. D., & Oram, G. (1994). Low incidence of behavior problems among elementary school students in gifted programs. Journal for the Education of the Gifted, 18, 4-19.

Cornell, D. G., Pelton, G. M., Bassin, L. E., Landrum, M., Ramsay, S. G., Cooley, M. R., Lynch, K. A., & Hamrick, E. (1990). Self-concept and peer status among gifted program youth. Journal of Educational Psychology, 82, 456-463.

Dauber, S. L. & Benbow, C. P. (1990). Aspects of personality and peer relations of extremely talented adolescents. Gifted Child Quarterly, 34, 10-15.

French, D. C., Waas, G. A., Stright, A. L., & Baker, J. A. (1986). Leadership asymmetries in mixed-age children's group. Child Development, 57, 1277-1283.

Gallagher, J. J. (1958). Peer acceptance of highly gifted children in elementary school. Elementary School Journal, 58, 465-470.

Gross, M. U. M. (1993). Exceptionally Gifted Children. NY: Routledge.

Grossberg, I. N., & Cornell, D. G. (1988). Relationship between personality adjustment and high intelligence: Terman versus Hollingworth. Exceptional Children, 5, 266-272.

Haier, R. J., & Denham, S. A. (1976). A summary profile of the nonintellectual correlates of mathematical precocity in boys and girls. In D. P. Keating (Ed.), Intellectual talent: Research and development (pp. 225-241). Baltimore, MD: Johns Hopkins University Press.

Harter, S. (1995). Manual for the self-perception profile for children. University of Denver.

Holliday, G. A., Koller, J. R., & Kunce, J. T. (1996). Personality attributes of high IQ/high achieving gifted adolescents: Implications of the personal styles model. Journal for the Education of the Gifted, 20, 84-102.

Hollingshead, A. B. (1975). Four factor index of social status. Unpublished paper. Yale University.

Hollingworth, L. (1942). Children above 180 IQ. NY: World Book.

Janos, P. M., Fung, H. C., & Robinson, N. M. (1985). Self-concept, self-esteem, and peer-relations among gifted children who feel "different". Gifted Child Quarterly 29, 78-82.

Janos, P. M., & Robinson, N. M. (1985). Psychosocial development in intellectually gifted children. In F. D. Horowitz & M. O'Brien (Eds.), The gifted and talented: Developmental perspectives (pp. 149-196). Washington, DC: American Psychological Association.

Karnes, F. A., & Wherry, J. N. (1981). Self-concepts of gifted students as measured by the Piers-Harris Children's Self-Concept Scale. Psychological Reports, 49, 903-906.

Lehman, E. B., & Erdwins, C. J. (1981). The social and emotional adjustment of young, intellectually-gifted children. Gifted Child Quarterly, 25, 134-137.

Lemerise, E. A. (1997). Patterns of peer acceptance, social status, and social reputation in mixed age preschool and primary classrooms. Merrill-Palmer Quarterly, 43, 199-218.

Lemerise, E. A, Harper, B. D., Caverly, S. L., & Howes, H. H. (1997). Stability of peer acceptance, social status, and social reputation in mixed-age primary classrooms. Manuscript submitted for publication.

Lemerise, E. A, Harper, B. D., & Howes, H. H. (1998). The transition from kindergarten to ungraded primary: Longitudinal predictors of popularity and social reputation. Early Education and Development, 9, 187-201.

Lombroso, C. (1893). The man of genius. London: Walter Scott.

Luftig, R. L, & Nichols, M. L. (1990). Assessing the social status of gifted students by their age peers. Gifted Child Quarterly, 34, 111-114.

Maas, P. (1992, March 11). Home-grown grandmasters. Washington Post, pp. B1, B4.

Marsh, H. W. (1990). Self Description Questionnaire H Manual. Campbelltown, Australia: University of Western Sydney.

Marsh, H. W., Parker, J. & Barnes, J. (1985). Multidimensional adolescent self-concepts: Their relationship to age, sex, and academic measures. American Educational Research Journal, 22, 422-444.

Marsh, H. W., & Peart, N. (1988). Competitive and cooperative physical fitness training programs for girls: Effects on physical fitness and multidimensional self-concepts. Journal of Sport and Exercise Psychology, 10, 390-407.

Montour, K. (1976). Success versus tragedy: Wiener and Sidis. Intellectually Talented Youth Bulletin, 1, 3-4.

Montour, K. (1977). William James Sidis, the broken twig. American Psychologist, 32, 265-279.

Norman, A. D. Ramsay, S. G., Martray, C. R., & Roberts, J. L. (1999). Relationship between levels of giftedness and psychosocial adjustment. Roeper Review, 22, 5-9.

Olszewski-Kubilius, P., Kulieke, M. J., & Krasney, N. (1988). Personality dimensions of gifted adolescents: A review of the empirical literature. Gifted Child Quarterly, 32, 347-352.

Otis, A. S., & Lennon, R. T. (1988). Otis-Lennon Multilevel Norms Booklet (6th edition). San Antonio, TX: Harcourt Brace Jovanovich.

Richardson, T. M., & Benbow, C. P. (1990). Long-term effects of acceleration on the social-emotional adjustment of mathematically precocious youths. Journal of Educational Psychology, 82, 464-470.

Rogosch, F. A., & Newcomb, A. F. (1989). Children's perceptions of peer reputations and their social reputations among peers. Child Development, 60, 597-610.

Sayler, M. F., & Brookshire, W. K. (1996). Social, emotional, and behavioral adjustment of accelerated students in gifted classes and regular students in eighth grade. Gifted Child Quarterly, 37, 150-154.

Schneider, B. H. et al (1989). Social relations of gifted children as a function of age and school program. Journal of Educational Psychology, 81, 48-56.

Swerdlik, M. E. (1992). Review of the Otis-Lennon School Ability Test, Sixth Edition [CD-ROM]. 11 Mental Measurements Yearbook.

Tannenbaum, A. J. (1983). Gifted children: Psychological and educational perspectives. NY: Macmillian.

Terman, L. (1925). Mental and physical traits of a thousand gifted children. Stanford, CA: Stanford University Press.

Terman, L. (Ed.). (1959). Genetic studies of genius (Vols. 1-4). Stanford, CA: Stanford University Press.

Manuscript submitted June, 1999.

Revision accepted January, 2000.

Antony D. Norman is an Associate Professor of Psychology at Western Kentucky University teaching in the area of Educational Psychology. Shula G. Ramsay is an Associate Professor of Psychology at Western Kentucky University teaching in the area of Educational Psychology. Julia L. Roberts is a Professor of Education and Director of The Center for Gifted Studies at Western Kentucky University. Carl R. Martray is Dean of the College of Education and Psychology at the University of Southern Mississippi.