Gifted adolescents' overexcitabilities and self-concepts: an analysis of gender and grade level.

Overexcitabilities, or enhanced levels of experience, have emerged as an area of study in relation to the gifted. Specifically, gender and age differences have been explored. Overexcitabilities could affect many facets of life, one of which could be self-concept. Self-concept is also a commonly researched construct in relation to the development of the gifted. Because the overexcitabilities are believed to contribute to the potential for further growth (Bouchard, 2004; Dabrowski, 1964; Piechowski & Miller, 1995), a relationship between overexcitabilities and self-concept may affect the likelihood that gifted adolescents with high overexcitabilities reach their developmental potential. Before discussing the current study, the following review of the literature will discuss what is known about overexcitabilities among the gifted, self-concept among the gifted, and the potential for a relationship between overexcitabilities and self-concept.

Overexcitabilities

Overexcitabilities are ways that an individual experiences the world (Piechowski, 1979) and allow her/him "to take in and process larger than usual amounts of stimuli from the environment" (Gallagher, 1986, p. 115). Overexcitabilities can be understood as "channels of information flow" (Piechowski, p. 29). The more channels that are open, the more diverse and intense experiences will be. How a person responds to a situation is determined by which overexcitability is most dominant, as overexcitabilities may be expressed in one or more of five dimensions: psychomotor, sensual, intellectual, imaginational, and emotional. Rather than viewing overexcitabilities as emotional extremes, though, some researchers view them as contributing to the potential for further growth (Bouchard, 2004; Dabrowski, 1964; Piechowski & Miller, 1995). "The richer and more complex their expression, the stronger the potential for development" (Piechowski & Cunningham, 1985, p. 156).

The notion of overexcitabilities stems from Dabrowski's (1964) Theory of Positive Disintegration, which is a theory of personality development. Dabrowski's theory defines five levels of personality development, as well as individual characteristics that contribute to developmental potential. Dabrowski believed developmental potential is contingent upon one's level of intelligence, one's talents, one's will to develop, and one's levels of overexcitabilities. For a thorough description of Dabrowski's ideas and the Theory of Positive Disintegration, the reader is referred to Dabrowski (1937, 1964, 1972) and Dabrowski, Kawczak, and Piechowski (1970).

The five forms of overexcitabilities may be described as follows: A psychomotor overexcitability refers to a surplus of energy and nervousness. Manifestations include extreme enthusiasm, rapid speech, love of intense activity, and impulsive actions. If experiencing stress, a person with a psychomotor overexcitability may act competitively, exhibit anxious behaviors, or act compulsively. A sensual overexcitability is an enhanced level of sensory experience and is marked by the pursuit of pleasure through senses such as tastes and smell. Manifestations may include enhancing stimuli, such as through seeking to become the focus of attention or overindulging oneself in food, or removing oneself from stimuli, such as by taking the tags out of ones' clothes or wearing ear plugs. An intellectual overexcitability is associated with striving for knowledge and truth through questioning, discovering, and analyzing, but differs from the construct of intelligence. An imaginational overexcitability is characterized by daydreaming, fantasizing, dramatization, and the use of imagery and metaphors. In order to escape boredom, those with an imaginational overexcitability may create an elaborate imaginary world. An emotional overexcitability is marked by the intensified level of interpersonal relations to people, things, and places, and compassionate feelings for others. Its many manifestations include somatic expressions, extreme and often complex feelings, inhibition, and empathy (Ackerman, 1997; Bouchet & Falk, 2001; Dabrowski & Piechowski, 1977; Piechowski, 1979; Piechowski & Colangelo, 1984).

Findings on Gifted Versus Average Ability Students

Various researchers have examined the relationship between giftedness and overexcitabilities. Although results vary, some researchers have found gifted students to score higher than average-ability students on intellectual, emotional, and imaginational overexcitabilities. For example, Piechowski and Colangelo (1984) examined the overexcitabilities of gifted adolescents, intellectually gifted adults, artists, and average-ability graduate students. Results indicate that both gifted adolescents and gifted adults were characterized by intellectual, emotional, and imaginational overexcitabilities. In addition, the gifted adolescents were found to have higher intellectual overexcitability scores than the graduate students. In line with these results, Gallagher (1986) found gifted, sixth-grade students to report higher intellectual, emotional, and imaginational overexcitability scores than a random sample of average-ability sixth-grade students. Differing results, however, have been reported. For example, a study of tenth and eleventh grade students found gifted students to be characterized by higher intellectual, emotional, and psychomotor overexcitabilities than average-ability students (Ackerman, 1997), and Bouchet and Falk (2001) found gifted college students to score higher than average-ability college students on measures of intellectual and emotional overexcitabilities.

The differing research findings between gifted and average-ability individuals may be due to a methodological issue. Some researchers have used an open-ended, interview framework to gather information related to overexcitabilities (e.g., Gallagher, 1986) and some researchers have used a Likert-style inventory to measure overexcitabilities (e.g., Bouchet & Falk, 2001). Although both groups of researchers are measuring overexcitabilities, the expression of each overexcitability may differ depending on the method used to assess it. Thus, research findings and interpretations may also differ.

Age and Overexeitabilities

A portion of prior research regarding the relationship between overexcitabilities and age has shown similar overexcitability scores across age groups. For instance, no age differences were found in gifted children ages 9-14 for sensual and psychomotor overexcitabilities (Piechowski & Miller, 1995). Piechowski and Colangelo (1984) found gilled adolescents aged 12-17 and gifted adults to report similar scores on the emotional and imaginational overexcitabilities. In addition, no age differences in overexcitability scores have been found among gifted college students (Bouchet & Falk, 2001).

Other studies have reported age differences in overexcitability scores. For example, Kitano (1990) found preschoolers, ages 32 to 55 months, exhibited fewer characteristics of overexcitabilities as they got older. Yet, a study by Piechowski and Miller (1995) reported children ages 12-14 scored higher than children ages 9-11 on emotional, intellectual, and imaginational overexcitabilities. However, Piechowski and Miller suggest this difference is likely due to an increased ability to respond in written form rather than a difference in the strength of the overexcitabilities. Piechowski and Colangelo (1984) found that gifted adolescents score lower than gifted adults in intellectual overexcitability. Piechowski and Colangelo also found gifted adolescents that score lower than gifted adults, and gifted adolescents score lower than average-ability graduate students on a measure of sensual overexcitability, concluding that sensual overexcitability may be a characteristic of age and maturity rather than giftedness.

Gender and Overexcitabilities

The relationship between gender and overexcitabilities has been examined for preschool children (Kitano, 1990), children ages 9-14 (Piechowski & Miller, 1995), college students (Bouchet & Falk, 2001), and adult samples (Miller, Silverman, & Falk, 1994), which leaves a gap in information related to adolescents. In general, no gender differences have been found in samples of young children. For example, Kitano's study of preschoolers found no gender differences in parents' rated scores on a measure used to indicate overexcitabilities. In addition, Piechowski and Miller found no gender differences in the overexcitability scores of gifted children ages 9-14.

While these researchers indicate no gender differences in samples of children under age 14, others report gender differences in adult samples. For instance, Bouchet and Falk (2001) examined the relationship between gender and overexcitabilities among university students. Gifted males scored higher than gifted females on intellectual, imaginational, and psychomotor overexcitabilities, while gifted females scored higher than gifted males on emotional and sensual overexcitabilities. Miller et al. (1994) examined intellectually gifted adults and average-ability graduate students and found females scored higher in emotional overexcitability, while men scored higher in intellectual overexcitability. No gender differences were found for imaginational, sensual, or psychomotor overexcitabilities.

Previous studies, therefore, have reported gender differences only in college students and adults. The limited amount of research performed in this realm warrants additional studies that would examine possible explanations for these differing results. One possible explanation, as proposed by Miller et al. (1994), suggests gender differences in overexcitabilities "seem to be related to areas in which males and females have been differentiated by traditional socialization" (p. 33).

Acceptability of Overexeitabilities

Gifted and average-ability students differ both socially and intellectually. These differences are often viewed as conflicting with their environment (Kitano, 1990). Overexcitabilities are typically not valued socially, but are rather interpreted as "nervousness, hyperactivity, neurotic temperament, excessive emotionality, and emotional intensity that most people find uncomfortable at close range" (Piechowski & Colangelo, 1984, p. 81). Consequently, stronger overexcitabilities typically mean less acceptability for gifted students by their peers and teachers. As a result, these children "are often made to feel different, apart from others, embarrassed, and guilty for being different" (Piechowski, 1992, p. 287). Even further, gifted children's overexcitabilities may be interpreted as psychological problems (e.g., ADHD; Hartnett, Nelson, & Rinn, 2004). Considering these negative consequences of overexcitabilities, it is reasonable to question whether or not gifted students' self-concepts may be affected by their overexcitabilities.

Self-Concept

Self-concept is commonly researched in the social sciences and has received a fair amount of attention in the realm of gifted adolescents. Self-concept can be understood as "the image we hold of ourselves" (Hoge & Renzulli, 1993, p. 449). Self-concept is both multifaceted and hierarchical (Shavelson, Hubner, & Stanton, 1976). Self-concept is multifaceted in that individuals view themselves in various dimensions that are related but can be measured independently of each other (Byrne & Shavelson, 1996). Self-concept is also "hierarchically organized, with perceptions of behavior at the base moving to inferences about self in sub areas (e.g., academic--English, science, history, mathematics), then to inferences about self in general" (Marsh & Shavelson, 1985, p. 107). One's self-concept is derived from various facets, rather than being a unidimensional construct (Shavelson et al.).

Global self-concept can be divided into academic and nonacademic aspects (Byrne & Shavelson, 1996). Academic self-concept, or one's perception of her/his academic ability, can be further broken into specific components, namely mathematics and verbal facets. Nonacademic self-concept can be divided into emotional, social, and physical facets (Bryne & Shavelson). Marsh (1990) describes emotional self-concept as "emotional well-being and freedom from psychopathology" (p. 2). Bryne and Shavelson define social self-concept as "one's perception of his or her social competence with respect to social interaction with others and derives from the assessment of one's behavior within a given social context" (p. 601). Physical self-concept is related to one's perceptions of her/his physical abilities and physical appearance (Marsh).

Grade Level and Self-Concept

While research has shown self-concept to become more differentiated with age (Bryne & Shavelson, 1996), researchers have reported mixed results regarding gifted adolescents. Research has provided support for the argument that general self-concept tends to increase over time as children move into adolescence (Marsh, 1989; Wilson, 1998). However, when the self-concepts of gifted students were exclusively considered, various researchers have found no age differences in the self-concept scores of gifted middle-school and high-school students at a summer program for the gifted (Gabelko, Roth, & Worrell, 1997; Stocking, Porter, & Goldstein, 1993; Worrell, Roth, Gabelko, 1998). Furthermore, Colangelo and Pfleger (1978) and Hoge and Renzulli (1993) reported grade level as having no significant effect on self-concept. However, other studies have found significant grade-level differences in gifted students' self-concepts. For example, Klein and Zehms (1996) found gifted girls' total sell-concept scores significantly decline from grades 3 to 5 to 8.

Gender and Self-Concept

Again, researchers have indicated mixed results regarding gender and the self-concepts of gifted students. Gifted males typically have higher global self-concepts than gifted females (Gabelko et al., 1997; Worrell et al., 1998). Previous research on gifted students in grades 5-12 have reported no gender differences in academic self-concept scores (Ablard, 1997; Colangelo & Pfleger, 1978; Gabelko et al.; Ross & Parker, 1980; Worrell et al.). However, Kelly and Jordan's (1990) study of gifted eighth-grade students revealed boys rated themselves more capable than gifted girls, even though there was no gender difference in actual academic achievement. Contrary to most findings on gifted American students, Dai (2001) found that gifted Chinese adolescent females reported higher academic self-concept than males. Regarding the specific facets of academic self-concept, Norman, Ramsay, Martray, and Roberts (1999) found gifted adolescent girls to score higher than gifted adolescent boys on a measure of verbal self-concept, while boys scored higher than girls on a measure of math self-concept. For Chinese adolescents at a school for the gifted, males and females reported similar scores on math self-concept (Dai).

Concerning gifted students' social self-concept, previous studies indicate mixed results as well. Some studies have reported no gender differences in social self-concept (Kelly & Jordan, 1990; Pyryt & Mendaglio, 1994; Ross & Parker, 1980). For example, Rinn's (2006) study of adolescents at a summer program for the gifted found no gender differences in perceived same-sex or opposite-sex peer relations at the beginning of the summer program. Yet, studies of gifted students in grades 7-11 revealed that females reported higher levels of social self-concept than males (Ablard, 1997; Gabelko et al., 1997; Worrell et al., 1998). In addition, Norman et al. (1999) found that gifted adolescent girls score higher than gifted adolescent boys on measures of perceived honesty/trustworthiness, opposite-sex relations, and same-sex relations. Regarding physical self-concept, gifted males typically report higher physical self-concept scores than gifted females (Gabelko et al.; Worrell et al.). However, Rinn and Wininger (2006) found no gender differences among gifted adolescents on measures of perceived physical abilities or physical appearance.

Current Study

The purpose of the current study is to explore the relationship between gifted adolescents' self-concepts and overexcitabilities. This study will also examine the effects of gender and grade level on the overexcitabilities of gifted students. There are both theoretical and practical implications for this study.

The existing body of research on overexcitabilities is limited, particularly when using samples of gifted adolescents. As self-concept becomes more differentiated with age (Bryne & Shavelson, 1996), early adolescence may be the time period when perceptions of self are emerging. Consequently, this study will offer insight into the initial stage of gifted adolescent self-concept formation, as well as how overexcitabilities may be affecting that self-concept formation. There is a lack of research that examines the possible relationship between overexcitabilities and self-concept, yet the relationship between these two constmcts is important because of the potential negative relationship between the two. Educators and practitioners should be aware of the potential relationship between overexcitabilities and self-concept, so that specific compensations can be made if gifted adolescents' self-concepts decrease as a function of their overexcitabilities. It is also important to be aware of any gender or grade-level effects so that adjustments can be made accordingly.

The current study is an exploratory study that examines the relationship between overexcitabilities and selfconcept. Given the negative consequences associated with overexcitabilities (Kitano, 1990; Piechowski, 1992; Piechowski & Colangelo, 1984), a relationship is hypothesized to exist. The relationship between gender and overexcitabilities is also explored; however, because of the lack of previous research, the direction of the relationship is not hypothesized. In accordance with prior research (e.g., Piechowski & Colangelo; Piechowski & Miller, 1995), no grade-level differences for gifted adolescents' overexcitabilities are hypothesized. Self-concept and gender have been examined recently (e.g., Gabelko et al., 1997; Norman et al., 1999; Rinn, 2006), as well as self-concept and grade level (Gabelko et al.; Klein & Zehms, 1996; Worrell et al., 1998). As a result, these relationships are not explored, as they are beyond the scope of this paper.

Method

Participants

Participants were recruited from two summer programs for gifted students held at a comprehensive university in the south. These particular summer programs have been in operation for over 20 years. The first summer program is a 2-week, largely residential program (1) for gifted students entering the seventh, eighth, or ninth grades the following school year. To qualify for participation in this summer program, students must show (a) high interest and/or achievement in one or more content areas; (b) be eligible for services as a gifted child or have an IQ score of 125 or above; (c) score at or above the 90th percentile on the total battery; or at or above the 95th percentile on the total mathematics or language/reading section of the most recent achievement test; or have scored at the proficient or distinguished level on performance assessment measures; and (d) be nominated by a teacher, counselor, or principal. This summer program involves 6 hours of class per day, 5 days a week, for 2 weeks. The students have a variety of courses from which to choose (e.g., acting, ecology, geography, science, etc.), and they enroll in four courses. The students also engage in various social activities (e.g., board games, athletic activities, a talent show, etc.) after class each day and on weekends.

The second summer camp is a 3-week residential program for gifted students entering the eighth, ninth, tenth, or eleventh grades the following school year. To qualify for participation in this summer program, students must have been eligible to attend talent search summer programs (e.g., through the Duke Talent Identification Program) within the past 4 years. This summer program involves 6 hours of class and 1 hour of study hall per day, 5 days a week, for 3 weeks. The students have a variety of courses from which to choose (e.g., humanities, genetics, theatre, mathematics, etc.), and they enroll in only one course. Similar to the other summer program, the students also engage in various social activities after class each day and on weekends.

A total of 248 gifted adolescents participated in this study. Although parental consent was obtained for 250 adolescents, 2 individuals declined participation. Of the 248 participants, 119 attended the first summer program while 129 attended the second summer program. There were 131 male participants and 116 female participants. The mean age of the participants was 13.6, with a range from 11 to 16. Approximately 85% of the participants were White. Complete demographic information can be found in Table 1.

Materials

Demographic information. Participants were given a demographic questionnaire to assess gender and age, among other information. Other data was gathered from participants' applications for summer camp participation, including ethnic background and grade level.

Overexcitabilities. The Overexcitabilities Questionnaire-Two (OEQII) was designed to measure the five forms of overexcitability: psychomotor, sensual, imaginational, intellectual, and emotional (Falk, Lind, Miller, Piechowski, & Silverman, 1999). Overexcitabilities derive from Dabrowski's theory of positive disintegration. The OEQII is a fairly new instrument that has been used to study giftedness (e.g., Bouchet & Falk, 2001).

The OEQII includes 50 self-report items using a 5-point Likert scale with responses ranging from 1 (not at all like me) to 5 (very much like me). High scores indicate higher levels of overexcitability. From the normative sample, internal reliability, using Cronbach's alpha, was high for each form of overexcitability: psychomotor (.86), sensual (.89), imaginational (.85), intellectual (.89), and emotional (.84). Each overexcitability subscale is made up of 10 items. An example from the psychomotor overexcitability subscale is, "When I have a lot of energy, I want to do something really physical." A sample item for the sensual overexcitability subscale is, "Viewing art is a totally absorbing experience." The intellectual overexcitability scale includes items such as, "Theories get my mind going," while an example from the imaginational overexcitability subscale is, "Things that I picture in my mind are so vivid that they seem real to me." A sample item for the emotional overexcitability subscale is, "I can be so happy that I want to laugh and cry at the same time."

Self-concept. The Self Description Questionnaire II (SDQII) was designed to measure the self-concepts of young adolescents, and is theoretically based on the notion that self-concept is multidimensional and hierarchically structured (Marsh, 1990; Shavelson et al., 1976). The SDQ-II measures self-concept in the following areas: mathematics, verbal, physical abilities, physical appearance, same-sex peer relations, opposite-sex peer relations, parent relations, emotional stability, honesty/trustworthiness, general academic, and general self. Extensive support for the reliability and validity of the SDQ-II has been reported in other research (see Gilman, Laughlin, & Huebner, 1999; Plucker, Taylor, Callahan, & Tomchin, 1997).

The SDQ-II consists of 11 subscales. The same-sex peer-relations subscale measures interactions with peers of the same sex. A sample item from this subscale is, "Not many people of my own sex like me" (Marsh, 1990, p. 4). From the normative sample, internal consistency was reported as .86 and factor loadings range from .57 to .68 (five items) (2). The opposite-sex peer-relations subscale measures interactions with peers of the opposite sex (e.g., "I have lots of friends of the opposite sex"; p. 3). Internal consistency was reported as .90 and factor loadings range from .69 to .78 (four items).

The mathematics subscale measures ability, enjoyment, and interest in mathematics and reasoning (e.g., "I do badly in tests of mathematics"; p. 5). From the normative sample, internal consistency was reported as .90 and factor loadings range from .72 to .80 (five items). The verbal subscale measures ability, enjoyment, and interest in English and reading (e.g., "I learn things quickly in English classes"; p. 5). Internal consistency was reported as .86 and factor loadings range from .53 to .75 (five items). The general school subscale measures one's interests and abilities in schoolwork (e.g., "I learn things quickly in most school subjects"; p. 6). From the normative sample, internal consistency was reported as .87 and factor loadings range from .48 to .64 (five items). The general self subscale measures one's feeling of self-worth, self-confidence, and self-satisfaction (e.g., "If I really try I can do almost anything I want to"; p. 6). From the normative sample, internal consistency was reported as .88 and factor loadings range from .49 to .64 (five items). The emotional stability subscale measures one's emotional well-being and freedom from emotional dysfunction (e.g., "Other people get more upset about things than I do"; p. 5). From the normative sample, internal consistency was reported as .83 and factor loadings range from .57 to .66 (five items). The physical abilities subscale measures skills and interest in physical activities and sports (e.g., "I'm good at things like sports, gym and dance"; p. 3). From the normative sample, internal consistency was reported as .85 and factor loadings range from .67 to .78 (four items). The physical appearance subscale measures perceived physical attractiveness (e.g., "Nobody thinks that I'm good looking"; p. 3). From the normative sample, internal consistency was reported as .91 and factor loadings range from .68 to .76 (four items). The parent relations subscale measures perceived interactions with parents (e.g., "I get along well with my parents"; p. 4). From the normative sample, internal consistency was reported as .87 and factor loadings range from .68 to .77 (four items). The honesty/trustworthiness subscale measures truthfulness and dependability (e.g., "When I make a promise I keep it"; p. 4). From the normative sample, internal consistency was reported as .84 and factor loadings range from .61 to .71 (five items).

Procedure

Parental consent was obtained prior to the start of the summer program. Adolescents whose parents gave consent were invited to take part in the study, but they were given the option to decline participation. Data was gathered at a single session during the first week of each summer program.

Results

Overexcitability and Self-Concept

To examine the relationship between overexcitability, self-concept, and grade level, correlation analyses were used. A correlation matrix of all the variables of interest can be found in Table 2. Significant correlations were found between grade level and opposite-sex peer relations self-concept (r =. 18, p < .01), physical appearance self-concept (r =. 14, p < .05), emotional stability self-concept (r =. 19, p < .01), the intellectual overexcitability scores (r = .22, p < .01), and the imaginational overexcitability scores (r = -.14, p < .05). Significant correlations were also found for the psychomotor overexcitability scores and same-sex peer relations self-concept (r = .32, p < .001), opposite-sex peer relations self-concept (r = .29, p < .001), physical appearance self-concept (r = .24, p < .001), general school self-concept (r = .22, p < .01), general self-concept (r = .28, p < .001), and physical abilities self-concept (r = .68, p < .001). Significant correlations were found for the sensual overexcitability scores and verbal self-concept (r = .29, p < .001). In addition, the intellectual overexcitability scores were significantly correlated with emotional stability self-concept (r = -.14, p < .05), general school self-concept (r = .29, p < .001), and verbal self-concept (r = .33, p < .001). The imaginational overexcitability scores were significantly correlated with same-sex peer relations self-concept (r = -.23, p < .001), opposite-sex peer relations self-concept (r = -.22, p < .01), physical appearance self-concept (r = -.28, p < .001), parent relations self-concept (r = -.26, p < .001), emotional stability self-concept (r = -.45, p < .001), math self-concept (r = -.25, p < .001), general self-concept (r = -.26, p < .001), honesty/trustworthiness self-concept (r = -. 18 p < .01), verbal self-concept (r =. 14, p < .05), and physical abilities self-concept (r = -.30, p < .001). Lastly, the emotional overexcitability scores were significantly correlated with emotional stability self-concept (r = -.41, p < .001), math self-concept (r=-.13, p < .05), honesty/trustworthiness self-concept (r =. 15, p < .05), and verbal self-concept (r = .21, p < .01).

Overexcitabilities, Grade Level, and Gender

To examine differences between gifted adolescents' grade level and gender with regards to each of the five overexcitabilities, a series of two-way factorial analyses of variance (ANOVAs) were utilized. Specifically, we examined whether each dependent variable (e.g., overexcitability subscale score) was affected by the adolescent's gender (male, female) and/or grade level (sixth, seventh, eighth, ninth, tenth). The means, standard deviations, and coefficient alphas of each dependent variable can be seen in Table 3. Although several of the dependent variables for this study were found to be somewhat intercorrelated, a MANOVA approach was not used because the dependent variables were not correlated beyond a moderate level. Results of the series of factorial ANOVAs suggested no significant main effects for adolescents' psychomotor overexcitability subscale scores. Significant findings were found for each of the following subscale scores: sensual, intellectual, imaginational, and emotional subscale scores.

Sensual overexcitability subscale scores. Regarding the sensual overexcitability, there was a main effect for gender, F(1,212) = 18.08, p < .001, partial [[eta].sup.2] = .08. Gifted adolescent females had higher sensual overexcitability scores (M = 3.44, SD = 0.79) than gifted adolescent males (M = 2.86, SD = 0.91). No other significant main effects or interactions were found. See Table 4 for more information.

Intellectual overexcitability subscale scores. Regarding the intellectual overexcitability, there was a main effect for grade level, F(4, 217) = 4.64, p < .01, partial [[eta].sup.2] = .08. A Tukey HSD post-hoc test revealed significant differences between the intellectual overexcitability subscale scores of sixth graders and eighth graders and between sixth graders and ninth graders. Gifted eighth-grade students had higher intellectual overexcitability scores (M = 3.85, SD = 0.76) than gifted sixth-grade students (M = 3.31, SD = 0.77). In addition, gifted ninth-grade students had higher intellectual overexcitability scores (M = 4.00, SD = 0.64) than gifted sixth-grade students (M = 3.31, SD = 0.77). No other significant main effects or interactions were found. See Table 5.

Imaginational overexcitability subscale scores. Regarding the imaginational overexcitability, there was a main effect for gender, F(1,213) = 9.36, p < .01, partial [[eta].sup.2] = .04. Gifted adolescent females had higher imaginational overexcitability scores (M = 2.96, SD = 1.01) than gifted adolescent males (M = 2.48, SD = 0.79). No other significant main effects or interactions were found. See Table 6.

Emotional overexcitability subscale scores. Regarding the emotional overexcitability, there was a main effect for gender, F(1,220) = 19.59, p < .001, partial [[eta].sup.2] = .15. Gifted adolescent females had higher emotional overexcitability scores (M = 3.63, SD = 0.67) than gifted adolescent males (M = 2.99, SD = 0.73). No other significant main effects or interactions were found. See Table 7.

Discussion

The purpose of this study was to explore the relationship between gifted adolescents' overexcitabilities and self-concepts, while also examining the effects of gender and grade level on gifted adolescents' overexcitabilities. Results from this study indicate significant gender differences for the sensual, imaginational, and emotional overexcitability subscale scores. In addition, there were significant grade-level differences for the intellectual overexcitability subscale scores. No gender or grade-level difference was found for the psychomotor overexcitability subscale scores. Further, multiple relationships were found between overexcitabilities, self-concept, and grade level.

Gender Differences in the Overexcitabilities

Psychomotor overexcitability. Regarding the psychomotor overexcitability, the current study of gifted adolescents found no gender differences, which is in accordance with some previous research (Kitano, 1990; Miller et al., 1994; Piechowski & Miller, 1995). Consequently, it appears as though gender differences in the psychomotor overexcitability are only evident in college students.

Sensual overexcitability. In the current study, gifted adolescent females report higher sensual overexcitabilities than males, which is consistent with previous research of gifted college students (Bouchet & Falk, 2001), yet inconsistent with research on gifted people of other ages (Kitano, 1990; Miller et al., 1994; Piechowski & Miller, 1995). This gender difference is consistent with typical stereotypes and may be a result of the particular developmental frame of time characterizing adolescence and the college years. Further, because the effect size for this finding is so small, gender differences may continue to emerge and grow stronger as students continue into college.

Intellectual overexcitability. Regarding the intellectual overexcitability, previous research has found no gender differences in younger samples (Kitano, 1990; Piechowski & Miller, 1995), yet males in older samples have reported higher scores on measures of intellectual overexcitabilities (Bouchet & Falk, 2001; Miller et al., 1994). The current study of gifted adolescents found no gender difference, which suggests gender differences may emerge in college.

Imaginational overexcitability. The current study of gifted adolescents found females scored higher than males on imaginational overexcitability, a finding which is inconsistent with previous research (Bouchet & Falk, 2001; Kitano, 1990; Miller et al., 1994; Piechowski & Miller, 1995). Gender differences are seemingly only evident in samples of adolescents and college students, with adolescent females and college males scoring higher than their respective counterparts. However, the effect size for this finding is very small. Before we can draw conclusions about these discrepancies in findings, more research needs to be conducted.

Emotional overexcitability. Gifted adolescent females in he current study reported higher emotional overexcitability scores than males. This finding is consistent with previous research on older samples (Bouchet & Falk, 2001; Miller et al., 1994), yet inconsistent with previous research on younger samples (Kitano, 1990; Piechowski & Miller, 1995). Thus, it appears as though the emotional overexcitability gender differences may emerge in adolescence. Again, though, the effect size for this finding is small, so the practical significance may be limited.

Grade Level Findings for the Overexcitabilities

Results of the current study suggest grade-level differences in only one of the five overexcitabilities. Specifically, gifted eighth- and ninth-grade students had higher intellectual overexcitability scores than gifted sixth-grade students. The current findings are consistent with the findings of Piechowski and Miller (1995), who found that 12- to 14-year-olds score higher than 9 to 11-year-olds on a measure of intellectual overexcitability. A possible explanation for the current findings may be the eighth- and ninth-grade students' increased anticipation of high-school and academic achievement, as compared to the sixth graders who are just entering middle school and have not yet been subject to the increasing intellectual demands and expectations of middle school and high school. Because the effect size related to these findings is very small, grade-level differences should be further examined.

Overexcitabilities and Self-Concept

As hypothesized, numerous correlations were found between the five overexcitabilities and the self-concept subscales. The results of the current study lend support to the proposed theory that the social value and consequences associated with overexcitabilities may affect gifted adolescents self-concepts (Kitano, 1990; Piecbowski, 1992; Piechowski & Colangelo, 1984). Each of the significant correlations is relatively weak, though, so the practical relationship between the two constructs may be limited.

Emotional stability self-concept. The emotional stability subscale scores are negatively correlated with three of the five overexcitabilities, including the intellectual, imaginational, and emotional overexcitabilities. These negative relationships may be explained in terms of the negative connotations often associated with overexcitabilities. As Piechowski and Colangelo (1984) suggest, overexcitabilities are not typically valued socially and are perceived negatively by others. As a result, peers and teachers of gifted students are less accepting of the manifestations of overexcitabilities. Piechowski (1992) suggests that gifted students may feel set apart from others because of their differences and even embarrassed. Consequently, it is possible that the acceptability of the overexcitabilities may be negatively affecting gifted adolescents' emotional stability self-concept. However, "Dabrowski (1972) believed that many individuals who exhibit neurotic characteristics (i.e., intense inner conflict, feelings of inferiority regarding one's self, dissatisfaction with one's self, feelings of inadequacy, disquietude, anxiety) also possess the greatest potential for development" (as cited in O'Conner, 2002, p. 53). Thus, gifted students who are experiencing lower emotional self-concepts may actually reach a higher level in Dabrowski's (1964) Theory of Positive Disintegration. Future research should explore the relationship between emotional self-concept, overexcitabilities, and the levels in Dabrowski's theory in greater detail.

Verbal self-concept. Of the 11 self-concept subscales cores, the verbal self-concept scores are the most correlated with the overexcitabilities. Specifically, it is positively correlated with each overexcitability except psychomotor. Prior research suggests gifted adolescent females perceive their verbal abilities as higher than males (Norman et al., 1999). Further, the current study found gifted adolescent females to score higher than males in the emotional, sensual, and imaginational overexcitabilities. Consequently, the correlations round between verbal self-concept and the emotional, sensual, and imaginational overexcitabilities may be explained by gender as a confounding variable. Future research should aim to examine gender as a moderator in the relationship between overexcitabilities and verbal self-concept. Regarding the positive correlation between verbal self-concept and the intellectual overexcitability, the academic nature of each is a likely explanation.

Psychomotor overexcitability. Of the five overexcitabilities, the psychomotor overexcitability scores are the most positively correlated with the self-concept subscale scores, namely the same-sex peer relations, opposite-sex peer relations, physical appearance, general school, general self, and physical abilities subscale scores. Except general school self-concept, all of the positive correlations are with nonacademic self-concept subscale scores. Although not the case for all students, one may expect a psychomotor overexcitability to inhibit academic success, as the manifestations are often distracting for a student. However, it seems as though a psychomotor overexcitability boosts self-concept, even general school self-concept. Educators and practitioners may benefit from the results of this study as they encounter students with a psychomotor overexcitability. As the psychomotor overexcitability may be positively related to various facets of self-concept, allowing the psychomotor overexcitability to flourish, rather than working to suppress it, may actually improve gifted adolescents' self-concept, which may in turn enhance other aspects of their lives, including academic performance.

Imaginational overexcitability. Of the five overexcitabiliies, the imaginational overexcitability is clearly the most negatively correlated with self-concept. More specifically, the imaginational overexcitability is negatively correlated with all but 1 of the 11 self-concept subscale scores. The overexcitabilities are supposed to contribute to the potential for further growth (Bouchard, 2004; Dabrowski, 1964; Piechowski & Miller, 1995), but given the negative association between the imaginational overexcitability and self-concept, gifted adolescents with this overexcitability may be at risk for not reaching their developmental potential. Consequently, specific compensations must be made while working with students who appear to have an imaginational overexcitability. For example, educators and practitioners should be intentional about recognizing the overexcitability and channeling their intensified experience of the world appropriately. Specific attention should be given to the self-concepts of gifted adolescents with an imaginational overexcitability, as the image they hold of themselves may be prone to be negatively affected. Also, given the negative correlation between grade level and the imaginational overexcitability, it may be vital for educators and practitioners to be aware of the manifestations associated with this overexcitability as students get older.

Limitations and Directions for Future Research

Results of the current study open a new window to the investigation of the relationship between gifted adolescents' overexcitabilities and self-concepts. Continued exploration of the effects of the overexcitabilities on self-concept is necessary to further solidify the current findings. However, future research is needed to determine if the relationships between overexcitabilities and self-concept are causal relationships rather than just correlational relationships. Future research should examine specific findings of the current study in greater detail using an experimental design in order to examine whether or not overexcitabilities predict and/or explain self-concept, or whether or not the relationship is reciprocal. In addition, a longitudinal study is suggested, especially when further exploring grade-level differences in overexcitabilities.

As the current sample consisted only of gifted students from a summer program, replication of the current study in settings other than a summer program, such as in the regular classroom, as well as using a more diverse sample is suggested. Future research is also needed to explore the relationships between overexcitabilities and self-concepts of younger and older gifted students, as well as compared to average-ability students. Examining different levels of giftedness with regard to overexcitabilities and self-concept might also provide interesting results.

Finally, when examining overexcitabilities and self-concept in any future research study, researchers should consider using varying methods to measures overexcitabilities. As mentioned earlier, researchers typically use one of two methods to examine overexcitabilites: an interview method or a Likert-style inventory. These two methods could result in differing responses. In addition to the OEQII, which was used in the current study, another commonly used measure to evaluate Dabrowski's five overexcitabilities is called the Overexcitability Questionnaire. It consists of 21 open-ended interview questions and the participant is encouraged to answer freely (Piechowski & Miller, 1995). Piechowski and Miller suggest interviewing a participant will result in a richer response than a standard questionnaire. Future research might include both a Likert-style inventory and an interview method to gain a more complete understanding of overexcitabilities.

Conclusion

The present study contributes to the developing literature on gifted adolescents' overexcitabilities, particularly with regard to the relationship between overexcitabilities and self-concepts. Results suggest gifted adolescent females score higher than males in the sensual, imaginational, and emotional overexcitabilities. In addition, results indicate significant grade-level differences for the intellectual overexcitability subscale scores, but no gender or grade-level difference for the psychomotor overexcitability subscale scores. Further, this study revealed multiple relationships among overexcitabilities, self-concept, and grade level. It is suggested that the acceptability of overexcitabilities may be affecting gifted adolescents' self-concepts. More research is needed, however, to clarify if the relationship is causal.

Manuscript submitted August 7, 2006.

Revision accepted December 7, 2006.

REFERENCES

Ablard, K. E. (1997). Self-perceptions and needs as a function of type of academic ability and gender. Roeper Review 20, 100-115.

Ackerman, C. M. (1997). Identifying gifted adolescents using personality characteristics: Dabrowski's overexcitabilities. Roeper Review, 19, 229-237.

Bouchard, L. L. (2004). An instrument for the measure of Dabrowskian overexcitabilities to identity gifted elementary students. Gifted Child Quarterly; 48, 339-350.

Bouchet, N., & Falk, R. F. (2001). The relationship among giftedness, gender, and overexcitability. Gifted Child Quarterly, 45, 260-267.

Byrne, B. M., & Shavelson, R. J. (1996). On the structure of social self-concept for pre-, early, and late adolescents: A test of the Shavelson, Hubner, and Stanton (1976) model. Journal of Personality of Social Psychology, 70, 599-613.

Colangelo, N., & Pfleger, L. R. (1978). Academic self-concept of gifted high school students. Roeper Review, I, 10-11.

Dabrowski, K. (1937). Psychological basis of sell-mutilation. Genetic Psychology Monographs, 19, 1-104.

Dabrowski, K. (1964). Positive disintegration. Boston: Little, Brown.

Dabrowski, K. (1972). Psychoneurosis is not an illness. London: Gryf.

Dabrowski, K., Kawczak, A., & Piechowski, M. M. (1970). Mental growth through positive disintegration. London: Gryf.

Dabrowski, K., & Piechowski, M. M. (1977). Theory of levels of emotional developmental (Vols. 1 & 2). Oceanside, NY: Dabor Science.

Dai, D. Y. (2001). A comparison of gender differences in academic self-concept and motivation between high-ability and average Chinese adolescents. Journal of Secondary Gifted Education, 13, 22-32.

Falk, R. E, Lind, S., Miller, N. B., Piechowski, M. M., & Silverman, L. K. (1999). The overexcitability questionnaire-two (OEQII): Manual, scoring system, and questionnaire. Denver, CO: Institute for the Study of Advanced Development.

Gabelko, N. H., Roth, D. A., & Worrell, F. C. (1997, March). Age and gender differences in global, academic, social, and athletic self-concepts. Paper presented at the meeting of the American Educational Research Association, Chicago, IL.

Gallagher, S. A. (1986). A comparison of the concept of overexcitabilities with measures of creativity and school achievement in sixth-grade students. Roeper Review, 8, 115-119.

Gilman, R., Laughlin, J. E., & Huebner. E. S. (1999). Validation of the Self-Description Questionnaire-II with an American sample. School Psychology International, 20, 300-307.

Hartnett, D. N., Nelson, J. M., & Rinn, A. N. (2004). Gifted or ADHD? The possibilities of misdiagnosis. Roeper Review, 26, 73-77.

Hoge, R. D., & Renzulli, J. S. (1993). Exploring the link between giftedness and self-concept. Review of Educational Research, 63, 449-465.

Kelly, K. R., & Colangelo, N. (1984). Academic and social self-concepts of gifted, general, and special students. Exceptional Children, 50, 551-554.

Kelly, K. R., & Jordan, L. K. (1990). Effects of academic achievement and gender on academic and social self-concept: A replication study. Journal of Counseling & Development, 69, 173-177.

Kitano, M. K. (1990). Intellectual abilities and psychological intensities in young children: Implications for the gifted. Roeper Review, 13, 5-11.

Klein, A. G., & Zebras, D. (1996). Self-concept and gifted girls: A cross sectional study of intellectually gifted females in grades 3, 5, 8. Roeper Review, 19, 30-35.

Marsh, H. W. (1989). Age and sex effects in multiple dimensions of self-concept: Preadolescence to early adulthood. Journal of Educational Psychology, 81, 417-430.

Marsh, H. W. (1990). Self Description Questionnaire (SDQ) II: Manual. New South Wales. Australia: University of Western Sydney.

Marsh, H. W., & O'Neill, R. (1984). Self Description Questionnaire III: The construct validity of multidimensional self-concept ratings by late adolescents. Journal of Educational Measurement, 21, 153-174.

Marsh, H. W., & Shavelson, R. (1985). Self-concept: Its multifaceted, hierarchical structure. Educational Psychologists. 20, 107-123.

Miller, N. B., Silverman, L. K., & Falk, R. F. (1994). Emotional development, intellectual ability, and gender. Journal for the Education of the Gifted, 18, 20-38.

Norman, A. D., Ramsay, S. G., Martray, C. R., & Roberts, J. L. (1999). Relationship between levels of giftedness and psychosocial adjustment. Roeper Review, 22, 5-10.

O'Conner, K. J. (2002). The application of Dabrowski's theory to the gifted. In M. Neihart, S. M. Reis, N. M. Robinson, & S. M. Moon (Eds.), The social and emotional development of gifted children: What do we know? (pp. 51-60). Washington, DC: Prufrock Press.

Piechowski, M. M. (1979). Developmental potential. In N. Colangelo & R. R. Zaffrann (Eds.), New voices in counseling the gifted (pp. 25-58). Dubuque, IA: Kendall/Hunt.

Piechowski, M. M. (1992). Emotional development and emotional giftedness. In N. Colangelo & G. Davis (Eds.), Handbook of gifted education (pp. 285-305). Needham Heights, MA: Allyn & Bacon.

Piechowski, M. M., & Colangelo, N. (1984). Developmental potential of the gifted. Gifted Child Quarterly, 28, 80-88.

Piechowski, M. M., & Cunningham, K. (1985). Patterns of overexcitability in a group of artists. Journal of Creative Behavior, 19, 153-174.

Piechowski, M. M., & Miller, N. (1995). Assessing developmental potential in gifted children: A comparison of methods. Roeper Review 17, 176-181.

Plucker, J. A., Taylor, J. W., V, Callahan, C. M., & Tomchin, E. M. (1997). Mirror, mirror, on the wall: Reliability and validity evidence for the Self-Description Questionnaire II with gifted students. Educational and Psychological Measurement, 57, 704-713.

Pyryt, M. C., & Mendaglio, S. (1994). The multidimensional self-concept: A comparison of gifted and average-ability adolescents. Journal for the Education of the Gifted, 17. 299-305.

Rinn, A. N. (2006). Effects of a summer program on the social self-concepts of gifted adolescents. Journal of Secondary Gifted Education, 17, 65-75.

Rinn, A. N., & Wininger, S. R. (2006). Sports participation among academically gifted adolescents: Relationship to the multidimensional self-concept. Manuscript submitted for publication.

Ross, A., & Parker, M. (1980). Academic and social self concepts of the academically gift ed. Exceptional Children, 47, 6-10.

Shavelson, R. J., Hubner, J. J., & Stanton, G. C. (1976). Validation of construct interpretations. Review of Educational Research, 46, 407-441.

Stocking, V. B., Porter, L. C., & Goldstein, D. (1993, April). The self-concept of talented adolescents in an intensive summer program: Implications for programming. Paper presented at the Annual Meeting of the American Educational Research Association, Atlanta, GA.

Wilson, P. L. (1998). Multidimensional Self Concept Scale: An examination of grade, race, and gender differences in third through sixth grade students' sell-concepts. Psychology in the Schools, 35, 317-326.

Worrell, F. C., Roth, D. A., & Gabelko, N. H. (1998). Age and gender differences in the self-concepts of academically talented students. Journal of Secondary Gifted Education, 9, 157-162.

Endnotes

(1) A non-residential option is chosen by about 20% of participants.

(2) Marsh (1990) uses" item pairs in factor analysis such that the eight or ten items from each subscale of the SDQ-II are divided into four or five item pairs. For more information, see Marsh and O'Neill (1984).

Candace M. Gross is a senior psychology major at Northwestern College in Orange City, IA. This research was conducted during the summer Research Experience for Undergraduates Program at Western Kentucky University. During the fall of 2007, she plans on attending graduate school in School Psychology. She is interested in focusing on parent-educator-community collaboration in students' education. E-mail: [e-mail]

Anne N. Rinn is an assistant professor of psychology at Western Kentucky University. She holds a PhD in educational psychology from Indiana University. Her research focuses on the academic, social, and emotional development of gifted adolescents and college students, as well as the effects of gifted programming on student development as a whole. E-mail: [e-mail]

Kelly M. Jamieson is a senior psychology major at Xavier University. She conducted this research as part of her participation in the Research Experience for Undergraduates Program at Western Kentucky University. She plans to enter a graduate psychology program in the fall of 2007. E-mail: [e-mail]

The authors greatly appreciate the assistance and support of Dr. Julia Link Roberts, Center for Gifted Studies, Western Kentucky University.

This research was supported in part by the Research Experience for Undergraduates Grant 0453454 from the National Science Foundation.

Correspondence concerning this article should be addressed to Anne N. Rinn, PhD, Department of Psychology, Western Kentucky University, 1906 College Heights Blvd. #21030, Bowling Green, KY 42101-1030. E-mail: [e-mail]

Table 1
Demographic Information
(N = 248)

                                      n (Percentage)
Gender
  Male                                  131 (52.8)
  Female                                116 (46.8)
  Not indicated                           1 (0.4)

Age
  11                                      8 (3.2)
  12                                     46 (18.5)
  13                                     74 (39.8)
  14                                     60 (24.2)
  15                                     31 (12.5)
  16                                     29 (11.7)

Grade Completed
  Sixth                                  34 (13.7)
  Seventh                                78 (31.5)
  Eighth                                 64 (25.8)
  Ninth                                  34 (13.7)
  Tenth                                  29 (11.7)
  Not indicated                           9 (3.6)

Race/Ethnicity
  American Indian/Alaska Native           1 (0.4)
  Asian or Pacific Islander              16 (6.5)
  Biracial                                2 (0.8)
  Black/African American                  3 (1.2)
  Hispanic or Latino                      2 (0.8)
  White                                 210 (84.7)
  Not indicated                           7 (2.8)

Program
  Two-Week Program                      119 (48.0)
  Three-Week Program                    129 (52.0)

Table 2
Correlations of Variables of Interest

                           Psychomotor       Sensual      Intellectual
                               OE              OE              OE

Grade level                 -.03             .10             .22 **
Same-sex SC                  .32 ***        -.03            -.04
Opposite-sex SC              .29 ***         .04            -.03
Physical appearance SC       .24 ***        -.09            -.10
Parent relations SC          .11            -.09            -.02
Emotional stability SC       .10            -.11            -.14 *
Math SC                      .07            -.12            -.03
General school SC            .22             .02             .29 ***
General SC                   .28 ***        -.06             .11
Honesty/Trust SC             .03             .01             .07
Verbal SC                   -.02             .29 ***         .33 ***
Physical abilities SC        .68 ***        -.10            -.06
Psychomotor OE              1                --              --
Sensual OE                   .08            1                --
Intellectual OE              .24 ***         .43 ***        1
Imaginational OE             .07             .50 ***         .34 ***
Emotional OE                 .21 **          .58 ***         .43 ***

                          Imaginational     Emotional
                               OE              OE

Grade level                 -.14 *            .03
Same-sex SC                 -.23 ***         -.01
Opposite-sex SC             -.22 **           .09
Physical appearance SC      -.28 ***         -.10
Parent relations SC         -.26 ***         -.02
Emotional stability SC      -.45 ***         -.41 ***
Math SC                     -.25 ***         -.13 *
General school SC           -.10              .02
General SC                  -.26 ***          .01
Honesty/Trust SC            -.18 **           .15 *
Verbal SC                    .14 *            .21 **
Physical abilities SC       -.30 ***         -.05
Psychomotor OE                --              --
Sensual OE                    --              --
Intellectual OE               --              --
Imaginational OE             1                --
Emotional OE                 .54 ***         1

Note. SC = Self-concept. OE = Overexcitability.

* p < .05. ** p < .01. *** p < .001.

Table 3
Means, Standard Deviations, and
Coefficient Alphas of Variables of Interest

Subscale scores                       Total          Coefficient
                                    mean (SD)         alpha (a)

Mathematics                        4.91 (1.04)           .91
Verbal                             4.99 (0.91)           .87
Physical abilities                 4.44 (l.30)           .93
Physical appearance                4.44 (l.08)           .93
Same-sex peer relations            5.16 (0.85)      .931, .89 (b)
Opposite-sex peer relations        3.75 (0.86)      .911, .92 (b)
Parent relations                   5.04 (0.96)           .90
Emotional stability                4.30 (l.06)           .88
Honesty/Trustworthiness            4.99 (0.86)           .87
General academic                   5.55 (0.49)           .80
General self                       5.32 (0.73)           .90
Psychomotor overexcitability       3.29 (0.86)           .87
Sensual overexcitability           3.15 (0.90)           .88
Intellectual overexcitability      3.70 (0.75)           .87
Imaginational overexcitability     2.73 (0.94)           .89
Emotional overexcitability         3.30 (0.76)           .82

(a) For males. (b) For females.

Table 4
Factorial ANOVA Results for Gender,
Grade Level, and the
Sensual Overexcitability Scores

                           Sum of                         Mean
Source                    squares           df           square

Grade level                 2.32            4             0.58
Gender                     13.24            1            13.24
Grade level x gender        1.40            4             0.35
Error                     155.24          212             0.73

                                   Partial

Source                    F ratio     [[eta].sup.2]

Grade level                0.79            .02
Gender                    18.08 ***        .08
Grade level x gender       0.48            .01
Error

Note. [R.sup.2] = .13 (Adjusted [R.sup.2] = .09)

*** p < .001.

Table 5
Factorial ANOVA Results for Gender,
Grade Level, and the
Intellectual Overexcitability Scores

                           Sum of                         Mean
Source                    squares           df           square

Grade level                 9.58            4             2.40
Gender                      0.01            1             0.01
Grade level x gender        2.62            4             0.66
Error                     111.91          217             0.52

                                   Partial

Source                    F ratio     [[eta].sup.2]

Grade level                 4.64            .08
Gender                      0.02            .00
Grade level x gender        1.27            .02
Error

Note. [R.sup.2] = .10 (Adjusted [R.sup.2] = .06)

** P < .01.

Table 6
Factorial ANOVA Results for Gender,
Grade Level, and the
Imaginational Overexcitability Scores

                           Sum of
Source                     square                         Mean
                                            df           square

Grade level                 3.74            4             0.94
Gender                      7.57            1             7.57
Grade level x gender        2.64            4             0.66
Error                     172.36          213             0.81

                                   Partial
Source
                          F ratio     [[eta].sup.2]

Grade level                 1.16           .02
Gender                      9.36 *         .04
Grade level x gender        0.82           .02
Error

Note. [R.sup.2] = .10 (Adjusted [R.sup.2] = .06)

* p < .05.

Table 7
Factorial ANOVA Results for Gender,
Grade Level, and the
Emotional Overexcitability Scores

                           Sum of                         Mean
Source                    squares           df           square

Grade level                 3.41            4             0.85
Gender                     19.59            1            19.59
Grade level x gender        0.48            4             0.12
Error                      108.50          220            0.49

                                  Partial

Source                    F ratio     [[eta].sup.2]

Grade level                 1.73           .03
Gender                     39.73           .15
Grade level x gender        0.25           .00
Error

Note. [R.sup.2] = .20 (Adjusted [R.sup.2] = .17)

*** p < .001.