Abstract
The purpose of this study was to examine the construct of amotivation. When amotivation is measured, negative items in a questionnaire cause attenuation of correlation, and as a result, it would give bias to the construct of amotivation. A questionnaire was made on the basis on Noels, Pelletier, Clement, and Vallerand (2000), and it included both positive items (P-type) and negative items of amotivation (N-type). By analyzing the data from the questionnaire using Confirmatory Factor Analysis to correct attenuation, we examined a systematic error caused by negative items. The result showed that (1) an artificial factor was identified when positive and negative items of amotivation were analyzed by Exploratory Factor Analysis; (2) the construct of amotivation was supported when 7-factor model of motivation was examined by conducting Confirmatory Factor Analysis to P-type questionnaire; and (3) P-type questionnaire was more appropriate than N-type questionnaire as a measure of amotivation, because bipolarity between amotivation and self-determined forms of motivation was clearly identified in P-type questionnaire.
