DOCUMENT RESUME ED 368 101 EC 302 846 AUTHOR Johnson, Nancy E.; And Others TITLE Understanding Gifted Underachievers in an Ethnically Diverse Population. PUB DATE 94 NOTE 23p.; In: Saccuzzo, Dennis P.; And Others. Identifying Underrepresented Disadvantaged Gifted and Talented Children: A Multifaceted Approach. (Volumes 1 and 2); see EC 302 840. PUB TYPE Reports Research/Technical (143) EDRS PRICE MF01/PC01 Plus Postage. DESCRIPTORS At Risk Persons; Elementary Secondary Education; Ethnic Groups; *Gifted; High Achievement; *Intelligence Quotient; *Knowledge Level; Performance Tests; Sex Differences; *Student Characteristics; Student Motivation; *Underachievement; Verbal Ability IDENTIFIERS Discrepancy Analysis; Wechsler Intelligence Scale for Children (Revised) ABSTRACT Gifted underachievers (n=108) were compared to gifted high achievers (n=96). All children had Wechsler Intelligence Scale for Children Revised (WISC-R) IQ scores of 130 or greater, but underachievers were performing at or below the 50th percentile in at least one major area of achievement, whereas high achievers were at the 96th percentile or greater in three areas of achievement: language, math, and reading. Results of analysis of variance of achievement level X WISC-R subtests revealed significant differences in scores on four verbal subtests: Information, Similarities, Vocabulary, and Comprehension. High achievers had significantly higher verbal, but not performance, IQ scores than underachievers. However, comparison of the verbal IQ-performance IQ discrepancy distributions for the two groups revealed no significant differences, negating the idea that a large verbal/performance IQ discrepancy can be used as an indicator of risk for low achievement in gifted children. Analysis of gender, ethnicity, and risk revealed a greater concentration of non-Caucasian males with at least two risk factors in the underachieving group. Present findings are consistent with earlier findings concerning the importance and discriminating power of the Information subtest in distinguishing high versus underachievers. The findings indicate that gifted underachievers are not as motivated or interested in acquiring traditional factual information as high achievers. (Contains.242 references.) (DB) *********************************************************************** Reproductions supplied by EDRS are the best that can be made from the original document. *********************************************************************** U.S. DEPARTMENT OF EDUCATION OThce ot Educational Research and irnnrovernent EDUCATIONAL RESOURCES INFORMATION CENTER IERICI (This document has been reoioduced as ece.ved from the Person 0, organization originating it r Minor changes nave been made to improve reproduction qualgy Points of view 0, opinions stated in this doCu CHAPTER 6 ment do nOt necessarily represent Why*, OE RI position or policy Understanding Gifted Underachievers in an Ethnically Diverse Population Nancy E. Johnson, Dennis P. Saccuzzo, & Tracey L. Guertin San Diego State University U.S. Department of Education, Jacob * This research was funded by Grant R206A00569, Javits Gifted and Talented Discretionary Grant. Diego City Schools, to Gifted The authors express their appreciation to the San Hermanson, and to the following and Talented Education (GATE) Adrninistrator David Dimaris Michalek, Ben Sy, and Daniel school psychologists: Will Boggess, Marcia Dome, Williams. McLaughlin for her assistance. The authors also wish to acknowledge Susan addressed to Nancy E. Johnson, Correspondence concerning this article should be Diego Clinical Training Program, 6363 Joint San Diego State/University of California, San 92120-4913 (Phone: 619-594-2845/FAX: Alvarado Court, Suite 103, San Diego, California 619-594-6780/E-mail: [email protected]). © 1994 nennission from the authors. Do not reproduce in any form without express written 2 Abstract sample of underachievers was compared to a A well defined sample or gifted of 130 or greater. had full scale WISC-R IQ scores gifted high-achievers. All children major a,.ea below the 50th percentile in at least one Underachievers were performing at or greater in three high-achievers were at the 96th percentile or of achievement, whereas children who had obtained language, math, and reading. Of 6,067 areas of achievement: ethnically diverse a nine year period in an full scale IQ scores of 130 or greater over for high underachievement, and 96 met criteria population, 108 met criteria for gifted repeated level) by 9 (WISC-R subtest) mixed achievement. Results of a 2 (achievement four verbal subtests: significant (p < .01) differences in scores on measures ANOVA revealed had and Comprehension. High-achievers Information, Similarities, Vocabulary, underachievers. but not Performance, IQ scores than significantly (p < .001) higher Verbal, children in the discrepancy ,_listributions for the However, comparison of the VIQ-PIQ the idea that a large significant differences. This finding negates two groups revealed no in gifted indicator of risk for low achievement VIQ-PIQ discrepancy can be used as an high-achievers as discrepancies were as likely to be seen in children, since large VIQ-PIQ concentration of ethnicity, and risk revealed a greater in low. Analysis of gender, Present risk factors in the underachieving group. nonCaucasian males with at least two the importance and confirm those of others concerning findings are consistent with, and high versus subtest in distinguishing discriminating power of the Information motivated or gifted underachievers are not as underachievers. The findings indicate that Creative teaching factual information as high-achievers. interested in acquiring traditional the talents of underachievers. strategies are recormnended to maximize 3 128 in an Ethnically Diverse Population Understanding Gifted Underachievers will not perform defined as those children who cannot or Traditionally, gifted underachievers are 1989; Fine, their intellectual potential (Emerick, achievement commensurate with at a level of academic performance discrepancy between a child's Underachievement manifests itself as a 1967; Gowan, 1955). is expected 1988); a discrepancy between what her intellectual ability (Rimm, in the classroom and his or (Newell, & d'Lberville, 1989). and what is actually accomplished represent one of the greatest time ago, gifted underachievers As Gowan (1955) noted some of exceptional ability who achieve at Gifted underachievers are children social wastes in our culture. tend to be overlooked Unfortunately, gifted underachievers levels. average or even below average Thus, while gifted children may relatively good levels (Wolfe, 1991). because such children perform at their children of normal intelligence, that cause underachievement as are be as susceptible to factors giftedness (Supplee, 1989). be recognized because of their underachievement is less likely to school dropouts are judged to between 10% and 20% of high According to various estimates, fifteen percent 1973; Whitmore, 1980). Ten to (Lajoie & Shore, 1981; Nyquist, have very superior ability 1985; below their potential (Gallagher, believed to achieve at a rate far of the academically gifted are these have called for more study of increasing number of researchers Ford, 1992). Consequently, an lazy, easily missed, or dismissed as their high potential and yet are so gifted children who fail to fulfill 1985; Waffle, 1991). (Emerick, 1989; Ford, 1992; Gallagher, manipulative or irresponsible difficult to identify, satisfactory levels, gifted underachievers are Because they usually perform at research inquiry. Much of the available resistant to systematic scientific and consequently have proved Some 1990; Mather & Udall, 1985). of identification (Ford & Harris, has been devoted to the problem 1989) or socioemotional consequences of the family (VanTassel-Baska, studies have looked at the role studies to examine gifted Still other reports have used case (Cornell, Callahan, & Lloyd, 1991). underachieving children (Hannel, 1990). in order to better understand has been concerned with test patterns A major line of investigation the gifted children. The basis for correlates of underachievement in and operationalize underlying suggest^ t:-,at gifted children process found, in part, in evidence that search for test patterns can be such as the WISC-R from average children on tests information in a qualitatively different manner gifted patterns of bright and Moreover, in a study of WISC test (Brown & Yakimowski, 1987). and underachievers differed on (1973) found that normal achievers underachievers, Bush and Mattson (1987) related study, Moffitt and Silva Arithmetic, and Digit Span. In a three subtests: Information, IQ WISC-R Verbal and Performance unselected birth cohort who had examined children from an sample Underachieving children in this beyond the 90th percentile. discrepancies that placed them the Performance IQ. depressed a Verbal IQ relative to were found to have patterns of gifted underachievers. continued the exploration of the test In the present study, we high- well-defined sample of gifted underachievers was compared to a A well-defined sample of gifted verbal-performance discrepancies. of subtest scores as well as achievers, to explicate differences in patterns Method in the Subjects: for an evaluation of giftedness children who were referred The subjects were drawn from consists of over Diego City School District 1984 and 1993. The San San Diego School District between geographic high schools across a wide 130 elementary, middle, and 123,000 children who attend more than evaluation by teachers, parents, be referred for a giftedness and ethnically diverse area. Children may by a school psychologist child referred is examined from the District office. Each or central nomination aptitude, and risk IQ achievement, including a consideration of who conducts a case study analysis of five whether the child has one or more the psychologist determines factors. For each child evaluated, 129 4 and health. Children cultural/language, economic, emotional, environmental, risk factors as follows: automatically certified above the mean on a standardized IQ test are who score two standard deviations combination of high achievement, high IQ also be certified as gifted based on a as gifted. Children may and risk factors. Procedure: (more than 95%) of children evaluated for giftedness Between 1984 and 1991 the vast majority Children-Revised (WISC-R). A total of 9,315 the Wechsler Intelligence Scale for were achninistered children, we identified during this time period. From these 9,315 children had been given the WISC-R deviations above the mean or greater). A of 130 or greater (i.e., two standard all who had a Full Scale IQ children, a group of underachievers criterion. From this group of high IQ total of 6,067 children met this the Total the mean (i.e., 50th percentile) or lower on selecting all children who scored at was obtained by Skills (CTBS). Such children of the Comprehensive Test of Basic Reading, Language, or Math scores IQ and achievement. A standard deviation discrepancy between would therefore have at least a two criterion. To obtain a high- ethnically diverse sample met this total of 108 children representing an 96th percentile or higher were all three achievement scores in the achieving group, all children who had 11 were Latino, 73 Caucasian, this criterion. For the underachievers, selected. A total of 96 children met Islander). Indochinese, Filipino, or Pacific and 7 Other (Native-American . 10 African-American, 7 Asian, 2 Asian, and 10 Other. Latino, 78 Caucasian, 4 African-American, For the high-achieving sample, 2 were nonCaucasians determine if there were significantly more Chi Square analysis was conducted to results revealed that Compared to an expected 50-50 sRlit, represented in the underachieving group. < x4(1, N = 108) = 5-45, nonCaucasians in the underachieving group there were significantly more the two groups (p > .05). In differences in numbers of Caucasians in .02). There were no significant and 17 had two. In the underachieving high-achievers had none, 23 had one, terms of risk factors, 56 statistically had two risk factors. There were no factors, 23 had one risk factor, and 24 group 61 had no risk risk category In the in the number of children in each significant differences between the two groups 40 were female and 68 male. In the underachieving group, high-achieving group, 49 were female, 47 X2(1, N = 108) = 726, P < overrepresented in the underachieving group, were male. Males were Results routinely given to performance for the nine subtests that were Table 1 shows the mean subtest performance between high and To evaluate differences in subtest the majority of the subjects. 9 (Subtests) mixed repeated in a 2 (Achievement Level) X underachievers, the groups were compared .001, and for Achievement, F(i, 131) = 17.16, p < There were significant main effects measures ANOVA. revealed that the high-achievers .001. Post-hoc multiple comparisons for Subtests, F(8, 1048) = 35.14, p < Similarities, Vocabulary, and .01) on 4 subtests: Information, scored significantly higher (p < Comprehension. Table 1. by High and Underachievers. Deviations for Subtest Pvfortnance WISC-R Means and Standard Entire Sample Underachievers High-achievers WISC-R Subtest (SD) (M) (SD) (M) (SD) (M) (2.02) 14.30 (1.87) 13.88 (2.09) 14.78 Information (2.01) 16.97 (1.97) 16.65 (1.99) 17.35 Similarities (2.24) 14.55 (2.15) 14.40 (2.33) 14.73 Arithmetic (2.09) 15.67 (2.08) 15.40 (2.08) 15.98 Vocabulary (1.94) 16.69 (1.97) 16.13 (1.76) 17.23 Comprehension (2.44) 13.83 (2.14) 13.90 (2.44) 13.76 Picture Completion (2.49) 14.68 (2.51) 14.67 (2.47) 14.70 Picture Arrangement (2.53) 15.09 (2.51) 14.82 (2.53) 15.39 Block Design (2.79) 14.21 (2.73) 14.01 (2.86) 14.44 Object Assembly 130 5 underachievers Since all of the differences found were for Verbal subtests, high-achievers and (Verbal versus Performance IQ) ANOVA. PIQ scores for were compared in a 2 (Achievement Level) X 2 significantly. high-achievers (M = 1322; SD = 9.7) and underachievers (M = 130.6; SD = 9.1) did not differ with a mean of 137.8 (SD = 8.5) for However, VIQ scores did differ significantly, F(1, 202) = 13.5, p < .001, underachievers. the high-achievers and a mean o 133.4 (SD = 7.8) for the differ in individual VIQ To investigate the possibility that high-achievers and underachievers for the two groups were compared. No PIQ discrepancy scores, VIQ - PIQ frequency distributions .263). As can be seen in Figure significant differences were found (Kolmogorov-Smirnov Z = 1.007, p = be seen in high-achievers as in low achievers. 1, relatively large VIQ - PIQ discrepancies were as likely to of achievement. Figure 1. Distribution of VIQ PIQ differences at the extremes 0.14 0.12 HIGH-ACHE:EVERS 0.10 0.08 10%. 0.06 0.04 0.02 nil 0.00 Cr, N in in 1-INNN rg m co N In in 0. co ° 0000 00 Cr% cr, e-e N IT 1-1 h.: he v-1 2 2 2 .2 0 0 0 0 0 0 0 en Ch N %ID 'a C N en rg h;1 in 1 VIQ - PIQ 0.14 0.12 UNDERACHIEVERS 0.10 0.08 .- 11,1 0.06 0.04 0.02 0.00 l vi in en in N N N os v. in et, ve Vi 1-1 co cr, e VI 2 2 2 2 0 0 0 0 0 0 N v-i 1-1 ' 0 0 0 0 0 0 2 " 0 '46 g A !1 '4 cm sez: 1 1 1 VIQ - PIQ . 131 . 6 analysis and stepwise multiple Performance was further analyzed through correlational intercorrelation matrix of the WISC-R subtests and C 1 BS regression (see Table 2). Table 2 shows the multiple linear regression was performed, with the (Language, Reading, and Math) scores. Stepwise and level of achievement (i.e., whether the child was in the nine Wechsler subtest scores as predictors the criterion. Three subtests were significant in predicting high versus underachievement group) as into the equation was the Information subtest, with a achievement level. The first variable that entered Comprehension subtest added significant variance, multiple R of .29, F(1, 131) = 12.08, p < .001. The R to .37. Finally, the Block Design subtest significantly, F(2, 30) = 10.1, p < .001, and increased the multiple the multiple R to .41. F(3, 129) = 8.5; p < .001, increased Table 2. Achievement Scores Correlation Matrix for WISC-R Subtests and OA BD PC PA VOCAB SIMS ARITH COMP INFO 1.00 INFO 1.00 COMP .22 1.00 .07 ARITH .11 1.00 .09 .18* .07 SIMS 1.00 .29** .11 .24** VOCAB .40** 1.00 .07 -.07 .02 -.11 -.01 PC 1.00 .12 .06 .02 .11 -.11 -.08 PA 1.00 .03 .06 -.06 -.07 -.12 -.16 .02 BD 1.00 .31** .18* .36** .09 .08 -.06 -.07 OA -.11 -.12 -.07 -.05 .01 -.03 -.06 COaNG -.05 .20* -.21* .10 .09 -.04 -.10 .15 .05 .10 .29** C1BSL .26** .11 .11 .01 .02 .12 .02 .10 .23** .23** C I BSR .03 .16 .01 -.04 -.04 .12 .07 .17 .05 C 1 BSM C 1 BSM CTBSR CODING C113SL 1.00 CODING 1.00 .08 C fl3SL 1.00 .71** .11 C113SR 1.00 .40** .52** .09 C I BSM C1 BSL = C 113S Total Language Note.: C1 BSR = C1.13S Total Reading C 1 BSM = CIBS Total Math *p < .05 ** p < .01 and ethnic differences between high-achievers To aid in understanding the gender and simultaneously. in terms of gender, risk, and ethnicity underachievers, the groups were further compared factors in the male, nonCaucasians with 1 or more risk Chi Square analysis revealed significantly more .019. underachieving group: x2(1, N = 33) = 5.50, p < 132 Discussion achieving at least two The present study compared intellectually gifted children who were sample. In general, the high-achieving standard deviations below expectation to a very high-achieving however, was attributable only to Verbal subtests. sample had slightly higher IQ scores. This superiority, Information, Similarities, Vocabulary, and The high-achievers had significantly higher scores on higher for the high-achieving group, Comprehension. Although mean Verbal IQ was significantly subtests or for the Performance IQ as a whole. there were no differences for any of the Performance plotted in frequency Differences in the pattern of individual V1Q PIQ discrepancy scores were between the two distributions. Indeed, VIQ distributions for each group. No differences were found found to be equally common in both high- PIQ discrepancies on the order of 15 points or greater were the fallacy of confusing statistical significance achievers and underachievers. This finding underscores with 95% certainty that an individual's VIQ and PIQ (i.e., the 15-point difference necessary to conclude conclusion that a 15-point V1Q-PIQ difference differ) with clinical significance (i.e., the mistaken risk for underachievement). Large VIQ-PIQ necessarily has prognostic significance and indicates low-achieving gifted children; only with the addition discrepancies are equally common in high- and be identified. In terms of predicting achievement of a low achievement test score can a low-achiever Information and Comprehension, with Block level using WISC-R subtests, the primary correlates were the variance. Design adding a small, but significant, contribution to concentration of nonCaucasian Analysis of gender, ethnicity, and risk further revealed a greater These findings are consistent with males with at least two risk factors in the underachieving group. underachievers, males outnumber females previrus studies, which indicate that of the intellectual ethnic minorities (Ford, 1992). Thus, (Gallagher, 1985; Wolf le, 1991), and that many of these students are particular focus on underachieving minority there is a need, as advocated by Gallagher (1985), to provide males. (1973) concerning the importance and Our results confirm the findings of Bush and Mattson high-achievers versus underaclz;evers. discriminating power of the Information subtest in distinguishing generalize to the WISC-R. The findings Present findings show that the older results with the WISC value, seem to suggest that gifted underachievers pertaining to the Information subtest, taken at face information as do high-achievers. simply do not have as much interest or motivation for acquiring factual teaching strategies, such as making This suggests that gifted underachievers may require creative abilities into more creative pursuits. information more relevant and interesting or channeling their by Moffitt and Silva (1987). Gifted Our findings are also consistent with those reported verbal skills; their Performance IQ's are underachievers are characterized by certain depressed characterize the gifted underachiever as an comparable to that of the high-achievers. Thus, we can Cattell (1963) would say, fluid intelligence, to individual who has not used his or her potential, or as intelligence). Again, the challenge for teachers acquire a traditional body of knowledge (i.e., crystallized make full use of their potential. is to find ways to motivate these underachievers to also relevant to a previous report of the Our findings pertaining to gifted underachievers are in 4,546 gifted African-American, direction of the difference between Verbal versus Performance IQ & Russell, 1992). This study showed Caucasian, Filipino, and Hispanic children (Saccuzzo, Johnson, IQ was actually higher than the Performance that for the typically gifted African-American, the Verbal roughly equivalent. Thus, the relevant For Hispanics, the Verbal and Performance 1Q's were IQ. Performance IQ relative to Verbal IQ for an dimension includes both direction and size; a very high the possibility of a gifted underachiever African-American and perhaps an Hispanic should signal Performance IQ's. For Filipinos, just the because these individuals tend to have higher Verbal than (Saccuzzo et al., tend to have higher Performance than Verbal IQ reverse is true since these individuals of selection (Johnson, 1992), it (or its relative, 1992). Therefore, while the WISC-R may be biased in terms underachieving African-American and Hispanic/ the WISC-III) may still have utility in identifying gifted Latino students. 8 133 three major strategies, researchers have pointed to Beyond modification of our educational (e.g., Boyd, 1990). The first focuses on motivational factors approaches to gifted underachievers. learning process in order need to add excitement or relevance to the According to this model there is a emphasizes the importance of fulfill their potential. A second approach to help gifted underachievers (VanTassel-Baska, 1989). and support for gifted underachievers families as a source of encouragement one's especially locus of control importance of personality variables, The third emphasizes the Jenkins-Friedman, & Tollefson, control the events of one's life (L,affoon, perceived ability to influence or 1990). Certainly any one, two, or Rosenbaum, 1987; Wi Rings & Greenwood, 1989; Waldron, Saphire, & need to be considered in addressing role in gifted underachievement and all three of these factors play a and unique student. the problems of each individual 9 134 . . References den Zusammenhang von Hochbegabung und Albrecht, H. T. & Rost, D. H. (1983). fiber San Diego, USA. / On the relationship between Wohnqualitat. Eine Wohnbezirksanalyse aus analysis of San Diego, USA. Psychologie in giftedness and housing quality: A social area Erziehung und Unterricht, 30, 281-290. gifted and (1981). National survey of identification practices in Alvino, J., McDonnel, R., & Richert, S. Children, 46, 124-132. talented education. Exceptional (Report No. EDO-UD bilingual students for placement and instruction. Ascher, C. (1990). Assessing Service No. University. (ERIC Document Reproduction 90-5). New York, NY: Columbia ED377773). the selection of magnet Raven Advanced Progressive Matrices for Baska, L. (1986A). The use of the Roeper Review, 8, 181-184. junior high school students. traditional testing. Roeper Review, 8, 181-184. Baska, L. (1986B). Alternatives to related to social class and Children's feelings of personal control as Battle, Ester,& Rotter, J. (1963). 31, 482-490. ethnic group. Journal of Personality, low economic bilingualism in children from Spanish-English Ben-Zeev, S. (1977). The effect of Papers on Bilingualism, and cognitive strategy. Working neighborhoods on cognitive development 14, 83-122. perspective. (Report No. American children: An ethnic-scientific Bemal, E. M. (1974). Gifted Mexican Reproduction Office of Education. (ERIC Document OEC4-7-062113-307). Washington, D. C.: Service No. ED091411). under backward masking. changes in visual information processing Blake, J. (1974). Developmental Psychology, 17, 133-146. Journal of Experimental Child 163-165. the baby. Roeper Review, 8, Throwing out the bathwater, saving Borland, J. H. (1986). IQ tests: Child lower- and middle-class children. information processing speed of Bosco, J. (1972). The visual Development, 43, 1418-1422. of high school. Gifted Education talented underachievers at the end Boyd, R. (1990). Academically International, 7, 23-26. role L. (1984). Locus of control, sex Furr, S., Ware, W., & Voight, N. Brown, D., Fulkerson, Fee, K., sixth-grade male and female leaders black and white third- and orientation, and self-concept in Developmental Psychology, 20, 717-721. in a rural community. WISC-R. of gifted students on the M. E. (1987). Intelligence scores Brown, S. W., & Yakimowski, Gifted Child Quarterly, 31, 130-134. and not-at- gender differences in at-risk (1991). Social, personality, and Browne, C. S., & Rife, J. C. 482-495. Journal of Early Adolescence, 11, risk sixth-grade students. Gifted of the disadvantaged gifted. of procedures for identification Bruch, C. B. (1971). Modification Child Quarterly, 15, 267-272. school children of Progressive Matrices among study on the Raven Coloured Burciaga, L.E. (1973). A research of El Paso, Texas. Doctoral dissertation, University the El Paso public schools. 135 1 0