The Development Of Gender Role Behavior Inventory For The College Students

Reference to gender role theory of two-dimensional model, Bem Sex Role Inventory (BSRI) and Qian Mingyi's Sex Role Inventory for College Students(CSRI) to get 155 basic adjectives, the use of open questionnaire to the subjects selected the adjectives which most describes negative characteristics in male and so on, and then write the behavior that best describe this adjective. After the formation of behavioral items, please be assessed 10 psychology graduate students, take most of a sentence adjective extent, the results obtained from 50 gender-role behavior items described in the formation of the initial measurement of the sentence form.Guilin for 545 scores on the four universities students carried out a first test object. The parallel analysis identified four factors exists. Exploratory factor analysis (EFA) initially set the inventory structure consists of four subscales are male negative, male positive, female negative and female positive, to explain the variance of 39.585%.Formal test with CSRI, the four universities 900 questionnaires,774 questionnaires were valid. Using confirmatory factor analysis (CFA) to verify the validity of the inventory and corrections. Then through the classical test theory (CTT) and Item Response Theory (IRT) of each component menu item for further screening. The results showed that:1. confirmatory factor analysis showed that 4-factor model revisedχ2/df, RMSEA, RMR, NFI, NNFI, RFI, IFI, CFI, respectively 2.930,0.051,0.130,0.930,0.940,0.920,0.950,0.950; four factors in the estimated value of the correlation coefficient between 0.110-0.680, indicating that the scale structure better; correlation coefficient estimates show that female negative scale and male positive scale, female positive scale and male negative scale had moderate negative correlation, while male positive scale and male negative scale, female positive scale and female negative scale are of the low negative correlation, supporting the theoretical model of Qian Mingyi, also consistent with the hypothesis that male positive characteristics and female negative temperament are the same dimension, but the poles are located in the dimension. Female positive temperament and male negative temperament are of the same dimension, but also the dimension of the poles are located;2. the reliability of gender role behavior inventory is good, test-retest reliability is between the 0.925-0.979, Cronbach's a coefficients are between 0.676-0.871;3. the criterion-related of the gender role behavior inventory for college students and CSRI is between 0.410-0.643, suggesting that good criterion validity;4. the four subscales of the CTT and IRT parameter estimates show that item difficnlty all meet the qualification. Formal scale of 30 items were good discrimination;5. college students were in the genders role behavior inventory on each factor divided equally between the 2-5 and the standard deviation between 0.886-1.056, that tends to choose more positive way;6. the subjects in the masculine, feminine, androgynous, undifferentiated four kinds of sex roles on the size distribution of 15%,17%,34%,34%, is androgynous, and a higher proportion of undifferentiated;7. boys and girls between the gender role behavior statistically significant (χ2=58.171, p= 0.000;χ2=46.733, p=0.000), boys on the negative and positive factors in the male scores are significantly higher than girls, while girls on the negative factors in female score higher than boys;8. a negative factor in the male, the liberal arts subjects scored the highest, lowest engineering; positive factor in the male, and engineering subjects scored the highest, lowest Arts; negative factor in female, and liberal arts subjects scored the highest, lowest medicine; positive factor in the female, the male and female students showed no statistically significant scores;9. the only-child subjects scored on the negative factors in male is significantly lower than that of the non-only-child, and both of them in negative male, female, and positive female scored difference was not statistically significant; the subjects from the country test scores in positive male were significantly higher than those from the city, and both of them in negative male, female, and positive female scored difference was not statistically significant.Comparative discussion of the preparation of the students in the study of gender role behavior inventory and the CSRI, CTT and IRT, and from the point of view of the development of gender roles have changed with the times, this study compared college students to discuss gender role behavior and status and Qian Mingyi & Bern, who studies the similarities and differences between the results and their possible causes.