Comparing creative thinking abilities and reasoning ability of deaf and hearing children.

While most research of creative thinking abilities has focused on hearing children, significant factors that may contribute to the creative thinking of deaf children are in need of further investigation. Investigating creativity in disabled individuals and the factors that contribute to it in disabled populations is very important. In general, children who are deaf or hard of hearing receive fewer special programs based on their abilities than programs based on disabilities (Laughton, 1988). In schools for deaf children, educational objectives and goals tend to focus on remediation or "normalization" related to the different aspects of their disability. Consequently, deaf children have been less likely than their hearing peers to be screened, identified, and served by special programs to assess and develop their creativity (Whitmore & Maker, 1985). Failure to identify and serve deaf children with creative thinking abilities is an indictment against the society and a problem that should not be tolerated (Johnson, Karnes, & Carr, 1997).

There are many factors that contribute to one's creativity. Feldman (1999) reported that an adequate analysis of creativity involves at least six dimensions: cognitive processes, social/emotional processes, family aspects, education and preparation, characteristics of the domain and field, and historical forces and events.

There are many cognitive abilities that relate to creative thinking abilities. The current research focuses on reasoning abilities for two reasons. First, there are some standardized nonverbal instruments to assess reasoning abilities. The instructions of these instruments can be delivered by using sign language and this provides valid assessment of the reasoning abilities of deaf children. Second, intervention programs to develop reasoning abilities could provide stable improvements in this area for deaf children.

The goal of the research was to investigate the relation between reasoning abilities and creative thinking abilities by using nonverbal instruments.

Creativity and Cognition 

Guilford (1950) considered convergent and divergent thinking operations as major components of creative thinking. He explained that the two operations require that the thinker produce information when given other information. Divergent thinking includes the abilities that are most significant in creative thinking and invention.

According to Guilford (1967), commonly used intelligence tests measure convergent forms of thinking. However, creativity involves divergent thought processes, which account for 30 of the 150 factors of intelligence described by the structure of intellect model. Guilford was able to procedurally identify over 100 out of 150 factors through factor analysis.

Within the creative cognition approach, many types of creative behavior are characterized as instances of conceptual expansion that push the boundaries of a conceptual domain by envisioning and bringing to fruition novel exemplars of such domains (Ward, Saunders, & Dodds, 1999). For example, when a composer writes a new symphony, or a student draws a unique picture, their creative products can be seen as instances of conceptual expansions.

Because the products of many creative persons are outgrowths of the concepts that have come before, they can be expected to share some important properties with previous exemplars of those concepts (Basala, 1988; Friedal & Israel, 1986; Perkins, 1988; Rothenberg, 1979; Weisberg, 1986).

The creative cognition approach tends to concentrate much attention on what is old or familiar within each individual to understand the dimensions and properties of the new creative products (Ward, 1994, 1995). So, by examining what creative new ideas have in common with their predecessors, the creative cognition approach can provide insights into the way in which individuals use their existing knowledge to invent some new creative products.

Cropley (2000) added another perspective to the creative cognition approach. He emphasized the processes involved in producing effective novelty, as well as the control mechanisms that regulate novelty production, and the structures that result. He explained that effective novelty can be produced at lower levels of cognitive development, but children's creativity is likely to differ qualitatively from that of adults. Cropley emphasized that novel structures should be meaningful and practiced to be effective.

Based on the creative cognition approach, a variety of cognitive skills have been targeted as major factors in shaping creativity. These include problem-finding abilities, idea generation skills, communication skills, spatial abilities, and information organization tendencies (Amabile, 1996; Eysenck, 1994; Gardner, 1992; Schooler & Melcher, 1995). Other researchers investigated the relationship between certain cognitive skills and specific creative abilities. Schooler and Melcher indicated that cognitive skills are differentially related to solving different types of verbal-analytic and graphic-perceptual problems. More specifically, skill to avoid mental sets (an ability to be original in thinking despite situational constraints) has a stronger relationship with performance on nonverbal-perceptually based problems than with verbal-analytic problems.

Cognition in Deaf Populations

Furth (1966) argued that cognitive ability of hearing and deaf people and their developmental patterns are essentially similar. He proposed that babies show evidence of mental activity and some forms of thinking before they learn to speak; therefore, they exhibit cognition without language. Children usually begin to talk in the second year of life, and this comes after many months of development during which they come to understand verbal concepts. Piaget's (1969) investigations revealed many situations in which children fail to understand the meaning of what adults are saying. Children understand the world and make sense of what adults do and say according to their developmental levels. The nature of their understanding may be different from that of adults because their views of the world are naturally and fundamentally different.

The proposed study is based on the views of Piaget (1969) and Furth (1966) that language is not the foundation for thought. Because cognitive development precedes linguistic understanding, we can conclude that the comprehension of language is based on cognition. Hence, deaf children, even if they lack facility with language, can be expected to develop the same nonverbal cognitive ability as their hearing peers (Bloom, 2001; Brown, 1991). Deaf children use Sign Language to compensate for the lack of speech. Most deaf children use Sign Language to communicate their thoughts and ideas to other deaf and hearing individuals. Wood (1991) pointed out that, whereas deaf adults may come to think in signs rather than in words, the operations that govern their thinking are the same as those found in the hearing adult's thinking.

There have been studies conducted to assess the cognitive ability of deaf individuals. Meadow (1980) found that the cognitive performance of deaf learners is similar to their hearing peers, although their actual performance in several cognitive categories (e.g., anticipation of images, rule inferences) might differ from hearing learners. Considering the language deficiency among deaf children, Martin (1989) tested the cognitive skills of deaf and hard-of-hear ing children by asking them to generate responses to selected visual problems. He found that deaf learners depend on visual-spatial perception and processing, and they are good at simultaneous visual processing. Also, Bond (1987) found that when language demands are reduced, the cognitive development of young deaf children is comparable to that of hearing children of the same age. Studies of cognitive ability in deaf individuals have focused on a wide range of issues, including the relationship between sign language and spoken language (Anderson & Reilly, 2002; Stokoe, 2001b), factors in memory span and memory coding (Flaherty, 2001; Fletcher-Flinn & Snelson, 1997), and reading ability (Byrne, 1993; Cossu, Rossini, & Marshall 1993; Hanson & Fowler, 1987). However, the relationship between cognitive ability and creativity in the deaf population has not received enough attention from researchers in either psychology or education (Marschark & Clark, 1987; Marschark & West, 1985).

When forcing deaf children to rely on verbal communication, this action may confuse the children's misunderstanding of the directions with cognitive failure. If attempts to test deaf children's understanding expose them to language demands they cannot meet, then failures on their part may be the result of the failure to establish mutual communication between the deaf children and the hearing assessors, rather than evidence of cognitive deficits.

Creativity in Deaf Populations

There are controversial findings as to whether deaf individuals possess the same creative thinking abilities as their hearing peers. Some investigators have cited that deaf children's poor performance on several nonverbal tests of cognitive and creative thinking is evidence of their conceptual concreteness and rigidity (Myklebust, 1964; Oleron, 1953; Singer & Lenahan, 1976; Templin, 1950).

Cornelius and Hornett (1990) reported that at a certain level of linguistic deficiency (85 dB), deaf children are limited in providing the anticipated quantity and diversity of their imaginary play compared to their hearing peers. Singer and Lenahan (1976) attempted to tap the imaginative abilities of deaf children by examining their daydreams, play, and fantasy reports. They obtained language samples of 20 profoundly deaf students using structured interviews and elicited story productions. Their reports indicated that the deaf children demonstrated less creativity and imagination than hearing peers. Also, they reported, "the deaf children tended to use past and present time frames, for the most part, the fantasies were very ordinary, and these children usually related experiences that they had encountered rather than experiences they wished to encounter" (Singer & Lenahan, p. 11). Silver (1977) concluded that deaf children lag behind hearing children in terms of abstract thinking, imaginary play, and originality in the verbal domain. However, she also argued that deaf children do not lag behind hearing peers when using nonverbal instruments to assess these capacities. Also, Johnson and Khatena (1975) found that deaf children scored significantly lower than their hearing peers on the Torrance Tests of Creative Thinking-Verbal. Deaf children signed to the experimenter their responses while the hearing children wrote their responses.

On the other hand, some investigators have observed that deaf individuals performed as well as their hearing peers and sometimes exceeded them on some creative tasks. Kaltsounis (1970) compared the creative thinking abilities of deaf students and hearing students by using the Torrance Tests of Creative Thinking-Figural. He found that deaf subjects surpassed their hearing age-mates on measures of nonverbal fluency and originality, whereas the hearing subjects were superior in nonverbal flexibility. Performance differences of the deaf subjects were especially large in the fourth- and fifth-grade samples compared to the sixth-grade sample. Marschark and West (1985) examined creative story productions signed and spoken by four severely to profoundly deaf and four hearing 12 to 15 year old students. They videotaped the children while they were telling stories on two experimenter-supplied fantasy themes. The most important finding was the use of several creative language constructions by deaf as well as hearing subjects. Marschark and West also found that their deaf subjects produced novel and figurative constructions just as often as their hearing age-mates in the frequency of using gesture, pantomime, nonliteral linguistic modifications, and linguistic inventions. Marschark and West suggested that in schools for the deaf, children are not necessarily tied to concrete, literal language when sign language is available as a communication mode.

Laughton (1988) compared a traditional approach to art education with a curriculum designed to develop creative abilities of deaf students. Twenty-eight profoundly deaf children between 8 and 10 years of age were exposed to one of two curricula for 12 weeks during their regularly scheduled art classes. Pretest and posttest scores on the Torrance Tests of Creative Thinking-Figural were the dependent measures. Data analyses indicated significant improvement in two of the creativity measures (flexibility and originality) by the group exposed to the creative thinking curriculum versus the group who received traditional art education lessons.

For many years, the preponderance of literature concerning the education of the deaf has centered on their disability and the communication deficits imposed by it (Gamble, 1985). In general, research in creativity of deaf populations is still minimal. More efforts are needed to investigate whether the creative thinking abilities are similar in deaf and hearing individuals. Investigating these similarities/differences may help in the way we instruct deaf children and help to revise curriculum and activities provided to deaf children.

Deaf individuals have a well-documented deficit in understanding and producing spoken language (Ross 1996). On the other hand, the extent to which this deficit in language development affects their cognitive ability and creative thinking is still open to more investigation--especially over a large sample of deaf individuals from culturally diverse populations. There are some studies indicating that hearing individuals exceed deaf individuals in performance on several types of cognitive tasks (Ehrlich & Bramaud du Boucheron, 1974; Ross, 1996; Rozanova, 1966). Such studies did not demonstrate clearly whether these differences would remain or disappear when adequate controls for the language deficit were used.

Significance of This Study

The goal of this study was to assess the relationship between creative thinking abilities and reasoning abilities of deaf and hearing children. There were 210 deaf children who participated in the study. There are few, if any, comparable numbers of deaf children who participated in any of the existing studies. Recruiting such numbers of deaf children is helpful in generalizing the results of the study and gives more insight about the nature of creative thinking abilities and reasoning abilities in the deaf population.

The findings of this study can advance the current knowledge concerning the nature of the reasoning and creative thinking abilities of deaf children, and overcome misconceptions about deaf populations that make reference to their rigidity, concreteness, lack of imagination, and deficiencies in abstract and divergent thinking. Also, this study can provide information that may assist in identifying highly creative (deaf and hearing) children for enrollment in some gifted programs in their schools in order to fully develop their mental abilities; provide directions to assist teachers and trainers in developing and delivering appropriate forms of assessment to deaf children; and enable teachers, special education counselors, and other educators to obtain information about the strengths of deaf children's thinking.

Research Questions

The purpose of the current study is to examine the relationship between creative abilities and reasoning abilities of a group of deaf children, and compare them with a group of hearing children [see Participants]. The following questions were addressed:

1. What is the relationship between six creative thinking abilities and four reasoning abilities of deaf children?

2. What is the relationship between six creative thinking abilities and four reasoning abilities of hearing children?

3. Is there a significant difference between deaf children and hearing children in the six creative thinking abilities?

4. Is there a significant difference between deaf children and hearing children in the four reasoning abilities?

Method

Participants

Deaf Students. Based on previous research (Laughton, 1988; Marschark and West, 1985; Ross 1996), a group of 210 deaf children was selected based on the following criteria: (a) unaided sensorineural pure tone average hearing loss for three frequencies (500, 1000, 2000 Hz) of 90 dB HTL (Hearing Threshold Level) or greater in the better ear; (b) the hearing loss of deaf children ranged from 90 to 131 dB HTL (with a mean of 110 and standard deviation of 5.8); (c) 95 of the deaf children had deaf parents (both father and mother), 28 had a deaf father, 17 had a deaf mother, and 70 had hearing parents (both father and mother); (d) they had pre-lingual onset of hearing loss at birth or prior to age 2 years; (e) no other diagnosed disabilities; (f) the children do not use any hearing aids; (g) most children were identified as deaf by the age of 6 and enrolled at deaf schools at the same age or prior; (h) they used American Sign Language (ASL) as the primary communication mode; and (i) their chronological age ranged from 8 to 11 years. These criteria were used to obtain as homogeneous a sample of deaf children as possible.

Hearing Students. A group of 200 hearing children was selected based on the following criteria: (a) they were enrolled in the elementary school (3rd and 4th grades); (b) their chronological age ranged from 8 to 11 years; (c) they enrolled in school at age 6; (d) none of them enrolled in kindergarten; (e) none of them had deaf parents; and (f) all of them attended public schools.

Research Design

Correlational research was chosen to conduct the investigation in this study because it allows a researcher to look for and describe relations that may exist among naturally occurring phenomena without trying in any way to alter these phenomena (Fraenkel & Wallen, 2000). This method was appropriate for this study because its goal was to identify the relationship between components of selected reasoning abilities and creative thinking abilities. No attempt was made to affect the presentation of these abilities by the participants in this study.

Instruments

Matrix Analogies Test-Expanded Form. Naglieri (1985) designed the Matrix Analogies Test-Expanded Form (MAT-EF) to assess nonverbal reasoning abilities of children (ages 5 to 17 years). The test was constructed to reduce the influence of impaired color vision by using the colors blue, yellow, black, and white. The MAT-EF has 64 items and was designed to provide an individually administered assessment of nonverbal reasoning ability that allows individuals to either point to or say aloud the number for his/her answer. The MAT-EF uses 64 abstract designs printed one per page to provide norms for a large, representative sample of individuals, ages 5 to 17 years living in the United States. The response to these designs require minimal motor involvement and minimal verbal comprehension requirements. Reliability and validity analyses were conducted on the MAT-EF. Internal consistency and test-retest analyses revealed high reliability coefficients for the Matrix Analogies Test- Expanded Form (.86 and .91 consecutively). These results support the evidence that the MAT-EF has good reliability coefficients in deaf populations. A useful way of conducting validity studies is analyzing the latent construct of the instrument. Item factor analysis was used to analyze the MAT-EF scores in the deaf population in order to understand the test's latent construct and to confirm its validity. Results from the/'actor analysis provide evidence that the four item groups of the MAT-EF were designed to measure a single construct: nonverbal reasoning ability.

The four groups of items in the MAT-EF include: Pattern Completion, Reasoning by Analogy, Serial Reasoning, and Spatial Visualization and each group consists of 16 items:

1. Pattern Completion: This subtest requires that students choose one of four options that accurately complete a pattern. These items require the individual to examine the directions and shapes in the diagram presented to determine which option fits the pattern and belongs on the question mark. The correct option should continue the pattern without interruption in a manner similar to that found in the rest of the diagram. These items are made more difficult when they require completion of a pattern that does not have contiguous parts.

2. Reasoning by Analogy: This subtest of items requires that the examinee investigate how the change(s) in one figure is (are) analogous to the change(s) in another. These items require the individual to analyze a matrix on the basis of specific variables (i.e. shape, size, shading) and determine how changes in two or more variables converge to result in a new figure. Complexity is achieved in this item group through increasing the number of changing variables in the item matrix.

3. Serial Reasoning: This subtest of items requires the student to discover the order in which items appear throughout a matrix. The boxes in the items included in this group have a specific order in which they appear, and the student has to decide which option completes the matrix according to the specific order. Complexity is achieved in these items through sequencing more than one variable in the matrix.

4. Spatial Visualization: This subtest of items requires the student to imagine how a figure would look when two or more components are combined. Difficulty is achieved in this item group by requiring manipulation of many figures at one time.

Torrance Tests of Creative Thinking-Figural, Form A. The Torrance Tests of Creative Thinking (TTCT) were first published by E. Paul Torrance and his associates in 1966. The tests have been renormed four times in 1974, 1984, 1990, and 1998. There are two forms (A and B) of the TTCT-Verbal and two forms (A and B) of the TTCT-Figural. This study used only the TTCT-Figural (form A) which has much evidence to support its use (e.g., Cropley, 2000). It has been translated into over 35 languages (Miller, 2002). The TTCT-Figural is the most widely used test of creativity (Colangelo & Davis, 1997), and has been used in more research than any other creativity test (Cramond, 1993; Lissitz & Willhoft, 1985). The standard administration and scoring procedures (Davis & Rimm, 1994), as well as the development and evaluation (Colangelo & Davis), have made the TTCT especially useful for identifying creativity among gifted and talented students. The TTCT-Figural has had 25 years of extensive development and evaluation (Miller). It has large norming samples, valuable longitudinal validations, and high predictive validity for a very wide age range (Cropley).

The TTCT-Figural, Form A and B are unbiased in terms of gender and race, and for persons who have various language, socioeconomic status, and cultural backgrounds (Cramond, 1993; Torrance, 1974). Each form of the test consists of three activities, each designed to tap somewhat different aspects of creative functioning: (a) Picture Construction consists of a single curved shape, (b) Picture Completion consists of 10 incomplete linear figures, and (c) Lines, in Figural Form A, consists of three pages of sets of parallel lines.

These three activities provide scores for five norm-referenced creative thinking abilities and 13 criterion-referenced abilities. Norm referenced creative thinking abilities are fluency, originality, abstraction of titles, elaboration, and resistance to premature closure. Fluency refers to the number of ideas a person expresses through interpretable responses that use the stimulus in a meaningful manner. Originality refers to the infrequency and unusualness of the response. Abstractness of titles refers to the ability to produce good titles and involves the thinking processes of synthesis and organization.

In scoring elaboration, credit is given for each pertinent detail (idea, piece of information, etc.) added to the original stimulus figure, its boundaries, and/or its surrounding space. Resistance to premature closure refers to the ability of a creative person to keep open and delay closure long enough to make the mental leap that makes possible original ideas. This is measured by the individual's tendency to close or not to close the incomplete figures immediately with straight or curved lines (Torrance, 1998).

In scoring for the criterion-referenced creative thinking strengths, any genuine appearance of a strength is indicated by a plus sign (+). If the strength appears three or more times, this is indicated by two plus signs (++). These creative strengths include: emotional expressiveness (e.g., in drawings, title); storytelling articulateness (context, environment); movement or action (e.g., running, dancing, flying, falling); expressiveness of titles; synthesis of incomplete figures (e.g., combination of 2 or more); synthesis of lines (e.g., combination of 2 or more); unusual visualization (above, below, at angle, etc.); internal visualization (e.g., inside, cross section); extending or breaking boundaries; humor (e.g., in titles, captions, drawings); richness of imagery (e.g., variety, vividness, strength); colorfulness of imagery (e.g., exactingness, earthiness); and fantasy (e.g., figures in myths, fables, fairly tales, science fiction).

Data Analysis

Analyses of these data were done using the canonical correlation analysis to investigate the relationship between the six creative-thinking abilities and the four reasoning abilities of both deaf and hearing children. Multivariate analysis of variance (MANOVA) was used to investigate the differences between deaf and hearing children in the four reasoning abilities and six creative thinking abilities.

Procedure

Two teachers for the deaf children volunteered to help administer the instruments that were used in this study. American Sign Language was used as a communication mode with the deaf children. Once consent for participation in the study was obtained from parents, either the researcher or the teachers reviewed the students' records to select the students who meet the assigned criteria for this study.

The tests were administered either during students' classes or study hall times during the school day. The MATEF was administered individually. The TTCT-Figural, Form A was administered in groups (10 children in each group) and two groups were tested each day. Given the small numbers of students completing the TTCT-F, Form A during any administration (10 students per administration session), both tests were administered simultaneously (i.e., some students were tested with the Torrance Tests of Creative Thinking-Figural, Form A, while some students were tested with the Matrix Analogies Test-Expanded Form.) Testing sessions typically took 1-2 class periods, depending on the school's schedule. The order of testing was randomly assigned across deaf and hearing children. Both tests were administered in two different school days. After each administration session, candy, pencils, and notebooks were given to the children.

Two certified independent raters (graduate students at The University of Georgia) helped in scoring the TTCT-Figural, Form A. However, the researcher scored the majority of the samples used in this study after being stringently trained at the Torrance Center and receiving a scoring certificate, and therefore, was considered the expert.

Results

Canonical correlation analysis was conducted to examine the relationship between two sets of variables (creative thinking abilities and reasoning abilities) as each set contains more than one variable (Thompson, 1984). The six dimensions of creative thinking abilities were used as one set of variables and the four reasoning abilities were used as the other multivariate profile. Deaf and hearing children's scores on both instruments met the basic distributional assumptions underlying multivariate analyses.

The canonical correlation analysis was conducted to answer the question regarding the relationship between the six creative thinking abilities and the four reasoning abilities of deaf children. The analysis revealed that four canonical correlations were statistically significant (p < .05). However, when the first pair of canonical functions was excluded, the remaining three pairs canonical functions combined were not statistically significant. Similarly, with the removal of the first and second canonical functions, the remaining canonical functions combined were not statistically significant. Further removal of canonical functions also produced no statistically significant results.

Together, the results suggest that the first canonical function was statistically significant but all the subsequent canonical functions were not. However, because the calculated probabilities are sensitive to sample size, particular attention should be paid to the educational (practical) significance of the obtained results (Thompson, 1980). The educational significance of canonical correlations typically is assessed by examining their size (Thompson, 1984). The canonical correlation indicates how much variance the two sets of weighted original variables share with each other (Thompson, 1984). In the current study, the first canonical correlation (.789) and adjusted canonical correlation (.773) were judged to be educationally significant (F statistics is 29.32), contributing 62% (i.e., [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) of the shared variance. Thus the focus was based on explaining the first canonical correlation.

Canonical loading is used to explain this canonical function. Table I contains the canonical loadings for the two sets of variables for the first canonical function.

This table indicates that the abstractness of titles has the highest loading .88, resulting in the shared variance with reasoning abilities. The reasoning variables have a loading ranging from .69 to .88; however, the loading for the creative thinking variables ranged from -.17 to .88.

Because standardized function coefficients typically are highly affected by the collinearity of the variables in a given set (Thompson, 1984), structure coefficients were interpreted. The coefficients are particularly useful for assessing the nature of the relationships between two sets of variables. These structure coefficients revealed that all six dimensions of creative thinking abilities made important contributions to the first canonical variate.

According to Thompson (1984), variables with small structure coefficients, but standardized coefficients that are large in absolute value magnitude, indicate that they are suppressor variables in the canonical correlation model. Suppressor variables are those that assist in the prediction of dependent variables because of their correlation with other sets of variables (Tabachnick & Fidell, 1996). In the current study, the abstractness of titles served as a suppressor variable. A variance of 68.5% in the pattern completion variable, 87.6 % in the reasoning by analogy variable, 71.4% in the serial reasoning variable, and 71.5% in the spatial visualization variable is explained by the abstractness of titles variable (creative thinking set). This canonical function has been labeled, "Abstractness of titles in creative thinking."

A canonical correlation analysis also was conducted to investigate the relationship between the six creative-thinking abilities and the four reasoning abilities of hearing children. The analysis revealed that only one pair of canonical correlations should be considered. For the hearing children sample, the first canonical correlation (.723), and adjusted canonical correlation (.711) both were judged to be educationally significant (F statistics = 27.97), contributing 52% (i.e., [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) of the shared variance. Canonical loadings are used to explain this canonical function. Table 2 contains the canonical loadings for the two sets of variables for the first canonical function for the hearing children.

This table indicates that only abstractness of titles has a loading of .91, resulting in the shared variance with reasoning abilities. The value of the canonical loading refers to the shared variance between the suppressor variable and canonical function variables. Table 2 shows that 46.1% of the variance in the pattern completion variable, 81% of the variance in the reasoning by analogy variable, 61.2% of the variance in the serial reasoning variable, and 31.3% of the variance in the spatial visualization variable are explained by the abstractness of titles variable (creative thinking set). For this group of children, the canonical function includes only two variables with large canonical loadings (reasoning by analogy and serial reasoning). Also, abstraction of titles appeared to serve as a suppressor variable and assist in the identification of the canonical function. This canonical function has been labeled visual reasoning. Abstractness of titles reflects a language element and it seemed to affect the nonverbal cognitive performance of the deaf children.

From the previous analyses, we could conclude that abstractness of titles was the dominant variable that was responsible for the identification of the canonical function in both hearing and deaf samples. Although the canonical function for the deaf sample includes all four variables of reasoning abilities, the canonical function for the hearing sample includes only two variables (reasoning by analogy and serial reasoning). The abstractness of titles' canonical loading is larger for the hearing children (.91) than for the deaf sample (.881). This result indicated that the abstractness of titles' variable has greater effect on the cognitive performance of hearing children than for deaf children in nonverbal context.

In order to investigate the differences between deaf and hearing children in creative thinking abilities, a Multivariate Analysis of Variance (MANOVA) was carried out by using SPSS 11.0. MANOVA is used to simultaneously compare mean differences on the two sets of scores on the outcome variables (Anderson, 2003). The test statistics employed were Wilks' Lambda, Pillai's trace, Hotelling's trace, and Roy's largest root. Table 3 shows the multivariate statistics were statistically significant.

The MANOVA revealed a significant overall effect (Wilks' Lambda =. 17, p < .05). Both tau squared ([[tau].sup.2]) and zeta squared ([[zeta].sup.2]) were computed as indices of effect size. Tau squared ([[tau].sup.2]) was equal to .74 and zeta squared (42) was equal to .76. The discriminant function that results from the descriptive discriminant analysis may be used to calculate a discriminant score for each child. The discriminant scores are then correlated with each variable. These resulting correlations are referred to as structure coefficients. In the current study, these structure coefficients were based upon the total sums of squares and cross-product matrices. Structure coefficients greater than .3 are considered to be meaningful. Table 4 represents the structure coefficients for creative thinking abilities.

The results from the MANOVA indicate that deaf children are different from hearing children in creative thinking abilities with respect to the abstractness of title variable. Tukey HSD tests revealed that the hearing children scored significantly (p < .05) higher than the deaf children in fluency, originality, and abstractness of titles. However, there were no significant differences between the deaf and hearing children in elaboration, resistance to premature closure, and creative strengths (p < .05) in all cases.

In order to investigate the expected differences between deaf and heating children in reasoning abilities, a MANOVA was carried out. An overall nonsignificant effect was found (Wilks' Lambda = .959). Both tau squared ([[tau].sup.2]) and zeta squared ([[zeta].sup.2]) were computed as indices of effect size. Tau squared ([[tau].sup.2]) was equal to .08 and zeta squared ([[zeta].sup.2]) was equal to. 11. These results show that there were no differences between deaf and hearing children in nonverbal reasoning abilities.

When comparing the means of the deaf and hearing children, the hearing children scored higher than the deaf children in pattern completion, reasoning by analogy, and serial reasoning. However, deaf children scored higher than the heating children in spatial visualization.

Conclusion

The canonical correlation analyses revealed one significant dimension in both deaf and heating children. This dimension links between the abstraction of titles and the four reasoning abilities parables in the deaf sample, and two reasoning variables (analogy by reasoning and serial reasoning) in the hearing sample.

Although the abstraction of titles seemed to have greater effect in the hearing sample (.91) than in the deaf children sample (.88), this variable affected the cognitive performance of deaf children more than the performance of hearing children. The results of the canonical analysis indicated that deaf children differed in only one of the creative thinking abilities investigated. The current findings do not support previous research (Conrad, 1979; Moores, 1978; Oleron, 1953; Quigley & Paul, 1984; Singer & Lenhan, 1976; Watts, 1979) that indicated the rigidity, concreteness, and lack of divergent thinking among deaf individuals.

Most of the previous investigations in creative thinking abilities of deaf children used the Torrance Tests of Creative Thinking-Figural with the old method of scoring, which included providing scores for flexibility, fluency, originality, and elaboration. The old scoring method did not provide scores for the abstraction of titles variable.

The current findings agree with Laughton (1976) in finding no differences in originality and elaboration; however, Laughton found that the deaf and hearing children differed only on fluency. The current findings agree with Horrocks and Pang (1968) that deaf students scored approximately the same as hearing students in fluency and originality. However, Horrocks and Pang found that deaf children scored higher in elaboration.

The findings of the multivariate analysis of variance revealed that there were no differences between deaf and hearing children in reasoning abilities. This result supports the finding of Martin (1989) that deaf learners depend on visual spatial perception and processing, and they are good at simultaneous visual processing. Also, this result supports some other previous research, which indicated that the cognitive development of young deaf children is comparable to that of hearing children of the same age (Al-Hilawani, 2000; Bond, 1987; Chovan, 1972; Craig and Gordon, 1989; Martin; Meadow, 1980).

The findings of the multivariate analysis of variance revealed that there were some similarities and some differences between the deaf and hearing samples regarding creative thinking abilities. The groups were different only in the abstraction of titles variable, and they were similar in the other five variables of creative thinking abilities.

The difference in the abstraction of titles variable between deaf and hearing children raises the issue of the effect of a language deficit in deaf children on the cognitive and creative performance. Earlier research examining deaf children's language abilities concluded that they were concrete and literal in their language abilities and, by extension, concrete and literal in their cognitive abilities (e.g., Blackwell, Engen, Fischgrund, & Zarcadoolas, 1978). Everhart and Marschark (1988), however, showed that when evaluated in sign language rather than English, deaf students showed language flexibility and creativity equal to or superior to that of hearing students of the same age.

Most obviously, deaf children as a group are more heterogeneous than hearing students. Beyond variability that may be directly related to their hearing losses, and beyond the normal variability found among children, deaf children frequently have different experiences, different language backgrounds, and perhaps different cognitive skills. This does not mean that deaf students are in any way deficient (Marschark, 2003). Much of the early research concerning the cognitive development of deaf children was presumed to be examining cognitive growth in the absence of language (e.g., Furth, 1966). Other investigations involved tasks that either required comprehension of the vernacular (e.g., English) in order to understand written instructions, or demanded a history of reading in order to have the knowledge necessary to perform according to normative standards.

Stokoe and Marschark (1999) noted that in the case of children acquiring their first language, the language used by adults around the child could affect the child's cognitive development and result in cognitive and cultural environmental connections. The underlying conceptual system is unlikely to be language-specific or even specific to a particular mode of language; although particular concepts may well vary in their availability or ease of communication in one language (or language mode) or another. It thus seems safe to assume that the more directly communication-whether by gesture, spoken language, or sign language--maps onto the world, the easier it will be to comprehend the cognitive tasks.

Implications

In this study, multiple facets of the creative thinking of deaf children became apparent. The TTCT-F provides valuable information to the teachers of the deaf who can thereby foster their students' creative thinking by integrating humor, thinking, feeling, intuition, and physical sensing into their instruction. By constantly adjusting their program through detailed observations over time, teachers can show different learners how to use their strengths to improve academic and social performance.

Assessment of creativity or creative potential, in addition to general intelligence and specific academic achievements, is generally acknowledged as a key component in the definition of giftedness. When an assessment confirms that a student is creative/deaf, this study suggests that programming should take into account those areas in which creative aspects resemble both those of the creative and the deaf. Like those who are creative heating children, children who are creative/deaf require enriching and stimulating cognitive experiences where they can use their problem solving abilities, independent research skills, initiative, ability to elaborate, and ability to express both their sense of humor and feelings.

Based on the findings of the current study, there exists a need to improve assessment procedures used to assess the creative thinking abilities and reasoning abilities of deaf learners. Considering this, educators of deaf children need to learn more about creative deaf individuals: how to facilitate their learning process, how to keep them on task, and how to provide them with the most appropriate educational environment. Assessment objectives must focus on specific, higher-order cognitive skills. This is important in guiding the development of instruments that ensure deaf learners have various opportunities to express their abilities at advanced levels. Curriculum and programming should emphasize building on strengths, encouraging motivation, and developing positive self-concept. Assessment of creativity in deaf children as well as hearing children should include: tests of divergent thinking; attitude and interest inventories; personality inventories; biographical inventories; ratings by teachers, peers, and supervisors; judgment of products; and self-reported creative activities and achievements.

Limitations

There are a few limitations that should be considered in this study:

1. No attempt was made to determine if any of the participants (deaf and hearing) had training in reasoning or in creative thinking.

2. Hearing children were selected on the basis of convenience sampling procedures. It would be more beneficial if these children were selected by random sampling procedures.

3. No attempt was made to investigate language variability among the deaf children. Further investigation will be made on the same sample of deaf children concerning the language variable and its relation to reasoning and creative thinking abilities.

4. The participants of the study were deaf and hearing children. If each group was divided into subgroups based on the age of the participants, it might reveal more significant underlying constructs that could provide more insights about the relationship between reasoning abilities and creative thinking abilities.

Manuscript submitted July 13, 2004.

Revision accepted October 19, 2005.

REFERENCES

Al-Hilawani, Y. A. (2000). A new approach to evaluate meta-cognition in hearing average achieving, hearing underachieving, and deaf/hard of hearing elementary school students. British Journal of Special Education, 27, 41-47.

Amabile, T. M. (1996). Creativity in contest. New York: Westview.

Anderson, D., & Reilly, J. (2002). The MacArther Communicative Development Inventory: Normative data for American Sign Language. Journal of Deaf Studies and Deaf Education, 7, 83-106.

Anderson, T. W. (2003). An introduction to multivariate statistical analysis (3rd ed.). Hoboken, NJ: Wiley-Interscience.

Basala, G. (1988). The evolution of technology: London: Cambridge University Press.

Blackwell, P., Engen, E., Fischgrund, J., & Zareadoolas, C. (1978). Sentences and other systems: A language and learning curriculum/or hearing-impaired children.

Washington, DC: National Association of the Deaf.

Bloom, L. (2001). The intentionality model and language acquisition: Engagement, effort, and the essential tension in development. Boston: Blackwell.

Bond, G. G. (1987). An assessment of cognitive abilities in hearing and hearing impaired preschool children. Journal of Speech and Hearing Disorders, 52, 319-323.

Brown. R. (1991). Cognition and the development of language. New York: Wiley.

Byrne, B. (1993). Learning to read in the absence of phonemic awareness. Cognition, 48, 285-288.

Chovan, W. L. (1972). Role of vocal labeling in memory for object arrangements by deaf and hearing children. Perceptual and Motor Skills, 34, 59-62.

Colangelo, N., & Davis, G. A. (1997). Handbook of gifted education. Needham Heights, MA: Viacom.

Conrad, R. (1979). The deaf school child. London: Harper & Row.

Cornelius, G,, & Hornett, D. (1990). The play behavior of hearing impaired kindergarten children. American Annals of the Deaf, 135, 316-321.

Cossu, G., Rossini, F., & Marshall. J. C. (1993). When reading is acquired but phonemic awareness is not: A study of literacy in Down's syndrome. Cognition, 46, 129-138.

Craig. H. B., & Gordon, H. W. (1989). Specialized cognitive function among deaf individuals: Implications for instruction. In D. S. Martin (Chair), International symposium on cognition, education, and deafness (pp. 112-126). Washington, DC: Gallaudet University Press.

Cramond, B. (1993). The Torrance Tests of Creative Thinking: From design through establishment of predictive validity. In R. F. Subotnik, & K. D. Arnold (Eds.). Beyond Terman: Contemporary longitudinal studies of giftedness and talent (pp. 229-254). Nor wood, NJ: Ablex.

Cropley, A. J. (2000). Defining and measuring creativity: Are creativity tests worth using? Roeper Review, 23, 72-79.

Davis, G. M., & Rimm S. B. (1994). Education of the gifted and talented. Needham Heights, MA: Simon & Schuster.

Ehrlich. S., & Bramaud du Boucheron, G. (1974). Memoire immedediate et apperentissage verbal chez les enfants sourds et entendants. Bulletin de Psychologic, 27, 238-245.

Everhart. V. S., & Marschark, M. (1988). Linguistic flexibility in signed and written language productions of deaf children. Journal of Experimental Child Psychology, 46, 174-193.

Eysenck, H. J. (1994). The measurement of creativity. In M. A. Boden (Ed.), Dimensions of creativity (pp. 199-242). Cambridge, MA: MIT.

Feldman, D. H. (1999). The development of creativity. In R. J. Sternberg (Ed.). Huudbook of Creativity (pp. 169-186). New York: Cambridge University Press.

Flaherty, M. (2001). Memory span for Arabic numerals and digit words in Japanese Kanji in deal signers. Japanese Psychological Research, 43, 63-71.

Fletcher-Finn, C., & Snelson, H. (1997). The relation between meta-linguistic ability, social reefs-cognition and reading: A developmental study. New Zealand Journal of Psychology, 26, 20-28.

Fraenkel. J. R., & Wallen, N. E. (2000). How to design and evaluate research in education (4th ed.). New York: McGraw-Hill.

Friedel, R., & Israel. P. (1986). Edison's electric light: Biography of an invention. New Brunswick, NJ: Rutgets University Press.

Furth, H. G. (1966). Thinking without language: Psychological implications pf deafness. New York: Free Press.

Gamble, H. (1985). A national survey of programs for intellectually and academically gifted hearing impaired students. American Annals o/the Deaf. 130, 508-513.

Gardner, H. (19927. Multiple intelligences: The theory in practice. New York: Basic Books.

Guildford, J. P. (1950). Creativity. American Psychologist, 5, 444-454.

Guilford. J. R (1967). The structure of intellect. New York: McGraw-Hill.

Hanson. V. L., & Fowler. C. A. (1987). Phonological coding in word reading: Evidence from hearing and deaf readers. Memory & Cognition, 15, 199-207.

Horrocks, C., & Pang, H. (1968). An explanatory study of creativity in deaf children. Perceptual and Motor Skills, 27, 844-846.

Johnson, L. J., Kames, M. B., & Carr, V. W. (1997). Providing services to children with gifts and disabilities: A critical need. In N. Colangelo & G. Davis (Eds.), Handbook of gifted education (pp.516-527). Needham Heights, MA: Allyn and Bacon.

Johnson, R. E., & Khatena, J. (1975). Comparative study of verbal originality in deaf and hearing children. Perceptual and Motor Skills, 40, 631-635.

Kaltsounis. B. (1970). Differences in verbal creative thinking abilities between deaf and hearing children. Psychological Reports, 26. 727-733.

Laughton, J. (1976). Nonverbal creative thinking abilities as predictors of linguistic abilities of hearing impaired children. Unpublished doctoral dissertation, Kent State University, Ohio.

Laughton, J. (1988). Strategies for developing creative abilities of hearing impaired children. American Annals of the Deaf 121. 258-263.

Lissitz, R. W., & Willhoft. J. L. (1985). A methodological study of the Torrance Tests of Creativity. Journal of Educational Measurement, 22(1), 1-11.

Marschark, M. (20113). Interactions of language and cognition in deaf learners: From research to practice. International Journal of Audiology, 42 (Supply. 1), 41-48.

Marschark. M., & Clark, D. (19877. Linguistic and nonlinguistic creativity of deaf children. Developmental Review, 7, 22-38.

Marschark, M., & West, S. A. (1985). Creative language abilities of deaf children. Journal of Speech and Hearing Research, 28, 73-78.

Martin, D. S. (1989) Cognitive process and hearing impaired learners. (Report No. 23). Washington DC: Office of Educational Research and Improvement. (ERIC Document Reproduction Service No. ED 353731).

Meadow, K. (1980). Deafness and child development. Berkeley: University of California Press.

Millar, G. W. (2002). The Torrance kids at mid-life. Westport, CT: Ablex.

Moores, D. (1978). Educating the deaf: Psychology, principles, and practices. Boston: Houghton Mifflin.

Myklebust, H. (1964). The psychology of deafness. New York: Grune and Stratton.

Naglieri, J. A. (1985). Matrix Analogies Test Expanded form. New York: The Psychological Corporation.

Oleron. P. (1953). Conceptual thinking of the deaf. American Annals of the Deaf, 98, 304-310.

Perkins, D. N. (1988). Creativity and the quest for a mechanism. In R. J. Sternberg & E. Smith (Eds.), The psychology of human thought (pp. 123-134). New York: Cambridge University Press.

Piaget, J. (1969). The psychology of the child. New York: Basic Books.

Quigley, S. P.. & Paul, P. V. (1984). Language and deafness. London: Groom Helm.

Ross. B. M. (1996). Probability concepts in deaf and hearing children. Child Development, 87, 917-928.

Rolhenberg, A. (1979). The emerging goddess: The creative process in art, science, and other fields. Chicago: University of Chicago Press.

Rozanova, T. V. (1966). Pictural memory of deaf children. In R. M. Boskis & A. 1. Meshcheryakov (Eds.), Sensory defects and mental development. Symposium 33, (pp. 97-133). Moscow: International Congress of Psychology.

Schooler, J. W., & Melcher, J. (1995). The ineffability of insight. In S. M. Smith, T. B. Ward, & R. A. Finke (Eds.), The creative cognition approach (pp. 97-133). Cambridge, MA: MIT.

Silver. R. A. (1977). The question of imagination, originality, and abstract thinking by deaf children. American Annals of the Deaf, 122, 349-354.

Singer, D., & Lenahan. M. (1976). Imagination content in dreams of deaf children. American Annals of the Deaf, 121, 44-48.

Stokoe, W. C. (2001b). Where should we look for language? Sign Language Studies, 1, 426-433.

Stokoe, W. C., & Marschark, M. (1999). Signs, gestures and signs. In L. Messing & R. Campbell (Eds.), Gesture, Speech and Sign (pp. 161-182). Oxford, UK: Oxford University Press.

Tabachnick, B., & Fidell, L. (1996). Using multivariate statistics (3rd ed.). New York: HarperCollins.

Templin, M. (1950). The development of reasoning in children with normal and defective hearing. Minneapolis: University of Minnesota Press.

Thompson, B. (1980, April). Canonical correlation: Recent extensions for modeling educational processes. Paper presented at the annual meeting of the American Educational Research Association, Boston, MA.

Thompson, B. (1984). Canonical correlation analysis: Uses and interpretations. Newbury Park, CA: Sage.

Torrance, E. P. (1974). Creativity in the classroom. Washington, De: National Education Association.

Torrance, E. P. (1998). Torrance Tests of Creative Thinking. Bensenville, IL: Scholastic Testing Service.

Ward, T. B. (1994). Structured imagination: The role of conceptual structure ill exemplar generation. Cognitive Psychology 27, 1-40.

Ward, T. B. (1995). What's old about new ideas? In S. M. Smith, T. B. Ward, & R. A. Finke (Eds.), The creative cognition approach (pp. 157-178). Cambridge. MA: MIT.

Ward. T. B., Saunders. K. N., & Dodds, R. A. (1999). Creative cognition in gifted adolescents. Roeper Review, 21, 260-265.

Watts, W. J. (1979). The influence of language on the development of quantitative, spatial, and social thinking in deaf children. American Annals of die Deaf, 12, 45-56.

Weisberg, R. W. (1986). Creativity, genius, aud other myths. New York: Freeman.

Whitmore, J. R., & Maker, C. J. (1985). Intellectual giftedness in disabled persons. Rockville, MD: Aspen.

Wood. D. (1991). Communication and cognition: How the communication style of hearing adults may hinder rather than help deaf learners. American Annals of the Deaf, 136, 247-251.

Fawzy Ebrahim is Assistant Professor in the Department of Instructional Technology and Research in the College of Education at Florida Atlantic University. He obtained his Ph.D. from the University of Georgia in the Spring of 2004. His research interests encompass deaf gifted children, instrument validation procedures, assessment of children and youth, and effects of testing accommodation on performance of students with exceptionalities. E-mail: [e-mail]

Table 1

Canonical Loadings of the
First Pair of Canonical
Functions in Deaf Children

                                    Canonical
Variables                           loadings

Pattern completion                     .69
Reasoning by analogy                   .88
Serial reasoning                       .71
Spatial visualization                  .72
Fluency                                .56
Originality                            .28
Elaboration                           -.43
Abstractness of titles                 .88
Resistance to premature closure       -.29
Creative strengths                    -.17

Table 2

Canonical Loadings of the
First Canonical Function
in Hearing Children

                                    Canonical
Variables                           loadings

Pattern completion                     .46
Reasoning by analogy                   .81
Serial reasoning                       .61
Spatial visualization                  .31
Fluency                                .46
Originality                            .42
Elaboration                           -.43
Abstraction of titles                 -.91
Resistance to premature closure       -.32
Creative strengths                    -.17

Table 3

Multivariate Test Statistics Comparing
Deaf and Hearing Children on
Creative Thinking Abilities

                                       Approximate

                                    F         Significant
Statistics            Value     statistics       of F

Wilks' Lambda         .17185      2.7635         .030
Pillai's trace        .18835      2.8123         .031
Hotelling's trace     .85049      2.705          .027
Roy's largest root    .13522

Table 4

Structure Coefficients for
Creative Thinking Abilities

                                     Structure
Variable                            Coefficient

Fluency                                -.55
Originality                            -.07
Elaboration                             .21
Abstractness of titles                  .83
Resistance to premature closure         .28
Creative strengths                      .23