Multiple intelligences of Chinese gifted students in Hong Kong: perspectives from students, parents, teachers, and peers.

It is generally acknowledged that identifying and educating gifted and talented children has always been a major concern in education. However, there is still a lack of relative consensus among experts in gifted education on how to conceptualize or define giftedness and talent (Pfeiffer, 2003; Winner, 2000). Evidently, an endorsement of, for example, Terman's (1925) psychometric theory of high intelligence, or Tannenbaum's (1983, 1997) five-factor predictive model, or Renzulli's (1978, 1986) three-ring conception of giftedness, or Piirto's (1999) pyramid model of talent development, would suggest very different approaches to how educators identify talent and promote talent development in children. Despite these differences, giftedness or talent has always been defined in association with high intelligence. Specifically, children's abilities have been traditionally assessed by measures that reflect a narrow definition of intelligence, and consequently giftedness is defined as high intelligence above a certain cutoff score (see Callahan, 1996).

This notion of a general unitary intelligence that cuts across all areas of human competence has not gone unchallenged (see Karolyi, Ramos-Ford, & Gardner, 2003; Guilford, 1967; Sternberg, 1986, 1997, 2000). Gardner (1983, 1993, 1999), in particular, believed that a unitary approach did not do justice to the strengths and weaknesses in assessing an individual, and proposed that each individual has multiple human abilities or multiple intelligences. On the basis of a set of eight criteria, Gardner identified eight intelligences that each individual possesses to varying degrees. These eight intelligences can be defined and summarized as follows: (1) verbal-linguistic intelligence represents the capacity to use words effectively, whether orally or in writing, (2) musical intelligence represents the capacity to perceive, discriminate, transform, and express musical forms, (3) logical-mathematical intelligence represents the capacity to use numbers effectively and to reason well, (4) visual-spatial intelligence is the ability to perceive the visual-spatial world accurately and to perform transformations on those perceptions, (5) bodily-kinesthetic intelligence includes the ability to use the body to express ideas and feelings, and the facility in using one's hands to produce or transform things, (6) intrapersonal intelligence is the ability to act adaptively on the basis of self-knowledge, (7) interpersonal intelligence is the ability to perceive and make distinctions in the moods, intentions, motivations, and feelings of other people, and (8) naturalist intelligence represents the ability to observe patterns in nature, identify and classify objects, and understand natural and human-made systems.

The theory of multiple intelligences (MI) provides a useful framework for understanding both the basic competencies of all people as well as the unique strengths of individuals. Specifically, a student can be considered gifted in a discipline of study or domain that draws on one or more of the eight intelligences (Gardner, 1993). Yet, there is no one-to-one mapping of intelligences and gifted ness. While mathematical activities, for example, would often harness logical-mathematical and visual-spatial intelligences, good performance on spatial tasks might not be necessarily associated with visual-spatial intelligence for some students who employ semantic strategies that rely primarily on verbal-linguistic and logical-mathematical intelligences. Nonetheless, MI theory enhances gifted education practitioners' awareness of the needs of gifted and multiply gifted children, and has implications for identification, instruction, and evaluation (Fasko, 2001).

More broadly conceptualized, MI theory supports the pluralistic view of intelligence, taking into consideration that there might be uneven or asynchronous development across different abilities, and not restricting its scope or intending to define it in its entirety. In addition, MI theory also highlights that students with well-developed conventional intelligences, such as the verbal-linguistic and logical-mathematical intelligences that are highly valued in traditional classrooms, might be privileged to be labeled gifted, while students with well-developed intelligences not assessed by conventional IQ measures might be underrepresented in the identification process. Indeed, MI theory has proved to be helpful in identifying underrepresented and culturally diverse groups of gifted students for participation in gifted education programs (see Maker, Nielson, & Rogers, 1994; Sarouphim, 1999).

The challenge to move away from the reliance on a single measure of intelligence has also been a major concern in gifted education in Hong Kong. Until relatively recently, identification of gifted students to be eligible for government enrichment activities required IQ assessment, and a score of 130 or above on the Hong Kong Wechsler Intelligence Scale for Children (HK-WISC; Psychological Corporation, 1981) or Raven's Standard Progressive Matrices (SPM; Hong Kong Education Department, 1986; Raven, Raven, & Court, 1998) was the sole criterion for admission to programs. This practice is undergoing changes as educators in Hong Kong start to incorporate and broaden the notion of giftedness to include different facets or aspects of giftedness (see Hong Kong Education Commission, 1990; Hong Kong Education Department, 2000). Notably, Gardner's notion of multiple intelligences appeals to educators as these intelligences bear directly on talent areas in the school setting (Armstrong, 1994; Campbell, Campbell, & Dickinson, 1999). According to MI theory, assessing giftedness in a student has to proceed with developing a profile of the student's intelligences, and psychometric and standardized measures may assess only a small part of the total spectrum of abilities. Based on this view, Gardner (1993) strongly supports the use of alternative assessment techniques, especially those involving performance-based assessment, to identify and evaluate student abilities and strengths with respect to multiple intelligences (see Chen & Gardner, 1997).

While the notion of multiple intelligences has gained increasing acceptance in the community of Hong Kong educators, questions have generally been raised regarding the viability of using self-report measures apart from the use of authentic and performance-based measures of multiple intelligences (Chan, 2001). With the development of the self-report scale, the Student Multiple Intelligences Profile (SMIP; Chan, 2001, 2003), the issue that needs to be addressed is whether the student perspective has been more privileged than other perspectives. Nonetheless, regarding different views and perspectives, it is believed that the information provided by parents, teachers, and peers on the spectrum of student abilities or multiple intelligences could be as revealing as self-reports in understanding the needs of gifted students and in comprehensive identification. In this connection, such information should not be overlooked (see Richert, 2003).

Specifically, parents are recognized to be good sources of information about their children's strengths, abilities, motivations, self-esteem, and creativity, especially when their children are young. They observe their children's behaviors in diverse situations, and could be well aware of their giftedness and needs that are frequently invisible to teachers (see Sabatella, 2003).

Regarding teachers, there were studies that indicated that teacher information did not provide reliable and useful information in addition to test scores from general intelligence tests (e.g., Gear, 1976, 1978; Pegnato & Birch, 1959). However, there were also other studies that attested to teachers' better identification of children's creativity (e.g., Kirschenbaum, 1983; Rimm, 1984), and of underachieving high-ability children, a group that might be missed by other means (e.g., Borland, 1978; Denton & Postlethwaite, 1984). In general, the results suggested that when teachers were trained and exposed to relevant information about students' characteristics that signal potential, they could provide useful information. Specifically, teachers could be sensitive to the multiple intelligences of students when they were exposed to a sufficient range of information about individual students (Guskin, Peng, & Simon, 1992).

While there were few studies focusing specifically on peer information, peer information has been regarded as useful. Peer information is especially helpful in finding students with leadership potential, as leaders usually emerge from and are recognized by peers, and is also useful in the area of creativity, as peers have a good basis for judging the imaginativeness and uniqueness of a fellow student's ideas (see Richert, 2003). Interestingly, with the diverse sources of information, the student as an expert in the knowledge about himself or herself provides a perspective that could not be matched by that of others. For example, Friedman, Jenkins-Friedman and Van Dyke (1984) reported a study in which outstanding leadership was assessed through nomination by self, peers, and teachers, and self-nomination was found to be the best single predictor.

While it is generally acknowledged that a comprehensive understanding of the needs and potential of gifted students requires an assessment that covers different aspects of giftedness and talents and incorporates multiple perspectives or multiple informants, there is little research on multiple perspectives, and no recent research on multiple informants in the gifted field. Against this background, this study aimed to assess gifted students' multiple intelligences from the perspectives of students, parents, teachers, and peers. Specifically, the profiles of student multiple intelligences derived from responses to the SMIP from students, mothers, fathers, teachers, and peers were examined and compared. In addition, the ability to predict students' perceived creativity and leadership using these profiles of multiple intelligences was explored.

Method

Participants

Gifted students, their parents, teachers, and peers were invited to participate in this study, which was a part of a larger research project investigating students' learning and emotional needs. First, over 1,200 invitations were sent out via mail to the gifted students registered with the government gifted education center for receiving educational services in enrichment program activities. These registered students have all been assessed prior to registration with IQ scores of 130 or above using the HK-WISC (Psychological Corporation, 1981) or Raven's SPM, the SPM scores being converted to IQ equivalents based on Hong Kong's norms (Hong Kong Education Department, 1986; Raven, Raven, & Court, 1998). Since the year 2000, the center has ceased registering students for services because of the change in government policy which no longer relies on a single IQ score to define giftedness and has been broadened to include, among others, creative and leadership giftedness (Hong Kong Education Department, 2000).

In accepting the invitation to participate in the project, each of these students had to request that his or her parents, one of his or her teachers, and one of his or her peers also agreed to participate to provide information about the student himself or herself. In addition, the designated teacher and peer had to be ones who knew the student well. All respondents were assured that participation was voluntary, and that the results of the assessment would be kept strictly confidential and for research purposes only.

Since not all registered students were active members of the center, initial responses were received from 448 students and their parents, 248 teachers, and 255 peers who indicated in their reply slips their willingness to participate to complete a questionnaire that would be subsequently sent to them via mail. As not all students, parents, teachers, and peers who initially agreed completed and returned the questionnaires, complete data were obtained from only 133 students, their mothers and fathers, teachers, and peers. These students were from grades 2 to 11 (92 boys and 41 girls), and were between the ages of 8 and 16 (M = 11.95, SD = 2.19). Ninety-five (71%) of the 133 responding teachers were currently teaching the target students, and 80 teachers (60%) knew the students as previous classroom teachers. The responding peers knew the target students for 1 to 13 years (M = 3.62, SD = 2.13). At the time of the study, 124 peers (93%) were from the same schools, 122 (92%) were from the same grades, and 84 (63%) were the target students' classmates.

Procedures

Each of the 133 students, their parents, designated teachers, and designated peers were separately sent a questionnaire that assessed the students' profiles of giftedness, talents or strengths in terms of multiple intelligences. Students were also requested to complete an additional self-report scale on their perceived creativity and leadership. All respondents received their questionnaires with the instructions that they were not to discuss their knowledge about the students concerned with other respondents, as the study valued independent views from varying perspectives, which were expected to be different. After completing the questionnaires, respondents were requested to mail back their questionnaires in separate pre-addressed envelopes. To ensure confidentiality of the students' information, data were collected by and with the help of the staff of the government gifted education center. Specifically, apart from the names of the students concerned, no other information, such as the addresses of the students or the names of their parents, their teachers or their peers, was made available. The only linkage among the set of questionnaires for a specific student was the name of the particular student.

Measures

Two instruments in Chinese, the SMIP (Chan, 2001), and a self-report scale assessing creativity and leadership were employed in this study. The SMIP is a 24-item checklist of characteristics and behaviors constructed to reflect perceived strengths in Gardner's (1999) multiple intelligences. The original 21-item SMIP was designed to assess students' seven intelligences (three items for each intelligence), that is, verbal-linguistic, musical, logical-mathematical, visual-spatial, bodily-kinesthetic, intrapersonal, and interpersonal intelligences. In the revised SMIP, three items have been added to incorporate the addition of naturalist intelligence (see Chan, 2003). The items of the SMIP are reproduced in the appendix of Chan (2001). For the purposes of this study, the self-report SMIP was modified to yield three additional versions for ratings by parents (mothers and fathers), by teachers, and by peers through changing the instructions from "I" to "My Child," "My Student," and "My Friend or Schoolmate, " respectively.

The brief self-report scale assessing creativity and leadership consisted of six items. The three creativity items translated into English are: "I can suggest many different solutions to problems;" "I can generate many different ideas such as ways to use a particular object;" and "I can think of unusual ideas or fashion products that would be regarded as novel by others." The three leadership items translated into English are: "I realize I can lead when I participate in group activities;" "I work to promote group activities to achieve the goal of the group;" and "Friends and classmates often elect me to lead them in accomplishing different tasks."

In completing the SMIP, respondents were requested to rate the students or themselves (in the case of self-report) on the 24 items using a five-point scale ranging from 1 (least descriptive) to 5 (most descriptive). The SMIP can be scored into eight scales that yield eight scores reflecting the eight intelligences. The self-report form of the SMIP has been used in studies with Chinese students and has demonstrated sound psychometric properties, with reliabilities ranging from .57 to .77 (see Chan, 2001, 2003), and construct validation using confirmatory factor analysis (see Chan, 2001). Students were also requested to complete the self-report scale on creativity and leadership using the same five-point rating scale, which can be scored as two subscales that yielded two scores reflecting creativity and leadership. This brief scale has been used in other pilot unpublished studies with over 400 Chinese students, demonstrating reasonably high internal consistency (creativity [alpha] = .78; leadership [alpha] = .79).

Results

To assess the profile of talents or strengths of the 133 gifted students, ratings on SMIP items from student self-reports and from responses of parents, teachers and peers were tabulated, and the 24 item responses were scored into eight scales representing the eight intelligences rated by students, mothers, fathers, teachers, and peers. Table 1 shows the mean ratings of the eight intelligences from the five perspectives, and the internal consistency measures of the eight intelligences scored as eight scales from the five perspectives. These eight scales had moderate to high internal consistency as reflected in the values of Cronbach's [alpha] (.50 to .94).

It can be seen from Table 1 that, in general, students' logical-mathematical intelligence received relatively higher ratings than the other seven intelligences across the five perspectives, as would be expected since students were selected on the basis of conventional IQ measures such as the HK-WISC and SPM that emphasize verbal and logical-mathematical abilities, whereas bodily-kinesthetic intelligence and naturalist intelligence received relatively lower ratings. Across different perspectives, students appeared to rate their interpersonal and musical intelligences higher than judges in other perspectives. Parents, teachers, and peers tended to rate the students' verbal-linguistic intelligence more highly than did the students. However, parents differed from teachers and peers in that they tended to rate the students' visual-spatial intelligence more highly than did teachers, peers, and students, whereas teachers and peers tended to rate the students' intrapersonal intelligence more highly than did parents and students.

Gender and Grade Level Differences in Student Multiple Intelligences

To explore whether there were gender or grade level (primary vs. secondary) differences on student self-ratings on the eight intelligences, a 2 by 2 (Gender by Grade Level) multivariate analyses of variance (MANOVA) was performed on the eight intelligences scores as dependent measures. The results indicated that the overall Gender main effect was significant, Wilks's [LAMBDA] = .86, F (8, 122) = 2.42, [[eta].sup.2] =. 14, p < .05, but the overall main effect of Grade Level, Wilks's [LAMBDA] = .90, F (8, 122) = 1.69, [[eta].sup.2] = .10, and the interaction effect of Gender by Grade Level, Wilks's [LAMBDA] = .98, F (8, 122) = 0.38, [[eta].sup.2] = .02, were nonsignificant (p > .05). Subsequent univariate ANOVA on each of the intelligences scores was conducted as a follow-up test to the significant MANOVA Gender main effect. Using the Bonferroni procedure to adjust for multiple tests, each ANOVA was evaluated at the .05/8 or .00625 level, and the results indicated that the differences between boys and girls in their self-ratings did not achieve statistical significance on any one of the eight intelligences.

To explore whether the gender or grade level (primary vs. secondary) of the students would affect the ratings by different judges on the eight intelligences, the above procedures for self-ratings were repeated for ratings by mothers, fathers, teachers, and peers as dependent measures in four separate MANOVAs followed up by ANOVAs on significant main or interaction effects. On the one hand, significant overall Gender main effect was indicated for all four perspectives. The specific results for ratings

by mothers, fathers, teachers, and peers yielded, respectively, Wilks's As (.81, .88, .84, and .88), F-values at df = 8/122 (3.60, p < .001; 2.18, p < .05; 2.90, p < .005; and 2.03, p < .05), and effect size indices [[eta].sup.2] (.19, .13, .16, and .12). On the other hand, significant overall main effect of Grade Level was indicated for only the mother and teacher perspectives. The specific results for ratings by mothers, fathers, teachers, and peers yielded, respectively, Wilks's [LAMBDA]s (.88, .93, .87, and .93), F-values at df = 8/122 (2.13, p < .05; 1.13, p > .05; 2.28, p < .05; and 1.20, p > .05), and effect size indices [[eta].sup.2] (.12, .07, .13, and .07). In contrast, none of the four perspectives yielded significant overall Gender/Grade-Level interaction effect (p > .05). The specific results for ratings by mothers, fathers, teachers, and peers yielded, respectively, Wilks's [LAMBDA]s (.94, .94, .96, and .98), F-values at df = 8/122 (1.00, .92, .58, and .35), and effect size indices [[eta].sup.2] (.06, .06, .04, and .02).

Thus, like the analysis with student self-ratings, significant Gender main effect was indicated relatively consistently for all four perspectives, whereas, unlike the analysis with student self ratings, significant main effect of Grade Level was also indicated, but only for the mother and teacher perspectives. Despite these differences, follow-up ANOVAs on the eight intelligences evaluated at the adjusted significance level using the Bonferroni procedure yielded only one significant gender difference, F (1, 129) = 15.17, p < .001, [[eta].sup.2] =.11, suggesting that mothers tended to give higher ratings on verbal-linguistic intelligence to girls (M = 11.56, SD = 2.31) than to boys (M = 9.83, SD = 2.43). In summary, while there appeared to be consistent gender differences in ratings on the eight intelligences across five perspectives, and some differences in ratings by mothers and teachers for students of different grade levels, strictly speaking, only one of these differences achieved statistical significance. On this basis, it could be suggested that self-ratings and ratings by parents, teachers, and peers were made relatively independent of the gender and grade levels of the students.

Comparing Levels of Student Multiple Intelligences Within Different Perspectives

To evaluate the effects of different judges (Perspectives) and of different intelligence domains (Talents), a two-way within-subjects ANOVA was conducted on ratings of the eight intelligences as the dependent variables. The within-subjects factors were Perspectives with five levels (self, mother, father, teacher, and peer), and Talents with eight dimensions (the eight intelligences). The multivariate criterion of Wilks's [LAMBDA] was employed to test the main and interaction effects. Specifically, the main effect of Perspectives was nonsignificant, Wilks's [LAMBDA] = .95, F (4, 129) = 1.57, [[eta].sup.2] = .05, p > .(15. The main effect of Talents, Wilks's [LAMBDA] = .43, F (7, 126) = 24.12, [[eta].sup.2] = .57, p < .001, and the interaction effect of Perspectives by Talents, Wilks's [LAMBDA] = .28, F (28, 105) = 9.89, [[eta].sup.2] = .73, p < .001, were significant.

Because the interaction effect of Perspectives by Talents was significant with a moderately large effect size, further tests were conducted to follow up on the significant interaction rather than the main effect of Talents. Specifically, follow-up one-way within-subjects ANOVAs were conducted to test the simple main effect of Talents for each perspective. The results indicated that the eight intelligences were rated significantly differently within each perspective. The specific results for ratings by students, mothers, fathers, teachers, and peers yielded, respectively, Wilks's [LAMBDA] (.73, .54, .44, .53, and .41), F-values at df = 7/126 (6.56, 15.52, 23.08, 16.09, and 25.85, all p < .001), and effect size indices [[eta].sup.2] (.27, .46, .56, .47, and .59). Further follow-up paired t-tests within each perspective for all 28 possible pairs of differences in intelligences scores, evaluated at .05128 or .00178 level of significance using the Bonferroni procedure, indicated that 6, 12, 17, 16, and 14 pairs out of the 28 pairs were significantly different in ratings by students, mothers, fathers, teachers, and peers, respectively, suggesting that judges distinguished among the intelligences and rated the target students' multiple intelligences differently.

Comparing Ratings of Student Multiple Intelligences From Different Perspectives

With a significant Perspectives-Talents interaction effect, it was of great interest to compare the multiple intelligences scores by different judges from the five perspectives, despite the lack of significant Perspectives main effect. Consequently, follow-up one-way within-subjects ANOVAs were also conducted to test the simple main effect of Perspectives for the intelligences. The results for ratings on verbal-linguistic, musical, logical-mathematical, visual-spatial, bodily-kinesthetic, intrapersonal, interpersonal, and naturalist intelligences yielded, respectively, Wilks's [LAMBDA]s (.90, .80, .93, .82, .90, .84, .92, and .71), F-values at df = 4/129 (3.70, p < .01 ; 8.13, p < .001 ; 2.60, p < .05; 6.87, p < .001 ; 3.62, p < .01; 6.18, p < .001; 3.01, p < .05; and 13.36, p < .001), and effect size indices [[eta].sup.2] (.10, .20, .07, .18, .10, .16, .09, and .29). Thus, the judges in the five perspectives gave significantly different ratings when each intelligence was considered separately. Further follow-up paired t-tests within each intelligence for all 10 possible pairs of differences in intelligences scores, evaluated at .05/10 or .005 level of significance using the Bonferroni procedure, indicated that 2, 5, 1, 2, 1, 4, l, and 5 pairs out of the 10 pairs were significantly different in ratings on verbal-linguistic, musical, logical-mathematical, visual-spatial, bodily-kinesthetic, intrapersonal, interpersonal, and naturalist intelligences, respectively. Thus, greater variability among perspectives was indicated for musical, intrapersonal, and naturalist intelligences.

Comparing Typical Profiles of Student Multiple Intelligences From Different Perspectives

Shifting the focus from individual intelligences scores to the profiles of eight intelligences scores, it was of interest to examine more closely the profile similarities and differences of the five perspectives. To this end, three indices of profile similarity were computed based on the average or typical profiles of the five perspectives (see Skinner, 1978). The profile elevation index referred to the average or mean ratings across the eight intelligences for the typical judge of a particular perspective. The profile scatter index referred to the dispersion or standard deviation of ratings of the eight intelligences for the typical profile of a particular perspective. The profile shape index referred to the correlation between pairs of typical profiles of mean ratings on the eight intelligences, with higher correlation suggesting greater similarity in profile shape between the two profiles of scores. The three indices to describe the five typical profiles of multiple intelligences representing the five perspectives are summarized in Table 2.

It can be seen from Table 2 that the student-perspective profile had the highest elevation with corresponding smallest scatter, and the peer-perspective profile yielded the lowest elevation with corresponding largest scatter, largely because peers tended to rate the naturalist intelligence of the target students much lower than they did the other seven intelligences. The profile shapes of the father and mother perspectives were most similar, whereas the profile shapes of the student and peer perspectives were least similar. Assuming that the student-perspective profile represented the unique pattern of self-perceived multiple intelligences and needs of students, the three profile similarity indices suggested that mothers could be most sensitive to the multiple intelligences and needs of students, (mothers' ratings on students' multiple intelligences being closest to students' self-ratings), followed by lathers, teachers, and peers. It is noteworthy that among the five typical profiles, the student profile and the peer profile were most dissimilar in terms of elevation, scatter, and shape indices.

Multiple Intelligences and Student Perceived Creativity and Leadership

Because creativity and leadership have been areas for training in government gifted programming in Hong Kong, it was of interest to examine students' self-perceived creativity and leadership, and see whether these self-perceptions could be predicted from ratings of multiple intelligences from the different perspectives. With this view, the relevant item responses of the 133 students were scored into the two subscales of creativity (M = 11.73, SD = 2.26) and leadership (M = 11.18, SD = 2.54). The two subscales correlated moderately and significantly (r = .45, p < .001), and both achieved moderately high internal consistency as measured by Cronbach's [alpha] (Creativity [alpha] = .78; Leadership [alpha] = .76).

To explore the relationship between self-perceived creativity and leadership and the eight intelligences from different perspectives, two sets of five multiple linear regression analyses (one for each perspective) were performed using scores on creativity and leadership as two separate criteria and the eight intelligence scores as predictors. It is believed that using the whole set of eight intelligences as predictors could help delineate whether certain intelligences expected to be good or poor predictors would indeed turn out to be convergent or discriminant measures of the criterion. The results of the regression analyses are presented in Table 3.

It can be seen from Table 3 that self-perceived creativity and leadership were most predictable from students' self-ratings on the eight intelligences, as about 40% of the variance of the two criterion measures could be accounted for by using the eight intelligences as predictors. Notably, creativity was significantly predicted by intelligences from the traditional cognitive domains (verbal-linguistic and logical-mathematical intelligences) and visual-spatial intelligence, whereas leadership was significantly predicted by interpersonal and naturalist intelligences. Multiple intelligences as rated by parents, teachers, and peers, however, did not significantly predict students' self-perceived creativity, but ratings by fathers and teachers did significantly predict students' self-perceived leadership with verbal-linguistic intelligence emerging as the significant predictor.

Discussion

The education system in Hong Kong has for many years emphasized verbal-linguistic and logical-mathematical intelligences in the formal curricula, thus bypassing, undervaluing, or underidentifying students with dominant abilities or giftedness in other intelligences. With the increasing acceptance of the notion of multiple intelligences among educators in Hong Kong, the education system needs to adapt to meet various individual differences for maximum intellectual gains (Kornhaber, Krechevsky, & Gardner, 1990; Waiters & Gardner, 1986).

Despite the lack of a precise mapping of intelligences on disciplines of study or talent domains, MI theory helps focus on students who might be outstanding in different talent areas such as art, music, mathematics, science, and other esthetic or nonacademic domains. Interestingly, some educators have even associated the development of multiple intelligences with the Chinese conceptualization of nurturing children in five domains of ethics, intellect, physique, social skills, and esthetics (de, zhi, ti, qun, and mei) to achieve educating the whole person, thus yielding a balanced development in children (see Chan, 2000).

This study employed the SMIP not only as a self-report checklist to assess the eight intelligences but also as a checklist for ratings by parents, teachers, and peers. Because target students for self-reports and ratings by judges in this study were all identified as gifted using conventional criterion of high IQ scores, the ratings of multiple intelligences obtained from the five perspectives could be regarded as giving voices to the different perspectives in describing giftedness in the sample of Hong Kong gifted students. This view is in line with the social constructionist notion of respecting different perspectives, while acknowledging that gifted students have expert knowledge about themselves. Indeed, the different perspectives might exercise influence on or jointly construct the lived experiences of gifted students.

As expected, target students' multiple intelligences were evidently rated as at different levels by judges from different perspectives. However, regardless of perspectives, logical-mathematical intelligence received in general the highest ratings whereas bodily-kinesthetic and naturalist intelligences received the lowest. What is perhaps more interesting is that the profiles of multiple intelligences from the five perspectives did have subtle differences. Unlike parents, teachers, and peers, the gifted students did not rated themselves highly on verbal-linguistic intelligence. Instead, gifted students gave themselves higher ratings on their interpersonal intelligence and their musical intelligence. Further, gifted students reported a typical profile that was more elevated and less scattered than the typical profiles perceived by other judges, suggesting that they might have a more inflated but less differentiated picture of their strengths and weaknesses when compared with typical profiles described by other perspectives.

On the other hand, the typical profiles based on ratings from mothers and fathers were more similar to each other than to the typical profiles of other perspectives, perhaps because mothers and fathers shared the same environments for interactions with their children. The parent perspectives emphasized that parents focused on students' high academic abilities as reflected in the conventional verbal-linguistic, logical-mathematical, and visual-spatial intelligences. In contrast, the typical profiles based on teacher and peer ratings were more similar to each other than to those based on parent ratings, again perhaps because teachers and peers shared the same school environments with the target students. Apart from rating highly the conventional verbal-linguistic and logical-mathematical intelligences, teachers and peers did appreciate the rich self-knowledge that students had about themselves in intrapersonal intelligence. Nonetheless, while students, their parents, teachers, and peers might make judgments that are far from being convergent, the information from different sources could be complementary in helping educators understand better the profiles of strengths and needs of gifted students. Future studies might focus on how communication of the various perspectives could affect gifted students' learning and talent development.

Putting aside the subtle differences in the profiles of multiple intelligences based on different perspectives, the consistent elevations and depressions of different intelligences across perspectives might truly indicate the relative strengths and weaknesses of this sample of gifted students, who were selected by test scores reflecting verbal and mathematical abilities. Viewed slightly differently, these consistent ratings might also reflect the valuing or emphasis of the Hong Kong school system on verbal and mathematical abilities and academic achievement, and the comparatively de-emphasis on physical or athletic activities and the appreciation of nature and natural environment. Thus, these findings also contributed to a better understanding of the pattern of giftedness as expected in a sample of Chinese students with high scores in traditional intelligence measures.

Taking seriously the notion that each student, including a gifted student, has a unique profile of strengths, weaknesses, and needs, the findings of this study highlighted that a student's profile might be perceived differently by judges from different perspectives. While acknowledging that all perspectives should be respected and they could be complementary, it is recognized that students do have expert knowledge about themselves and the student perspective might have more meaning for students. Contrary to the conjecture that gifted students might be more vulnerable to having problems related to their interpersonal and intrapersonal intelligences, gifted students in this sample judged themselves favorably in interpersonal intelligence but somewhat poorly in intrapersonal intelligence, suggesting that they might have problems in emotional regulation and self-appraisal. Accordingly, parents and teachers need to strive to be more sensitive to the student-perspective profile, helping students recognize their strong and less developed intelligences, and explore how they might develop their individual intelligences and use different ways of knowing in their formal schooling and daily living.

Admittedly, this sample of gifted students represented only about 11% of the total number of gifted students registered with the government up to the year 2000, and can hardly be claimed to represent gifted students as traditionally defined in Hong Kong. Although this study was intended to study all registered students, not all registered students were actively involved in the center's activities, and not all of them could locate parents, teachers, and peers who agreed to participate. The attrition was inevitably large. In this connection, the small sample size could be regarded as a limitation of the study, especially when one considers the large number of multiple tests performed.

In predicting students' perceived creativity and leadership using ratings on the multiple intelligences, the students' self-ratings yielded the best predictors. Specifically, the conventional intelligences (verbal-linguistic, logical-mathematical, and visual-spatial intelligences) emerged as significant predictors for creativity, whereas interpersonal and naturalist intelligences emerged as significant predictors for leadership. These results suggested that gifted students who rated themselves highly on the conventional intelligences also perceived themselves as more creative and capable of divergent thinking, and those who rated themselves highly on interpersonal and naturalist intelligences also perceived themselves as having leadership skills in team work.

Somewhat unexpectedly, self-perceived creativity and leadership were not particularly predictable using the profiles of multiple intelligences from the different perspectives other than the student perspective. Assuming that parents, teachers, and peers could have good opportunities to observe students' creative behaviors and products as well as leadership behaviors and skills in a variety of situations in the home and school settings, information from these perspectives should be valuable. One plausible reason for the poor prediction is that self-perception might not correspond very well with performance and behavioral measures. Thus, future studies might consider students' creative products, experiences in leading or managerial positions in school clubs and societies, or other evidence of creative productivity and leadership as objective, performance-based, or product-based measures for prediction.

Table 1
Mean Ratings on
Students' Multiple Intelligences (N= 133)

                               Judgment Ratings (Perspective)

Talent (Intelligence)  Student    Mother    Father   Teacher     Peer

Verbal-Linguistic
  M                      9.74     10.36     10.38     10.41     10.12
  SD                     2.35      2.51      2.60      2.69      2.64
  [alpha]                0.56      0.53      0.61      0.73      0.66
Musical
  M                     10.38      9.94      9.85      8.80      9.13
  SD                     3.17      3.33      3.31      3.13      3.14
  [alpha]                0.80      0.84      0.85      0.88      0.76
Logical-Mathematical
  M                     10.68     10.94     11.05     10.79     10.21
  SD                     2.47      2.45      2.54      2.87      2.64
  [alpha]                0.50      0.54      0.62      0.80      0.60
Visual-Spatial
  M                      9.98     10.65     10.53      9.97      9.46
  SD                     2.84      2.56      2.55      2.55      2.65
  [alpha]                0.68      0.61      0.64      0.79      0.69
Bodily-Kinesthetic
  M                      9.38      8.62      8.50      8.94      9.14
  SD                     2.62      2.91      2.60      2.95      2.34
  [alpha]                0.69      0.69      0.68      0.85      0.50
Intrapersonal
  M                      9.90      9.68      9.38     10.35     10.79
  SD                     2.52      2.95      3.00      3.38      2.86
  [alpha]                0.72      0.82      0.83      0.91      0.77
Interpersonal
  M                     10.46     10.00      9.53     10.16     10.02
  SD                     2.52      2.93      2.91      3.33      3.37
  [alpha]                0.76      0.85      0.87      0.94      0.89
Naturalist
  M                      9.67      9.14      8.56      9.02      7.50
  SD                     3.32      3.50      3.32      2.87      2.98
  [alpha]                0.77      0.87      0.86      0.90      0.84

Note. [alpha] is Cronbach's alpha to assess internal consistency of
the scales.

Table 2
Multiple Intelligences
Profile Characteristics of
Students From Five Perspectives

                          Perspective of Multiple Intelligences Profile

Profile Similarity Index     Student  Mother   Father   Teacher   Peer

Elevation (M)                 10.02    9.92     9.72     9.81     9.55
Scatter (SD)                   0.45    0.77     0.92     0.77     1.01
Shape
  Student
  Mother                        .71
  Father                        .65     .96 **
  Teacher                       .43     .73 *    .69
  Peer                          .37     .50      .54      .78 *

Note. Profile elevation and scatter as indices of profile similarity
refer to the mean ratings and standard deviations across the eight
intelligences. Profile shape as an index of profile similarity
refers to the correlation between pairs of profile scores of
multiple intelligences.

* p<.05; p<.01(2-tailed).

Table 3
Regression Analysis for the Prediction of
Students' Self-Perceived Creativity and
Leadership Using the Eight Intelligences
from Different Perspectives

                                    Multiple

Criterion                  [R.sup.2]        F(8, 124)

Creativity
Student perspective           .41           10.92 ***

Mother perspective            .08            1.32
Father perspective            .12            2.17 *
Teacher perspective           .09            1.59
Peer perspective              .03            0.47

Leadership
Student perspective           .40           10.17 ***

Mother perspective            .10            1.71
Father perspective            .12            2.15 *
Teacher perspective           .19            3.68 ***
Peer perspective              .10            1.70

                                    Significant

Criterion                  predictor            t

Creativity
Student perspective   Verbal-linguistic     2.24 *
                      Logical-mathematical  2.55 *
                      Visual-spatial        2.82 **
Mother perspective    --
Father perspective    --
Teacher perspective   --
Peer perspective      --

Leadership
Student perspective   Interpersonal         2.00 *
                      Naturalist            2.63 **
Mother perspective    --
Father perspective    Verbal-linguistic     2.64 **
Teacher perspective   Verbal-linguistic     2.51 *
Peer perspective      --

* p < .05, ** p < .01, *** p < 0.001

REFERENCES

Armstrong, T. (1994). Multiple intelligences in the class room. Alexandria, VA: Association for Supervision and Curriculum Development.

Borland, J. H. (1978). Teacher identification of the gifted: A new look. Journal for the Education of the Gifted, 2, 22-32.

Callahan. C. M. (1996). A critical self study of gifted education: Healthy practice, necessary evil, or sedition? Journal for the Education of the Gifted, 19. 148-163.

Campbell, L. Campbell, B., & Dickinson, D. (1999). Teaching and learning through multiple intelligences (2nd ed.). Boston: Allyn and Bacon.

Chan, D. W. (2000). Learning and teaching through the multiple intelligences perspective: Implications for curriculum reform in Hong Kong. Educational Research Journal. 15, 187-201.

Chan, D. W. (2001). Assessing giftedness of Chinese secondary students in Hong Kong: A multiple intelligences perspective. High Ability Studies, 12, 215-234.

Chan, D. W. (2003). Adjustment problems and multiple intelligences among gifted students in Hong Kong: The development of the revised Student Adjustment Problems Inventory. High Ability Studies, 14, 41-54.

Chen, J., & Gardner, H. (1997). Alternative assessment from a multiple intelligences theoretical perspective. In D. P. Flanagan. J. L. Genshaft. & P. L. Harrison (Eds.), Contemporary intellectual assessment: Theories, tests, and issues (pp. 105-121). New York: Guilford Press.

Denton, C., & Postlethwaite, K. (1984). A study of the effectiveness of teacher-based identification of pupils with high ability in the secondary school. Gifted Education International, 2, 100-106.

Fasko. D. Jr. (2001). An analysis of multiple intelligences theory and its use with the gifted and talented. Roeper Review, 23, 126-130.

Friedman. P. G., Jenkins-Friedman, R., & Van Dyke, M. (1984). Identifying the leadership gifted: Self, peer, or teacher nominations? Roeper Review. 7, 91-94.

Gardner, H. (1983). Frames of mind: The theory of multiple intelligences. New York: BasicBooks.

Gardner, H. (1993). Multiple intelligences: The theory in practice. New York: BasicBooks.

Gardner, n. (1999). Intelligence reframed. New York: BasicBooks.

Gear. G. (1976). Accuracy of teacher judgment in identifying intellectually gifted children. Gifted Child Quarterly, 20, 478-489.

Gear. G. (1978). Effects of training on teachers' accuracy in the identification of gifted children, Gifted Child Quarterly, 22. 90 97.

Guilford, J. P. (1967). The nature of human intelligence. New York: McGraw-Hall.

Guskin, S. L., Peng, C. Y. J., & Simon, M. (1992). Do teachers react to "Multiple Intelligences"? Effects of teachers' stereotypes on judgments and expectancies for students with diverse patterns of giftedness/talent. Gifted Child Quarterly. 36, 32-37.

Hong Kong Education Commission. (1990). Education Commission Report No. 4. Hong Kong: Hong Kong Government.

Hong Kong Education Department. (1986). Hong Kong supplement to guide to the Standard Progressive Matrices. Hong Kong: Hong Kong Government.

Hong Kong Education Department. (2000). Gifted education in Hong Kong. Hong Kong: Special Educational Needs Section, Curriculum Development Institute, Hong Kong.

Karolyi, C. V., Ramos-Ford, V., & Gardner, n. (2003). Multiple intelligences: A perspective on giftedness. In N. Colangelo & G. A. David (Eds.). Handbook of gifted education (3rd ed., pp. 100-112). Boston. MA: Allyn and Bacon.

Kirschenbaum. R. J. (1983). Let's cut out the cut off score in the identification of the gifted. Roeper Review, 5, 6-9.

Kornhaber. M., Krechevsky, M., & Gardner. H. (1990), Engaging intelligence, Educational Psychologist, 25, 177-199.

Maker, C. J., Nielson, A. B., & Rogers. J. A. (1994), Giftedness, diversity, and problem-solving. Teaching Exceptional Children, 27. 4-19.

Pegnato, C. W., & Birch. J. W. (1959). Locating gifted children in junior high schools: A comparison of methods. Exceptional Children, 25, 300-304.

Pfeiffer, S. I. (2003). Challenges and opportunities for students who are gifted: What the experts say. Gifted Child Quarterly, 47, 161-169.

Piirto, J. (1999). Talented children and adults: Their development and education (2nd ed.). New York: Macmillan.

Psychological Corporation. (1981). Hong Kong Wechsler Intelligence Scale for Children manual. New York: Author.

Raven, J., Raven, J. C., & Court, J. H. (1998). Manual for Raven's Progressive Matrices and vocabulary scale. Oxford, England: Oxford Psychologists Press.

Renzullk J. S. (1978). What makes giftedness? Reexamining a definition. Phi Delta Kappan, 60, 18-24.

Renzulli, J. S. (1986). The three-ring conception of giftedness: A developmental model for creative productivity. In R. J. Sternberg & J. E. Davidson (Eds.), Conceptions of giftedness (pp. 53-92). New York: Cambridge University Press.

Richert. E. S. (2003). Excellence with justice in identification and programming. In N. Colangelo & G. A. David (Eds.). Handbook of gifted education (3rd ed., pp, 146-158). Boston, MA: Allyn and Bacon.

Rimm, S. B. (1984). The characteristics approach: Identification and beyond. Gifted Child Quarterly, 28, 181-187.

Sabatella. M. L. (2003). Role of programs: Relationships with parents, schools, and communities. In J. F. Smutny Ed.), Designing and developing programs for gifted students (pp. 119-128). Thousand Oaks, CA: Corwin Press.

Sarouphim, K. M. (1999). DISCOVER: A promising alternative assessment for the identification of gifted minorities. Gifted Child Quarterly. 43, 244-251.

Skinner, H. A. (1978). Differentiating the contribution of elevation, scatter and shape in profile similarity. Educational and Psychological Measurement, 38, 297-308.

Sternberg, R. J. (1986). A triarchic theory of intellectual giftedness. In R. J. Sternberg & J. E. Davidson (Eds.). Conceptions of giftedness (pp. 223-243), Cambridge, MA: Cambridge University Press.

Sternberg, R. J. (1997). A triarchic view of giftedness: Theory and practice. In N. Colangelo & G. A. Davis (Eds.), Handbook of gifted education (2nd ed., pp. 43-53). Boston: Allyn and Bacon.

Sternberg, R. J. (2000). Patterns of giftedness: A triarchic analysis. Roeper Review, 22. 231-235.

Tannenbaum, A. J. (1983). Gifted children: Psychological and educational perspectives. New York: Macmillan.

Tannenbaum, A. J. (1997). The meaning and making of giftedness. In N. Colangelo & G. A. Davis (Eds.), Handbook of gifted education (2nd ed., pp. 27-42). Boston: Allyn and Bacon.

Terman, L. M. (1925). Genetic studies of genius (Vol. 1). Stanford, CA: Stanford University Press.

Walters, J., & Gardner, H. (1986). The theory of multiple intelligences: Some issues and answers. In R. J. Sternberg & R. Wagner (Eds.), Practical intelligences (pp. 163-182). New York: Cambridge University Press.

Winner, E. (2000). Giftedness: Current theory and research. Current Directions in Psychological Science, 9, 153-156.

This study was supported in part by a Hong Kong Quality Education Fund for Research. I thank Ms. L. K. Chan and Mr. K. K. Yung of the government gifted education center for assistance in data collection.

Manuscript submitted December 3, 2003. Revision accepted April 19, 2004.

David W. Chan is professor in the Department of Educational Psychology and Program Supervisor of the Programs for the Gifted and Talented at the Chinese University of Hong Kong. His research interests include stress and coping, psychopathology and health, creativity and intelligence, and gifted education and talent development.

E-mail: [e-mail]