The emotional adjustment of gifted adolescents: a view of global functioning.

The purpose of the study was to determine if academically gifted adolescents differ from nongifted adolescents with regard to their perception of overall emotional adjustment as measured by a composite score (Emotional Symptoms Index or ESI) on the Self Report of Personality (SRP) of the Behavior Assessment System for Children (BASC). Gifted adolescents (N = 115) from a resource support program in grades 9 through 12 were compared with nongifted peers (N = 97) from a suburban high school within the Atlanta metropolitan area. The comparison group was composed of volunteers from one regular education English class from each grade level (9-12) with a random, computer-assigned enrollment. A 2 x 4 analysis of variance factorial design was used to analyze data obtained from the study in assessing whether the means of the eight groups differed significantly with regard to giftedness and/or grade on the composite measure (ESI) of emotional adjustment. Results revealed the overall emotional adjustment of both groups falls within normal limits when compared with the General norms established by the authors of the BASC. The gifted group obtained significantly lower T-scores than the nongifted group, suggesting the presence of fewer indicators of behavioral-emotional maladjustment. No difference was obtained between groups with regard to grade. Results support previous findings that, when a difference exists between the overall emotional adjustment of gifted and nongifted students, the gifted students show better adjustment on self-report instruments.

Parents and professionals in education and mental health can often be heard espousing the belief that the global or overall emotional functioning of gifted individuals is characterized by more emotional maladjustment than that of their nongifted peers. In spite of numerous years of research comparing the emotional adjustment of gifted and nongifted populations, dismissal of the stereotype has not been accomplished; thus, the issue remains an unresolved subject of investigation and debate. Furthermore, the association of giftedness and emotional maladjustment still continues to influence the perceptions and decisions of adults and professionals regarding gifted youth; therefore, consequences of such a view may prove grave and detrimental when symptoms of mal-adjustment in gifted adolescents are perfunctorily attributed to giftedness and are ignored. Continued investigation in this area will facilitate better understanding of the gifted population, better identification of those at risk for emotional problems, and more effective treatment and guidance. Utilization of a valid composite index of emotional adjustment to compare gifted and nongifted groups will provide additional insight into the appropriateness of commonly held views regarding the global adjustment of gifted adolescents.

The purpose of this study was to compare the overall emotional adjustment of a gifted adolescent sample in a resource gifted program with a sample of nongifted students through the use of the adolescent version of the Self-Report of Personality (SRP) of the Behavioral Assessment System for Children (BASC) (Reynolds & Kamphaus, 1992). The main research hypothesis for this study, which is based on current research, is that the overall emotional adjustment of gifted students will not vary significantly from that of a nongifted group of peers as measured by the global Emotional Symptoms Index (ESI) of the SRP of the BASC. A second hypothesis for this study is that the overall emotional adjustment of gifted adolescents falls within normal limits when compared with general population norms on a measure of emotional and personality functioning.

The pairing of high intelligence with emotional maladjustment had its beginnings in the views of people such as the criminologist Lombroso (1891), who purported as early as 1891 that there is an association of high intelligence with insanity and emotional instability. According to Galluci (1988), Clinical and school lore has suggested that some gifted children were prone to interpersonal isolation .... low self esteem .... depression, suicide, and hypersensitivity... (p. 273).

The longitudinal studies of Terman and his associates refuted the stereotype associating maladjustment and giftedness and served as a genesis for the point of view that the emotional adjustment of gifted children is similar to or better than that of nongifted peers (Grossberg & Cornell, 1988). Results of the studies by Terman in actuality suggested that the emotional adjustment of gifted individuals was better than their nongifted counterparts (Gallagher, 1990). Terman's subjects identified as having an IQ greater than 140 on the Stanford-Binet tended to fare better on wide ranging adjustment variables than the norm group. Delisle (1990) relates that Terman's work so strongly stressed the positive adjustment of gifted persons that a new myth arose that gifted individuals are practically immune to mental problems.

Galluci (1988, 1989) stated that other studies employing standardized tests of personality have concurred with the findings of Terman and his associates, thus providing additional evidence that the personality' functioning of gifted subjects is similar to or more developmentally mature than cohorts of average ability and comparable chronological age. Galluci points out that these results have been obtained although differences in the studies have existed in standardized tests of personality, tests of intelligence, the mean and range of intellectual functioning, and the mean and range of the chronological ages of subjects. Grossberg and Cornell (1988) and Freeman (1994) also reported that results similar to Terman's have been obtained by other studies.

Olszewski-Kubilius, Kulieke, and Krasney (1988) in their review of the literature on personality dimensions of gifted individuals, concluded that differences tend to favor gifted students when they are found in studies comparing gifted students with nongifted students of the same age. The review of the literature conducted by Olszewski-Kubilius et al. included studies of same-age comparisons of four groups including kindergartners, elementary school children, adolescents, and college age students. Gifted students displayed lower levels of anxiety and tended to be well adjusted with fewer indicators of psychological maladjustment. Gifted students obtained higher scores than their nongifted agemates on positive qualities when multidimensional instruments of personality were used. According to Olszewski-Kubilius et al., the research is uncertain and inconsistent when specific personality dimensions, such as self-concept and locus of control, are investigated.

Galluci (1989) employed the Rorschach and the Child Behavior Checklist to assess indicators of psychopathology in children with IQs greater than 135 on the Stanford-Binet Form L-M. Galluci's findings revealed that the average and superior intelligence samples displayed a comparable incidence of psychopathology. Galluci also conducted another comparison between the group of children with IQs greater than 150 and the group with IQs falling between 136 and 140. The results of the two groups were analogous, suggesting no significant difference in the occurrence of psychopathology.

Luthar, Zigler, and Goldstein (1992) compared gifted adolescents with chronologically matched individuals not identified as gifted, chronologically matched peers who were athletically talented, and older adolescents matched on cognitive maturity. Results of the study by Luthar et al. employing multiple indices of adjustment corroborate earlier findings indicating that gifted adolescents who are high achievers display more positive psychological adjustment than nongifted age-mates. According to the researchers, the association of positive psychological adjustment with giftedness and high achievement suggested by all of the comparisons may be further attributed to greater cognitive maturity and experiential factors, such as those that relate to successes experienced in the past.

Beer (1991) investigated the emotional adjustment of 27 gifted adolescents in junior high and high school on a variety of measures, including Children's Depression Inventory, the Beck Depression Scale, Test Anxiety Scale, General Anxiety Scale, and Breskin's Rigidity Scale. Results concur with research indicating that the adjustment of gifted children is generally better than that of peers of average intelligence and below. Beer found that gifted adolescents may display some rigidity in their outlook, but they do not tend to be depressed nor do they experience much general anxiety or test anxiety. According to Beer, gifted students overall are characterized by mental flexibility, emotional resilience and the ability to think positively, which may account for the better emotional adjustment.

Lessinger and Martinson (1961) employed the use of the California Psychological Inventory to compare eighth grade girls in a gifted program with a random sample of eighth grade girls, and likewise to compare eighth grade gifted boys with randomly chosen eighth grade boys. The gifted males obtained higher average scores on all scales of the CPI than their counterparts; whereas, the gifted females scored higher on all CPI subtests with the exception of femininity than did the randomly chosen counterparts.

Grossberg and Cornell (1988) indicate that a second position derived from the earlier studies of Hollingworth proposes that the emotional adjustment of highly gifted children is less than that of their nongifted peers. According to Grossberg and Cornell and Luthar et al. (1992), others have repeated concerns similar to Hollingworth's findings, suggesting that highly intelligent children were vulnerable to the development of social and emotional adjustment problems.

Grossberg and Cornell (1988) investigated the relationship between intelligence and personality adjustment in gifted children. The researchers employed both student- and parent-report measures to assess the emotional adjustment of 83 7- to 11-year-old children with IQs within the range of 120 and 168. Grossberg and Cornell reported a modest support for the view that intelligence is positively related to adjustment within the gifted range; however, the study did not support the view of Hollingworth that very high IQ is associated with poor adjustment.

Olszewski-Kubilius, Kulieke, and Krasney (1988), in their review of the literature on personality dimensions of gifted individuals, categorized the previous research according to four perspectives: same-age comparisons between gifted and nongifted individuals, comparisons of gifted individuals with nongifted, chronologically older persons, gifted male-female comparisons, and gifted achiever-nonachiever comparisons. Olszewski-Kubilius et al. pointed out several methodological problems that account for inconsistencies in research results. These related to the following: difficulty making comparisons across studies due to varying criteria for giftedness which involve the use of IQ, off-level tests, and achievement tests; deceptive significant findings resulting from the exclusive employment of univariate statistical tests as opposed to multivariate tests with multidimensional personality instruments; sampling error such as small sample size and confounding variables such as unexamined male-female differences and/or SES that result from comparisons between relatively homogeneous samples of gifted students and heterogeneous norming groups.

Several researchers (e.g., Gallagher, 1990; Freeman, 1994) cited sampling error in the studies of Terman in particular, which were heavily biased by educational and economic factors. Gallagher hypothesized that these factors would more likely create an environment more conducive to superior social acceptance and emotional status of gifted individuals.

Additional research investigating personality variables on standardized self-report measures should prove helpful in further eliminating the myths regarding the emotional adjustment of gifted individuals, which in turn will facilitate greater understanding of their needs and utilization of more appropriate interventions and services.

Methodology

Subjects

The data set was collected in a suburban public high school with an enrollment of approximately 2,000 students and located within the metropolitan Atlanta area. The persons (N = 212) who served as subjects for the study were 115 gifted and 97 nongifted students in grades 9 through 12.

The 60 females and 55 males who served as subjects for the gifted group were volunteers from the school's gifted program, which is based on a resource model. Placement in the gifted program was determined by ability and achievement indicators which include a minimal IQ score of 128 on a group or individual measure of intelligence and at least one in-grade achievement test level falling at or above the 90th percentile in reading or math or the 85th percentile in overall achievement. The composition of the group consisted of 32 ninth graders, 33 tenth graders, 24 eleventh graders, and 26 twelfth graders. Three percent were African-American, 91 percent were Caucasian, 1 percent was Hispanic American, 3 percent were Asian American, and 2 percent were unspecified. Seventy four percent lived with both of their biological parents.

A comparison group (N = 97) of nongifted students consisted of 39 males and 58 females who were obtained from one regular education English class in each grade level 9 through 12. English classes were chosen since every student is required to enroll in English at each grade level. Students had been randomly assigned to the English classes by a computer. The comparison group consisted of 20 ninth graders, 24 tenth graders, 30 eleventh graders, and 23 twelfth graders. Eighteen percent of the nongifted students were African Americans, 67 percent were Caucasian, 3 percent were Native Americans, 5 percent Hispanic Americans, 1 percent were Asian Americans, and 6 percent were unspecified. Sixty-four percent resided with both biological parents.

A total of 212 students with valid profiles (Gifted N = 115; Nongifted N = 97) participated in the study. Validity of the profiles was determined by age parameters, missing data, and random guessing. The subjects were divided into eight groups that were determined by grade level and participation in the gifted program.

Instruments

The adolescent level of the Self-Report of Personality (SRP), which is one component of the multimethod assessment system of the Behavior Assessment System for Children (BASC), was selected because it was empirically derived and was created to measure the personality, emotions and self-perceptions of children (Reynolds & Kamphaus, 1992). The adolescent level (ages 12 - 18) of the SRP, a 186-item inventory, was designed to yield a number of scores reflecting maladjustment and adjustment, including 4 composite scores, 14 individual scale scores, 3 validity and 2 caution indexes. The most global indicator of serious emotional disturbance, according to Reynolds and Kamphaus, is the T Score represented by the Emotional Symptoms Index (ESI); therefore, it was chosen as the measure of overall emotional adjustment in the present study.

T-scores on this instrument are based on X = 50, SD = 10. The ESI incorporates the individual scales of Depression, Sense of Inadequacy, Social Stress, Anxiety, Interpersonal Relations, and Self-Esteem. The inventory's individual scales were developed to assess both adjustment (Self-Esteem, Self-Reliance, Relations with Parents, and Interpersonal Relations) and maladjustment (Atypicality; Locus of Control; Somatization; Social Stress; Anxiety; Depression; Sense of Inadequacy; Attitude to School; Attitude to Teachers; Sensation Seeking); therefore, each of the individual scales was selected to measure specific indicators of emotional adjustment and maladjustment.

The SRP BASC was normed separately on both general and clinical populations. Reynolds and Kamphaus (1992) recommend the use of the general norms for general use. The present study employed the general-population norms for ages 12 to 18 years which were based on 4,448 male and female subjects. The norms of the adolescent level of the SRP were subdivided into ages 12 -14 and 15 - 18 due to the differences that exist between the older and younger ends of the overall adolescent age range. The standardization samples were chosen to match the current U.S. demography for children and also included representative proportions of children with special education classifications. The samples were weighted by race/ethnicity and gender.

Reynolds and Kamphaus (1992) reported that the resulting scales have relatively high internal consistency reliabilities, averaging about .8 for each gender at both the adolescent and child age levels. The authors indicated that the internal-consistency reliabilities are very high for the composite scores of the General norm sample, extending from the mid-80s to the mid-90s. Test-retest reliabilities reported for the individual scales reflect a median value of .76 at each age level; whereas, test-retest reliabilities for the composites were found to fall in the low to middle 80s with the exception of one.

Reynolds and Kamphaus (1992) report that test-retest reliability correlations for the individual scales at each age level over a period of several weeks have a median value of .76. The retest correlations for the composite scores are relatively high, falling in the low-to-middle .80s with one exception. The ESI's test-retest reliabilities are .82 and .84.

Procedure

The subjects in both the experimental and control groups were administered the SRP in groups by the Gifted Program and English teachers respectively. Each teacher had previously been given both oral and written instructions with regard to administration of the SRP and the demographic questionnaire. The completed protocols were scored through the use of the microcomputer program, "BASC Plus." Overall emotional adjustment was measured by the ESI on the SRP of the BASC. Specific dimensions of emotional adjustment and maladjustment were measured by the T scores on the individual scales for the SRP.

Data Analysis

A 2 x 4 analysis of variance factorial design was used to analyze the data obtained from the study in assessing whether the means of the 8 groups differed significantly with regard to giftedness and/or grade on the composite measure (ESI) of emotional adjustment. Significance was set at the .05 level. The group sizes were roughly equal which ensures that results are robust to violations of homogeneity of variance and nonnormality of distribution.

An additional comparison was made with the General norm sample obtained for the SRP of the BASC by the developers of the instrument. Reynolds and Kamphaus (1992) report in the BASC manual that a T-score range of 41 to 59 demarcates normal limits ("Average"), and they recommend use of the General norms as previously mentioned.

Results

Table 1 presents the means and standard deviations of the subjects according to giftedness and grade level and the total mean and standard deviation of the gifted and nongifted groups as measured by the ESI. Examination reveals that the gifted group obtained lower mean T-score values on the global measure of emotional symptoms (ESI). It should be noted that there is a positive correlation between the ESI score and the presence of symptoms/indicators of emotional maladjustment.

Table 1
Means and Standard Deviations of the Gifted & Nongifted
Groups by Grade on Emotional Symptoms Index

Group                                 Grade
                  Total             9       10     11       12
                  T-Score
Gifted              47.18        44.63    48.79   48.75   46.85
                  (SD = 8.49)
Nongifted           51.06        52.50    53.17   50.77   48.00
                  (SD = 10.12)

Lower mean ESI T-scores were also obtained by the gifted group at each grade level when compared with the nongifted group at the comparable grade levels; however, the two groups did not vary significantly at any single grade level. ESI scores obtained by both groups examined in the study fall within the average range or normal limits when compared with the General norm population of the BASC-SRP.

Results of the 2 x 4 ANOVA which tested the difference between the group means are presented in Table 2. The calculated F scores for main effect (4.137) and grouping (11.513) according to giftedness were significant at p [is less than] .003 and p [is less than] .001 respectively. No significance was indicated by the calculated F scores for grade factor nor two-way interaction of grade and giftedness.

Table 2
Analysis of Variance of Gifted and Nongifted Groups by
Grade on the ESI

SOURCE               df     SS        MS      F RATIO   PROBABILITY
Gifted/Nongifted      1   763.355   763.355    8.984      .003(*)
Grade                 3   385.392   128.464    1.512      .213
2 Way Interactions    3   336.803   112.268    1.321      .268

(*) p < .01

Discussion

This study was conducted primarily to investigate whether academically gifted and nongifted adolescents differ in overall emotional adjustment. Results do clearly reflect that the gifted students are emotionally better adjusted, a finding that is inconsistent with the current study's primary hypothesis and previous studies that have yielded analogous adjustment levels between the two groups (e.g., Galluci, 1988, 1989). Outcomes of this research, however, provide substantial support for the study's second hypothesis that the overall emotional adjustment of gifted adolescents falls within average limits when compared with the general population norm group on a self report measure of emotional and personality functioning. The results of this study, therefore, contribute most notably to the body of empirical evidence that gifted adolescents display a greater degree of emotional adjustment than nongifted adolescents (Terman;. Luthar et al., 1992)

Several questions, however, remain unanswered by the current study, as well as previous investigations, in attempting to more clearly delineate the differences in overall emotional adjustment of gifted and nongifted adolescents. Most significantly, these pertain to the possible role of gender differences and differences that may arise on specific dimensions impacting global measures of psychological adjustment, such as anxiety, depression, interpersonal relations, self-esteem, and locus of control. Future studies academically gifted adolescents are needed to clarify the role of gender and to gain additional insight into the multidimensional profile of global emotional functioning.

The present investigation provides substantial evidence that challenges the longstanding myth equating giftedness and emotional maladjustment. With regard to the research problem, findings of this study suggest that symptoms of behavioral and/or emotional maladjustment must not be perfunctorily attributed to the idiosyncracies of giftedness. The adjustment of gifted students presenting with behavioral-emotional difficulties should receive careful attention in developing appropriate interventions which specifically address relevant academic and emotional needs. Gifted students are multidimensional and complex as are any nongifted peers, particularly during adolescence. Being academically gifted and academically successful may provide the student with many opportunities and successes in life; however, true success is achieved by adjustment in all areas of life, including one's mental health. Accolades and achievements mean little if one's inner peace is lost in the process. Continued research with improved methodology (e.g., use of multidimensional personality measures examining both global and specific domains; increased sample size; comparable normative and gifted samples) should prove beneficial in understanding and addressing the needs of the gifted student.

REFERENCES

Beer, J. (1991). Depression, general anxiety, test anxiety, and rigidity of gifted junior high and high school children. Psychological Reports. 69, 1128-1130.

Delisle, J. R. (1990). The gifted adolescent at risk: Strategies and resources for suicide prevention among gifted youth. Journal for the Education of the Gifted, 13, 212-228.

Freeman, J. (1994). Some emotional aspects of being gifted. Journal for the Education of the Gifted, 17, 180-197.

Gallagher, J. J. (1990). Editorial: The public and professional perception of the emotional status of gifted children. Journal for the Education of the Gifted, 13(3), 202-211.

Gallucci, N. T. (1988). Emotional adjustment of gifted children. Gifted Child Quarterly, 32, 273-276.

Galluci, N. T. (1989). Personality assessment with children of superior intelligence: Divergence versus psychopathology. Journal of Personality Assessment, 53, 749-760.

Grossberg, I. N., & Cornell, D. G. (1988). Relationship between personality adjustment and high intelligence: Terman versus Hollingworth. Exceptional Children, 55, 266-272.

Lessinger, L. M. & Martinson, R. A. (1961, March). The use of the California Psychological Inventory with gifted pupils. Personnel and Guidance Journal, 772-575.

Lombroso, C. (1891). The man of genius. London: Scott.

Luthar, S., Zigler, E., & Goldstein, D. (1992). Psychosocial adjustment among intellectually gifted adolescents: the role of cognitive-developmental and experiential factors. Journal of Child Psychology and Psychiatry, 33, 361-373.

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

Reynolds, C.R., & Kamphaus, R.W. (1992). BASC: Behavior Assessment System for Children. Circle Pines, MN: American Guidance Service, Inc.

Joe M. Nail, a doctoral student at Georgia State University is employed as the lead psychologist for the psychological services department of a school system in metropolitan Atlanta. J. Gary Evans is an Assistant Professor of School Psychology within the Counseling and Psychological Services Department of Georgia State University.

Manuscript submitted July, 1996. Revision accepted May, 1997.