A preliminary form of a scale on mother attitudes toward highly competitive school entrance exams for fifth grade students

Abstract

The purpose of the present study is to develop a scale that would be used in measuring some parent variables which are assumed to have relationships with attitudes toward highly competitive entrance exams for high schools. Based on earlier research findings and informal observations of the researcher, four parent variables which seem to contribute to stressful exams experienced by children are suggested. These variables are conceptualized as (1) democratic/authoritarian child rearing practices, (2) identification with children, (3) conformity, and (4) social mobility. To measure the four suggested variables, an instrument composed of four subscales for a mother form and three subscales for a child form was constructed. A questionnaire on sociode: graphic family characteristics was also inclu'The newly constructed instrument and the questionnaire were administered to 75 mothers and their children from the fifth grades of two upper-middle/high SES elementary schools. The internal consistency and validity of the scale forms were conducted in four stages. In the first stage, item-remainder correlations and alpha coefficients for each mother and child subscale were computed. These initial coefficients were generally at an unacceptable level. In the second stage of the study. the most reliable items obtained from the first analysis were selected and reanalyzed. Using the selected pool of items produced more satisfactory results, though the internal consistency of most subscales Here still below the expected level. In the third stage, short farms for the mother and the child scales, composed of the best subscale items, were derived and analyzed. The total 30-item mother form yielded an acceptable (a=.81 and median item-remainder r=.34) outcome. However, a similar analysis for the child form did not give satisfactory results, therev fore, this scale was dropped. In the final stage, the 30-item mother form was subjected to factor analysis for validation. A three-factor varimax rotation showed clear evidence for the content validity of one of the subscales (namely, the identification with children subscale). This is the first factor, which accounted for 17 percent of the total variance and converged on the original ÖZDEŞ subscale. Factors two and three, on the other hand, accounted for 9 and 7 percent, respectively, but did not converge on a particular subscale. They included items from the DEMO (democratic/authoritarian child rearing practices), UYMA (conformity), and SOSHAR (social mobility) subscales showing no particular pattern, and therefore, not possible to label conceptually. There are two major limitations of this study: (1) related to the sample size and characteristics and (2) to the initial number of items. Further experimentation with additional items and cr08 validation on larger and varying samples may mitigate these limitations. The three-factor short mother form of the scale and the mother ÖZDEŞ subscale are recommended for these trials before they can be used for research purposes.