Analyses of variance were performed to compare the gifted and nongifted students' responses on the scales. Ethnicity and parental education levels were entered on covariates. Gifted students, as compared with nongifted peers, perceived themselves as being more intimate with best friends, as assuming fewer family responsibilities, and as taking greater risks in sports and dangerous activities (see Table 2).
Consistent with rating themselves as being more intimate with friends, gifted students also indicated on relationship items from the Background Information Questionnaire that they were closer to friends than to family. For example, approximately 62% indicated "friend" in response to the statement, "There is an important person in my life," while an equivalent number of nongifted students indicated "parent and friend" ([[Chi].sup.2] = 9.0, p [less than] .005). Similarly, when answering the question, "I feel closer to my friends than to my family," 42% of the gifted students responded often or very often as compared with 22% of the nongifted students ([[Chi].sup.2] = 18.4, p [less than] .001).
Goodness-of-fit chi-square tests were conducted on each of the items on the Perceptions about Giftedness Scale (see Table 3). On all of the perceived academic and social skills items, the gifted students rated themselves as the same as, or better than, their nongifted peers. The distribution of responses was generally equal across "same as peer" and "better than peer" categories. On several items, however, they thought they were superior to their peers, including spelling/writing performance, originality, and creativity, as well as getting along with teachers, getting along with adults, and understanding people.
Independent t tests were used for post hoc comparisons between the responses of gifted students and their teachers on the Self-Perceived Academic Skills, Self-Perceived Social Skills, and the Unhappy with Giftedness subscales. Analyses of academic and social skills yielded no significant differences. However, the teachers' mean score for the Unhappy with Giftedness subscale was significantly higher than that of the students (teacher = 10.7, student = 8.2; t = 5.9, p [less than] .001), indicating that the teachers viewed the gifted students as being less happy than the students viewed themselves.
DISCUSSION
Demographic differences between the gifted and nongifted students were noted on ethnicity and parental education. The higher percentage of white, non-Hispanic students and lower percentage of black and Hispanic students is a common attribute of gifted programs. This uneven ethnic distribution has been noted in much of the research on gifted students (Masten, 1985). The higher education level of the gifted students' parents was not surprising and may relate to more highly educated parents realizing the importance of superior education and having the resources and resourcefulness to ensure their children' s placement in these programs.
