Gender differences in spatial ability: relationship to spatial experience among Chinese gifted students in Hong Kong.

Conceptions of giftedness and talent in gifted education have shifted from an emphasis on general cognitive ability (general intelligence or "g") to an appreciation of the unique information afforded by multiple abilities (Colangelo & Davis, 2003; Davis & Rimm, 2004; Winner, 1996). Indeed, some of the more recent conceptions have gained popularity among educators and the general public. They include the notions that intelligence encompasses practical ability (Sternberg, Wagner, Williams, & Horvath, 1995), musical and bodily-kinesthetic abilities (Gardner, 1983, 1999), and emotional capabilities (Goleman, 1995), and that these abilities could even be regarded as independent intelligences. Yet, few of these abilities are assessed in the most commonly used standardized intelligence tests.

The notion of multiple abilities is not novel in the history of psychometric intelligence testing. Vernon (1961), for example, sought to interpret intellectual performances in terms of a hierarchical group-factor model, in which "g" represents the general factor under which major group factors, minor group factors, and specific factors are organized. The major group factors are the verbal-numerical-educational (v: ed) factors and the practical-mechanical-spatial-physical (k: m) factors. These broad factors are, in turn, subdivided into minor group factors: The v: ed factor divides into verbal and number group components, and the k: m factor divides into mechanical information, spatial information, and manual group factors. Vernon's conceptualization parallels the work of Wechsler (1958), who designed an intelligence test with subtests categorized by Verbal and Performance sections. The more recent fourth edition of the Wechsler Intelligence Scale for Children (Wechsler, 2003) has subtests related to index scores that include the verbal comprehension index, perceptual reasoning index, working memory index, and processing speed index. Support for a hierarchical factor model is also provided in the large-scale factor analysis study of 460 data sets by Carroll (1993) who identified three strata of cognitive abilities. The highest stratum is a factor of general intelligence, often designated by "g"; the second stratum consists of broad factors that include fluid intelligence, crystallized intelligence, general memory and learning, broad visual perception, broad auditory perception, broad retrieval ability, broad cognitive speediness, and processing speed; and the third stratum consists of narrow abilities related to the broad areas of the second stratum.

In providing educational opportunities commensurate with students' level and pattern of abilities, the focus has often been on students' verbal and quantitative aptitudes (v: ed factor), given that the standard educational curriculum requires students to operate in learning environments that stress reasoning with numbers and words. Consequently, students with strong mechanical and spatial abilities (k: m factor) but relatively modest quantitative and verbal reasoning abilities might be deprived of opportunities for their talent development, and could be at risk for academic underachievement (Gohm, Humphreys, & Yao, 1998; Humphreys & Lubinski, 1996; Humphreys, Lubinski, & Yao, 1993; Shea, Lubinski, & Benbow, 2001).

With the increasing realization that talent search procedures might have overlooked an underappreciated dimension of cognitive functioning--notably spatial ability in students--numerous studies have started to address the need to assess spatial ability, and to identify spatially talented students whose gifts involve reasoning with figures and shapes; the kind of reasoning that characterizes architecture, engineering, the physical sciences, and many of the creative arts (Gohm et al., 1998; Humphreys et al., 1993; Humphreys & Lubinski, 1996). Since spatial ability is normally not specifically assessed by most standardized tests employed in talent search, a spatial ability instrument needs to be included in the talent search procedures. This addition would help provide a more comprehensive profile of students' cognitive abilities that could help inform academic and career planning.

Broadly defined, spatial ability can be described as the ability to represent and transform symbolic or nonlinguistic information through space. There are many spatial ability tasks or measures available for assessing students (Bennett, Seashore, & Wesman, 1974; Carroll, 1993; Stumpf & Eliot, 1999). In a large-scale meta-analysis, Linn and Petersen (1985) distinguished three qualitatively different types of spatial ability: spatial perception, mental rotation, and spatial visualization. Spatial perception requires participants to locate the horizontal or the vertical in a stationary display while ignoring distracting information. A good example is the rod and frame task in which the participant confronts a rod and a rectangular frame that are tilted (nothing else being visible), and has to adjust the rod so that it appears vertical to him or her. Mental rotation involves the ability to imagine how objects will appear when they are rotated in two- or three-dimensional space. Spatial visualization refers to complex, analytic multistep processing of spatial information. Typical tasks include the embedded figures test and paper folding. To these three types, Halpern and LaMay (2000) added two other distinct types: spatiotemporal ability and the generation and maintenance of a spatial image. The former involves judgments about the responses to dynamic or moving visual displays, and the latter requires participants to generate an image such as the shape of a particular letter of the alphabet, and then use the information in the image to perform a specified cognitive task.

From a slightly different perspective, Stumpf and Eliot (1995, 1999) classified a collection of extant figural spatial tests into 16 categories within the two divisions, recognition and manipulation. Recognition tests require the perception, retention, and transformation of visual forms within a two-dimensional plane, whereas manipulation tests require the mental manipulation of visual shapes across a plane. The eight recognition categories include perceptual speed tasks, copying tasks, maze tasks, figure-ground tasks, visual memory tasks, Gestalt resolution tasks, paper form board tasks, and figural rotation tasks. The eight manipulation categories include block tasks, intersection tasks, external block rotation, internal block rotation, paper folding tasks, pattern assembly tasks, surface development tasks, and perspective tasks. In reviewing these spatial ability tasks, Stumpf and Eliot (1999) concluded that the structure underlying performance on figural-spatial tests can largely be described by four facets: dimensionality, mental rotation, memory load, and speededness. In this connection, it is no surprise that mental rotation has often been regarded as a measure representing general spatial reasoning ability (Casey, Nuttall, Pezaris, & Benbow, 1995).

In using mental rotation tasks to represent spatial ability, men show a consistent and substantial advantage over women (Linn & Petersen, 1985; Stumpf & Eliot, 1995; D. Voyer, S. Voyer, & Bryden, 1995), leading to the conclusion by some researchers that mental rotation is one of the two loci of gender differences in spatial ability (Karl, Carter, & Pellegrino, 1979); the other locus being visualization of perspectives (Stumpf & Eliot). In contrast, women have only a small advantage over men on visual memory tasks (Stumpf & Eliot).

To explain these gender differences from an evolutionary psychology perspective, Silverman et al. (2000) and Silverman and Eals (1992) have proposed a hunter-gatherer theory of spatial ability. The theory predicts that men will be better at abilities for hunting adaptation such as navigational abilities, map reading, and the sort of mental rotations that requires hurling a spear through space to take down an animal. On the other hand, women will be better at spatial location memory as a gathering adaptation. The theory has led to empirical findings consistent with the adaptation conceptualization (Buss, 2004; Dabbs, Chang, Strong, & Milun, 1998; Geary, 1995).

Parallel to the adaptation conceptualization, Baenninger and Newcombe (1989) have proposed that spatial ability is, in part, experientially determined and that gender differences in spatial ability can be explained by gender differences in the differential exposure or experiencing of spatial activities by men and women (D. Voyer, Nolan, & S. Voyer, 2000). Accordingly, the role of experience in spatial ability can be studied by examining the relationship between spatial activity participation and spatial-ability test performance for men and women and for boys and girls. To understand gender differences in spatial ability, it is important to know if, and to what extent, differential spatial activities or experiences may foster differences in spatial ability. Such knowledge might help educators enhance spatial ability and close the performance gap between men and women by improved shaping of spatial experience (Terlecki & Newcombe, 2005).

With this view, the present study aimed to explore the relationship between spatial experience and spatial ability in a sample of Chinese, gifted students. Specifically, self-reported spatial experience of students in activities outside the formal curriculum was assessed and related to their spatial ability. The role of spatial experience in contributing to gender differences in spatial ability among students was explored.

Method

Participants

A total of 337 primary (grades 3 to 6) and secondary (grades 7 to 12) Chinese students participated in this study. These students (212 boys and 125 girls), aged 7 to 17 (M = 10.99, SD = 2.11), had already been identified by their schools and were participating in several gifted programs provided at different times at the Chinese University of Hong Kong over a period of 3 months. Schools were requested to nominate students who they judged to be gifted intellectually (e.g., with a high IQ score) or academically (e.g., with outstanding performances in school subjects), or who had demonstrated talents in other specific, nonacademic areas. Teachers nominated a total of 341 students, and 337 students responded positively to participate in this study, yielding a response rate of 99%. Since there were no generally accepted standard measures in Hong Kong schools, and schools often did not have access to information on specific IQ scores of students, teachers making recommendations would make their own judgment based on their knowledge of their students, bearing in mind that students could be regarded as gifted in different domains (Education Commission, 1990). Thus, this sample of participants could be regarded as relatively heterogeneous in terms of their giftedness or talents, and represented students from a broad age range. All 337 students voluntarily participated with parental consent.

Measures

Spatial test of the Multidimensional Aptitude Battery II (MAB-II). The spatial test of Jackson's (2003) MAB-II was used as a measure of spatial ability. The test assesses the respondents' ability to visualize abstract objects in different positions in two-dimensional space and to be sensitive to critical differences among alternatives. More generally, it requires reasoning in the figural-spatial domain combined with visual and imaginal processes that need to be evoked quickly and automatically. The spatial test thus involves testing respondents' mental rotation ability in a two-dimensional space. Each of the 50 items of the test consists of a criterion figure presented to the left and five alternatives to match on the right. Respondents were given 10 minutes to complete the 50 items. Hence, an excessive cautiousness (checking responses) will impair the speed of performance. Points were given for each correct response. The maximum score on the test was 50 points. Age has been found to affect performance substantially (Jackson).

Jackson (2003) reported that the spatial test correlated with Block Design of the Wechsler Adult Intelligence Scale--Revised (Wechsler, 1981) in the original study of test development, r = .44, N = 145, and subsequently in different studies, r = .74 to .97, M = .88. It correlated, r = .44, N = 103, with Raven's Advanced Progressive Matrices (J. Raven, J. C. Raven, & Court, 1998) in a study originally published by Kranzler and Jensen (1991). Psychometric properties of the tests have been reported in different studies (Kranzler, 1991 ; Krieshok & Harrington, 1985; Wallbrown, Carmin, & Barnett, 1988, 1989), and summarized by Jackson.

Spatial experience and activities. A 6-item scale was constructed for the purpose of this study. Four schoolteachers, who were enrolled in the author's class on gifted education for in-service training, were enlisted to suggest and generate typical students' extracurricular activities that required some types of spatial ability. From over 20 activities generated, the teachers agreed that 6 activities were typical, and these were retained for item writing in Chinese in first person statements. Three items were developed to describe in-school extracurricular activities: specifically, doing well in drawing or sculpting, taking part in designing posters or school magazine covers, and receiving awards in competitions on drawing or designing. Another three items were developed to describe extracurricular activities in outdoor events: remembering buildings and places visited by students, reading maps well and having a sense of direction, and being relied on by schoolmates in finding routes and directions.

In completing the 6 items, students were requested to rate the degree to which each of the items was descriptive of them using a scale ranging from 1 (least descriptive) to 5 (most descriptive). Initially, it was thought that these 6-item responses could be summed to yield an index or score on spatial experience and activities. A closer examination, however, revealed that the two types of activities did not fit together empirically, and might be better treated as separate indices on visual-arts experience and spatial-orientation experience, respectively.

Procedure

To assess the spatial ability of gifted students as well as their experience in spatial activities, the students were tested in groups of 30 to 50 on their spatial ability using the spatial test of the MAB-II, and on their spatial experience using the 6-item list of activities.

Results

Spatial Ability

Responses to the 50-item mental rotation test of the MAB-II were scored to yield a score on spatial ability. The mean score and standard deviation of students' responses are summarized in Table 1, which also shows the high internal consistency of the scale. To explore whether there were gender- or grade-level (primary vs. secondary students) differences on students' spatial ability, a two-factor (gender by grade level) analysis of variance (ANOVA) was performed. Results indicated that the gender main effect was significant, F(1,333) = 21.66, p < .001, partial [[eta].sup.2] = .06, because boys' scores were significantly higher than girls' scores. Also significant was the grade-level main effect, F(1,333) = 101.22, p < .001, partial [[eta].sup.2] = .23, indicating that secondary students scored significantly higher than primary students--a result that was expected from the reported age effect of the test (Jackson, 2003). The interaction effect of gender by grade level was nonsignificant (p > .05), showing that gender differences occurred in both primary and secondary students, and grade-level differences occurred irrespective of gender. The effect size ([[eta].sup.2]) of grade level was relatively large compared with the very small effect size of gender. The significant group differences are summarized in Table 1.

Spatial experience and activities

To assess students' spatial experience in extracurricular activities, the relevant 6-item responses to the activity checklist were tabulated. Preliminary maximum-likelihood factor analysis of the 6-item correlation matrix was used to explore whether the 6 items fell into a single dimension or a general factor describing spatial experience. The initial estimation yielded two factors with eigenvalues exceeding unity, accounting for 49% of the total variance. The goodness of fit test, [chi square] (4) = 17.09, p < .005, was significant and suggested that it was appropriate to consider a solution with more than one factor, or even more than two factors, because a nonsignificant chi-square would require a solution with a larger number of dimensions. The varimax-rotated two-factor solution also indicated that the three items describing visual-arts experience and the three items describing spatial-orientation experience loaded saliently on two separate factors. For comparison, the one-factor solution was also performed. The amount of variance accounted for was 36%, and the goodness of fit test, [chi square] (9) = 87.43, p < .001, suggested that the reduction in chi-square values was substantial when the two-factor solution was chosen over the one-factor solution. On the basis of the results of the factor analysis, it was deemed appropriate to consider two indices of spatial experience rather than one global score.

Table 1 shows the means and standard deviations of students' ratings on the two scales of spatial experience; the two scales had moderate internal consistency, Cronbach's [alpha] = 63 and .79, and were correlated, r = .44, p < .001. It can be seen from Table 1 that students generally reported engaging more often in spatial-orientation activities than they did in visual-arts activities. Paired t-test indicated the difference was significant, t(336) = 9.04, p < .001, partial [[eta].sup.2] = .20.

Gender and Grade-Level Differences in Spatial Experience

To explore whether there were gender or grade level (primary vs. secondary students) differences on students' spatial experience, a two-factor (gender by grade level) multivariate analysis of variance (MANOVA) was performed on the two types of spatial experience. Results indicated that for the overall spatial experience the gender main effect was significant, Wilks' [LAMBDA] = .92, F(2, 332) = 15.44, p < .001, partial [[eta].sup.2] = .09, as was the grade-level main effect, Wilks' [LAMBDA] = .97, F(2, 332) = 4.66, p < .01, partial [[eta].sup.2] = .03. The interaction effect of gender by grade level was also significant, Wilks' [LAMBDA] = .98, F(2, 332) = 3.11, p < .05, partial [[eta].sup.2] = .02. Subsequent univariate ANOVA on each of the two types of spatial experience was conducted to further investigate the significant overall gender, grade level, and interaction effects found with the MANOVA. Using the Bonferroni procedure to adjust for multiple tests, each ANOVA was evaluated at the value of .05/2, or .025. Results indicated that on the visual-arts experience, girls rated themselves significantly higher than boys did, F(1,333) = 16.52, p < .001, partial [[eta].sup.2] = .05, and the grade-level differences were not significant on either type of spatial experience. The interaction effect was significant for the spatial-orientation experience, F(1, 333) = 5.09, p < .025, partial [[eta].sup.2] = .02. To explore further on the interaction effect, additional follow-up tests on gender differences of spatial-orientation experience within each grade level, and grade-level differences within each gender were conducted. Results indicated that gender differences favored boys for secondary students, F(1, 134) = 6.86, p < .01, partial [[eta].sup.2] = .05, but not for primary students (p > .05). On the other hand, grade-level difference favoring secondary students was significant only for boys, F(1,210) = 7.83, p < .01, partial [[eta].sup.2] = .04, but not for girls (p > .05). Thus, there were significant overall gender differences on visual-arts experience favoring girls. In addition, the condition of boys having more spatial-orientation experience than girls was found to be significant only in secondary schools, and boys in secondary schools reported greater involvement in spatial-orientation activities than did boys in primary schools. However, the effect sizes ([[eta].sup.2]) of these differences were relatively modest. These differences are summarized in Table 1.

Relationships Between Spatial Experience and Spatial Ability

Table 2 shows the correlations between spatial ability and the two types of spatial experience both based on the total sample of 337 students and separately for boys and girls. It can be seen that spatial ability correlated only minimally and nonsignificantly with school visual-arts experience, but substantially and significantly with spatial-orientation experience, especially for the total sample and for male students. This difference might have to do with the testing of spatial ability based mainly on mental rotation, which might improve with more exposure to spatial-orientation activities. Since gender and age were also included in the matrix of correlations for the total sample, it can also be seen that both gender and age correlated substantially and significantly with spatial ability, and gender correlated significantly with visual-arts experience but not with spatial-orientation experience, whereas age correlated minimally with the two types of spatial experience. Unlike in the total sample, however, age correlated substantially and significantly with spatial-orientation experience for boys.

To examine more closely how the two types of spatial experience were related to spatial ability, a series of multiple linear regression analyses were conducted. Specifically, separate multiple regression analyses were performed to predict spatial ability in the total sample, and for boys and girls. Table 3 summarizes the results of the regression analyses.

From Table 3, it can be seen that age and gender emerged as significant predictors for spatial ability for all students, whereas spatial experience contributed minimally to the prediction of spatial ability. Thus, students with higher scores on spatial ability tended to be boys and older in age. Regarding the two types of spatial experience, spatial orientation appeared to contribute more to overall spatial ability, and it approached being a significant predictor of spatial ability for the total sample (p = .08) as well as for gifts (p = .07).

Spatial Experience Differences in Spatial Ability

Because substantial association was observed between spatial ability and spatial-orientation experience, and both gender and age were significant predictors of spatial ability, it was of interest to explore whether the status of spatial-orientation experience (low or high experience) could impact spatial ability while taking into account the effects of gender and age. To this end, one-way analyses of covariance (ANCOVA) were conducted separately for each gender using spatial ability as the dependent variable, spatial-orientation experience (median split) as the grouping variable, and age as the covariate. The preliminary analysis evaluating the homogeneity-of-slopes assumption indicated nonsignificance in the interaction between age and experience group for both boys, F(1,208) = 1.69, p = .20, partial [[eta].sup.2] = .01, and girls, F(1,121) = 0.01, p = .94, partial [[eta].sup.2] = .00. The ANCOVA results on the effect of spatial-orientation experience indicated a nonsignificant effect for boys, F(1,209) = 0.05, p = .82, partial [[eta].sup.2] = .00, but a significant effect for girls, F(1,122) = 5.64, p < .05, partial [[eta].sup.2] = .04, favoring girls with greater spatial-orientation experience (adjusted mean ability score, M = 30.02) over girls with less spatial-orientation experience (M = 26.39). Taken together, the results suggested that even though girls reported less involvement in spatial-orientation activities, compared to boys, spatial-orientation experience could be of greater benefit for girls.

Discussion

The present findings that there are clear gender differences in spatial ability (especially mental rotation) favoring males are consistent with past findings in numerous research studies (Linn & Petersen, 1985; D. Voyer et al., 1995), and serve to extend past findings to Chinese, gifted students in Hong Kong. While the existence of gender differences in spatial ability is not at issue, there are many competing explanations for such differences between males and females. In this regard, a host of biological and environmental factors have been implicated and explored for their contribution to the prevalence of gender differences in spatial ability. For example, variables such as rate of maturation, cerebral lateralization, sex hormones, differential experience and socialization, and gender-role identification have all been considered relevant (D. Voyer et al.). Of particular interest is the role of experiential factors explored in the present study.

Based on the findings that participation in spatial activities was positively related to performance on spatial tasks and training improved performance in both genders (Baenninger & Newcombe, 1989, 1995), it was hypothesized that gender differences might be partly accounted for by the differential amounts of participation in spatial activities. With this view, a set of extracurricular school activities was constructed to assess students' involvement in activities that require spatial ability; and these activities were categorized either as visual-arts experience or spatial-orientation experience. In general, students engaged more in activities requiring spatial-orientation skills, but compared with boys, girls engaged more in visual-arts activities--a result which goes against the assumption that differential experiences foster differential ability in boys and girls. Compared to gifts, however, secondary-school boys reported greater involvement in the other type of spatial experience, spatial-orientation activities. Despite these differences in participation in spatial activities, the effect sizes of these gender differences were relatively modest, thus providing no strong support for the idea that gender differences in spatial experience might create gender differences in spatial ability.

Also somewhat puzzling were the findings that spatial ability was not strongly related to spatial experience, or reported participation in spatial activities. The low correlation of spatial ability with visual-arts experience as opposed to the substantial correlation between spatial ability and spatial-orientation experience suggested that direct correspondence of experience (e.g., experience with mental manipulation builds mental-rotation skills) was necessary to see a more substantial or significant relationship between experience and ability. Conversely, wider or more comprehensive sampling of types of spatial experience beyond the present selection might more accurately reflect the experience-ability relationship. Thus, future studies examining the experience-ability relationship should consider more representative sampling of the domains of spatial activities as well as the domain of measures or tasks tapping spatial ability.

In the present study, the experience-ability relationship was also examined in relation to whether self-reported spatial experience could predict spatial ability as represented by performance on a measure of mental rotation. Specifically, gender and age were included as predictors to provide opportunities for assessing their contributions to the relationship between spatial experience and spatial ability. Despite the result that gender and age were the predominantly significant predictors for all students, and age was the predominantly significant predictor for boys and for girls, the greater contribution of spatial-orientation experience, when compared with visual-arts experience, to spatial ability in all students and in girls was noteworthy. In this regard, one would expect that the role of spatial experience on spatial ability, or performance on spatial tasks, could be more clearly delineated when analyses could be conducted separately for male and female subjects and with adjustment for age effects.

The present findings suggest that spatial-orientation experience could influence spatial ability or mental rotation ability, especially for female students. Accordingly, encouraging women and girls to gain spatial experience or to participate in activities requiring, for example, the finding of directions and routes, might help bridge the gap in spatial ability between the genders. One might even suggest that the magnitude of gender differences in spatial ability could be a result of the accumulation of differential spatial experience for male and female subjects. From a different perspective, there is no denying that there might be an inborn disposition for spatial ability in boys but not in girls. This same disposition might prompt boys to self-select participation in spatial activities that contribute to the further widening of the gap in gender differences. This conjecture could be tested by allowing research participants to list their favorite spatial and nonspatial activities, with the expectation that boys more than girls tend to prefer spatial activities.

Admittedly, there are many limitations in the present study. One limitation, among others, is the restricted domain sampling of both spatial activities and measures of spatial ability. Regarding spatial activities, one might raise the issue that the restriction of considering spatial activities to students' typical extracurricular activities that required some types of spatial ability could be too narrow. One might further question whether asking about students' spatial experience based on outcomes of involvement in spatial activities could be indirect, and might be less valid than asking about students' actual involvement in terms of time spent in engaging in these activities. Regarding spatial tasks, this study used a timed mental-rotation task to assess spatial ability. A different spatial task, under timed or untimed conditions, might yield different results. Thus, given the variety of spatial activities and spatial-ability measures, the challenge in future studies would be to introduce a greater variety of spatial activities and spatial tasks or to sample representative activities and representative spatial-ability measures in efforts to examine the experience-ability relationships.

Another related important limitation of this study is the possibly limited generalizability of the findings of this study. Despite that gender differences were evident in the present findings with gifted students who were likely to be talented academically, as teachers tended to nominate students with high academic achievement, it is not known whether the pattern of gender differences would hold in the population of gifted students talented in domains that require a high level of spatial ability. The wide age range of the participants in this study could also be a problem. Since the same spatial task of mental rotation was administered to all students with heterogeneous talents, it is not known whether ceiling and floor effects could affect performance on the spatial task. These issues need to be carefully considered in future studies.

The inference of a causal link between spatial experience and spatial ability based on cross-sectional data could also be an important limitation. One might argue that the causal link could be justified by assuming that references were made to the accumulating experience in spatial activities leading to current performance on a spatial task. However, the cross-sectional nature of the correlation data did not warrant causal inference from experience to ability, since the causal direction from ability to experience would be equally likely. Nonetheless, the bidirectional influence of experience and ability is likely, and future studies might consider longitudinal data that allow more thorough testing of the causal ordering of experience and ability as well as explicating developmental changes over time in the relationships between experience and ability.

Manuscript submitted May 18, 2006.

Revision accepted September 18, 2006.

REFERENCES

Baenninger, M., & Newcombe, N. (1989). The role of experience in spatial test performance: A meta-analysis. Sex Roles. 20. 327-344.

Baenninger, M., & Newcombe, N. (1995). Environmental input to the development of sex-related differences in spatial and mathematical ability. Learning and Individual Differences, 7, 363-379.

Bennett. G. K., Seashore, H. G., & Wesman, A. G. (1974). Manual for the Differential Aptitude Tests (5th ed.). New York: Psychological Corporation.

Buss, D. M. (2004). Evolutionary psychology: The new science of the mind (2nd ed.). Boston: Allyn and Bacon.

Carroll, J. B. (1993). Human cognitive abilities: A survey of factor-analytic studies. New York: Cambridge University Press.

Casey, M. B.. Nuttall, R., Pezaris. E., & Benbow, C. (1995). The influence of spatial ability on gender differences in mathematics college entrance test scores across diverse samples. Developmental Psychology. 31, 697-705.

Colangelo. N., & Davis. G. A. (Eds.). (2003). Handbook o[ gifted education (2nd ed.). Boston: Allyn and Bacon.

Dabbs, J. M, Chang. E. L., Strong, R. A., & Milun, R. (1998). Spatial ability, navigation strategy, and geographic knowledge among men and women. Evolution and Human Behavior: 19, 89-98.

Davis, G. A., & Rimm. S. B. (2004). Education of the gifted and talented (5th ed.). Boston: Allyn and Bacon.

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

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

Gardner, H. (1999). Intelligence reframed. New York: Basic Books.

Geary, D. C. (1995). Sexual selection and sex differences in spatial cognition. Learning and Individual Differences, 7, 289-301.

Gohm, C. L., Humphreys, L. G., & Yao, G. (1998). Underachievement among spatially gifted students. American Educational Research Journal, 35, 515-531.

Goleman, D. (1995). Emotional intelligence. New York: Bantam Books.

Halpern, D. F., & LaMay, M. L. (2000). The smarter sex: A critical review of sex differences in intelligence. Educational Psychology Review, 12, 229-246.

Humphreys, L. G., & Lubinski, D. (1996). Assessing spatial visualization: An underappreciated ability for many school and work settings. In C. R Benbow & D. Lubinski (Eds.), Intellectual talent: Psychometric and social issues. Baltimore: Johns Hopkins University Press.

Humphreys, L. G., Lubinski, D., & Yao, G. (1993). Utility of predicting group membership and the role of spatial visualization in becoming an engineer, physical scientist, or artist. Journal of Applied Psychology, 78, 250-261.

Jackson, D. N. (2003). Multidimensional Aptitude Battery II Manual. Port Huron, MI: Sigma Assessment Systems.

Kail, R., Carter, P, & Pellegrino, J. (1979). The locus of sex differences in spatial ability. Perception and Psychophysics, 26, 182-186.

Kranzler, J. H. (1991). The construct validity of the Multidimensional Aptitude Battery: A word of caution. Journal of Clinical Psychology, 47, 691-697.

Kranzler, J. H., & Jensen, A. R. (1991). The nature of psychometric g: Unitary process or a number of independent processes? Intelligence, 15, 397-422.

Krieshok, T. S., & Harrington, R. G. (1985). A review of the Multidimensional Aptitude Battery. Journal of Counseling and Development, 64, 87-89.

Linn, M., & Petersen, A. (1985). Emergence and characterization of sex differences in spatial ability: A metaanalysis. Child Development, 56, 1479-1498.

Raven, J., Raven, J. C., & Court, J. H. (1998). Manual.[or Raven's Progressive Matrices and Vocabulary Scale. Oxford, England: Oxford Psychologists Press.

Shea, D. L., Lubinski, D., & Benbow. C. P. (2001). Importance of assessing spatial ability in intellectually talented young adolescents: A 20-year longitudinal study. Journal of Educational Psychology, 93, 604-614.

Silverman. I., Choi, J., Mackewn, A., Fisher, M., Moro, J., & Olshansky, E. (2000). Evolved mechanisms underlying way finding: Further studies on the hunter-gatherer theory of spatial sex differences. Evolution and Human Behavior, 21, 201-213.

Silverman, I., & Eals, M. (1992). Sex differences in spatial abilities: Evolutionary theory and data. In J. H. Barkow, L. Cosmides, & J. Tooby (Eds.). The adapted mind (pp. 533-549). New York: Oxford University Press.

Sternberg, R. J., Wagner, R. K., Williams, W. M., & Horvath, J. A. (1995). Testing common sense. American Psychologist, 50, 912-927.

Stumpf, H., & Eliot, J. (1995). Gender-related differences in spatial ability and the k factor of general spatial ability in a population of academically talented students. Personality and Individual Differences, 19, 33-45.

Stumpf, H., & Eliot, J. (1999). A structural analysis of visual spatial ability in academically talented students. Learning and Individual Differences, 11, 137-151.

Terlecki, M. S., & Newcombe. N. S. (2005). How important is the digital divide? The relation of computer and videogame usage to gender differences in mental rotation ability. Set Roles, 53, 433-441.

Vernon, P. E. (1961). The structure of human abilities (2nd ed.). London: Methuen.

Voyer. D, Nolan, C., & Voyer, S. (2000). The relation between experience and spatial performance in men and women. Sex Roles, 43, 891-915.

Voyer, D., Voyer, S., & Bryden, M. (1995). Magnitude of sex differences in spatial abilities: A meta-analysis and consideration of critical variables. Psychological Bulletin, 117, 250-270.

Wallbrown. F. H., Carmin, C. N., & Barnett, R. W. (1988). Investigating the construct validity of the Multidimensional Aptitude Battery. Psychological Reports, 62, 871-878.

Wallbrown, F. H., Carmin, C. N., & Barnett, R. W. (1989). A further note on the construct validity of the Multidimensional Aptitude Battery. Journal of Clinical Psychology, 45, 429-433.

Wechsler, D. (1958). The measurement and appraisal of adult intelligence (4th ed.). Baltimore: Williams & Wilkins.

Wechsler, D. (1981). Manual of the Wechsler Adult Intelligence Scale Revised. New York: Psychological Corporation.

Wechsler, D. (2003). Wechsler Intelligence Scale for Children (4th ed.). San Antonio, TX: Psychological Corporation.

Winner, E. (1996). Gifted children: Myths and realities. New York: Basic Books.

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. He has diverse research interests. He has published in different areas that include stress and coping, psychopathology and health, creativity and intelligence, and gifted education and talent development. E-mail: [e-mail]

This study was supported in part by a direct grant for research from the Chinese University of Hong Kong. Correspondence concerning this article should be addressed to David W. Chan, Department of Educational Psychology, Faculty of Education, the Chinese University of Hong Kong, Shatin, NT, Hong Kong. E-mail: [e-mail]

Table 1
Means, Standard Deviations, and Internal Consistency
of Measures of Students' Spatial Ability and
Spatial Experience (N = 337)

Measures                   Number
                          of items       M               SD

Spatial ability
  Mental rotation            50        30.18            11.27

Spatial experience
  visual-arts                 3         9.43            3.56
  Spatial orientation         3        11.08            2.58

                         Cronbach's                  Significant
Measures                   [alpha]                group difference

Spatial ability              .96                    Boys > Girls
  Mental rotation                               Secondary students >
                                                  Primary students

Spatial experience           .79                    Girls > Boys
  visual-arts                .63                  Secondary boys >
  Spatial orientation                              Secondary girls
                                                  Secondary boys >
                                                    Primary boys

Note. Visual-arts experience and spatial-orientation experience were
assessed using the 6-item rating scale of spatial experience and
scored in the range of 3 to 15. Spatial ability was assessed using
the spatial test of the Multidimensional Aptitude Battery II, and
each test item is scored 0 or 1. a is the Cronbach's alpha internal
consistency measure. Significant group differences were determined
through univariate analyses of variance.

Table 2
Correlation Matrix of Gender, Age,
Spatial Experience, and Spatial Ability

                                              Age          VAE

Total sample (N=337)
  Age
  visual-arts experience (VAE)                -10
  Spatial-orientation experience (SOE)         10         44 ***
  Spatial ability (SA)                         63 ***    -09
  Gender                                       05         21 ***

Boys (n = 212)
  Age
  visual-arts experience (VAE)                -07
  Spatial-orientation experience (SOE)         22 **      49 ***
  Spatial ability (SA)                         66 ***    -07

Girls (n =125)
  Age
  visual-arts experience (VAE)                -17
  Spatial-orientation experience (SOE)        -07         41 ***
  Spatial ability (SA)                         62 ***    -03

                                              SOE           SA

Total sample (N=337)
  Age
  visual-arts experience (VAE)
  Spatial-orientation experience (SOE)
  Spatial ability (SA)                         15 **
  Gender                                      -05        -15 **

Boys (n = 212)
  Age
  visual-arts experience (VAE)
  Spatial-orientation experience (SOE)
  Spatial ability (SA)                         16 *

Girls (n =125)
  Age
  visual-arts experience (VAE)
  Spatial-orientation experience (SOE)
  Spatial ability (SA)                         11

Note. Decimals are omitted.

* p < .05. ** p < .01. *** p < .001 (two-tailed).

Table 3
Multiple Regression Analyses for the Prediction of
Spatial Ability Using Gender, Age, and Spatial Experience

                                           Adjusted

Criterion                [R.sup.2]        [R.sup.2]          F

Total sample (N=337)
  Spatial ability           .44              .43          64.13 *

Boys (n = 212)
  Spatial ability           .44              .43           53.87

Girls (n=125)
  Spatial ability           .41              .39           27.72

Criterion              Predictor                             t

Total sample (N=337)
  Spatial ability      Gender                             -4.03 *
                       Age                                14.87 *
                       Visual-arts experience             -0.52
                       Spatial-orientation experience      1.77
Boys (n = 212)
  Spatial ability      Age                                11.89 *
                       Visual-arts experience             -0.68
                       Spatial-orientation experience      0.66
Girls (n=125)
  Spatial ability      Age                                 8.93 *
                       Visual-arts experience              0.28
                       Spatial-orientation experience      1.83

Note. The enter procedure was used in all regression analyses
in predicting spatial ability.

* p < .001.