The Effects on Item Phrasing of the Self-report Personality Scale for the Elementary and Junior High School Students

• Main Authors: Chen-Su Hsu, 徐貞淑
• Other Authors: 李克明
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/77009221060794781773

碩士 === 臺中師範學院 === 諮商與教育心理研究所 === 93 === The main purposes of the study were to investigate the consistency and reaction time of item response of different item phrasing for the elementary and junior high school students. Experimental design was employed in the study. The four treatments of item phrasing were: 1) positive trait with affirmative phrasing, 2) positive trait with denying phrasing, 3) negative trait with affirmative phrasing and 4) negative trait with denying phrasing. All the scale items were presented by the experimental lab software and experiment generator – SuperLab pro 2.0 on the laptop in Windows. The subjects (202 students) were sampled from the elementary and junior high schools in Tainan City and Pingtung County. Statistical data were analyzed by t-test, Pearson product-moment correlation, mix design two-way ANOVA and stepwise multiple regression. The findings are as follows： 1.The consistency of item response was found to have significant differences among the four item phrasing treatments. 2.The reaction time of item response was found to have significant differences among the four item phrasing treatments. 3.The consistency and reaction time of item response of gender were found to have no significant differences. 4.The consistency and reaction time of item response of school grade were found to have significant differences. 5.The interaction effects of the four item phrasing treatments and gender were found to have no significant differences. 6.The interaction effects of the four item phrasing treatments and school grade were found to have no significant differences. 7.The length of item phrasing, positive trait with denying phrasing, negative trait with denying phrasing, and the score of self-concern sub-scale can significantly predict the reaction time.