AbstractThe present study set out to examine the epistemological beliefs of 1, 224 African-American and Latino college students. The initial purpose of the study was to first validate an existing instrument’s usefulness in measuring the epistemological beliefs of African-American and Latino college students.The Jehng Epistemological Beliefs Instrument (JEQ) (Jehng, Johnson & Anderson, 1993) was statistically analyzed to determine whether the original hypothesized five-factor structure would remain the same for these two populations or whether different factor structures would emerge thereby reflecting alternative belief patterns.A concern of this study was to investigate the construct validity of the instrument prior to its use in further analyses of the study. Assuming successful evidence of validity of the JehngEpistemological Beliefs Instrument (JEQ),a second aim of the study was to examine three variables: education level, college major, and gender, said, in the extant literature, to influence epistemological beliefs.These three areas have been established in the epistemological beliefs literature for other ethnic groups, but remain unknown for African-American and Latino college students.In the present study, participants’ epistemological beliefs were to be compared across educational levels, college majors, and gender through a multivariate analysis of variance (MANOVA), and the computation of factor scores to facilitate comparison. However, the initial exploratory factor analysis revealed that Jehng’s 5-factor model of epistemological beliefs was not replicated for the African-American andLatino college students in the presentstudy; instead a 3-factor model emerged for both groups respectively.In other words, the factor structures which emerged for the African-American and Latino college students were not the same as the factor structure of the Jehng Epistemological Beliefs Instrument.Thus, the validity of the Jehng Epistemological Beliefs Instrument for measuring the epistemological beliefs of African-American and Latino college students in this study was called into question.Moreover, based on the results of the factor analyses, any subsequent statistical analysis, previously intended, e.g., MANOVA, factor scores, etc., were not supported. Hence, the goals and direction of this study changed in a significant manner.To support the exploratory factor analyses, confirmatory factor analyses were conducted on data for both African-American and Latino college students. Three-factor models were accepted as viable for both groups, however, this acceptance should be regarded as marginal. The statistical results, while acceptable, still fall far below what is considered adequate according to customary standards. The results of this study coincide with previous research which has concluded that self-report epistemological beliefs instruments should be used with extreme caution, and interpretations based on data gain from these instruments should be corroborated with other reliable measures (e.g., rigorous qualitative methods).
【 预 览 】
附件列表
Files
Size
Format
View
African-American and Latino students' beliefs about knowledge and learning