Abstract:
The purpose of the present study is to establish the validity, reliability, and partial norms of the Turkish Teacher Acceptance-Rejection/Control Questionnaire Child short form (Turkish Child TARQ/Control) that was developed by Rohner (2002) and revised by him (2004) to assess the perceived teacher acceptance-rejection and behavioral control. The instrument consists of four subscales, namely; Warmth/Affection, Hostility/Aggression, Indifference/Neglect, and Undifferentiated Rejection, and a separate Control scale. It has 29 items. The sample of the study was composed of 503, 4th and 5th grade students coming from six different primary schools and three different SES levels in Istanbul, and 253 of them were readministered the instrument for investigating the temporal stability of the instrument. Internal consistency of the instrument was determined by computing the corrected item-total correlation, which was found to be ranging between .31 and .64, item-subscale, subscale-total, subscale-subscale correlations and Cronbach Alpha values for the total scale, that is .90 (p<.01), for its subscales, and for the separate Control scale. Test-retest reliability coefficient was found to be .76 (p<.01) for the total scale.The concurrent validity of the instrument was computed by correlating the subjects̕ Child TARQ scores with their scores on the Perceived Teacher Behavior Inventory (PTBI) and The Teacher Support Subscale of the Perceived Social Support Scale-Revised (PSSS-R/TSss). The Pearson Product Moment Correlation technique yielded a correlation coefficient of .82 with the PTBI and -.79 with the PSSS-R/TSss (p<.01). Factor analysis that was carried out for determining construct validity of the instrument yielded four factors. When analyzed with two factors, rejection items clustered around F 1 and warmth/affection items clustered around F 2, as expected. The partial norms of the instrument were established on age, gender, school SES, self-reported semester point average, teacher gender, mother and father education, mother and father employment status, and number of children in the family by using the means and standard deviations. One-way Analysis of Variance and Scheffe procedure results indicated significant differences in perception of teacher rejection in relation to gender (F=15.81; p<.001), teacher gender (F=6.04;p<.05), and SPA (F= 4.51; p<.001), such that males perceived higher rejection, male teachers were perceived to be more rejecting, and low SPA students perceived higher rejection. Overall, the results of the present study indicated that the Turkish Child TARQ is a highly reliable and valid instrument to be used with children for assessing their perceptions in terms of teacher acceptance-rejection while the Control scale assessing perception of behavioral control by teachers needs some revision.
