Investigating the Validity of Student’s Portfolio in Classroom Assessment at the First Secondary Class in the Kingdom of Bahrain

This research aimed to investigate the validity of the secondary student's portfolio by utilizing Messick's six-facet unified model of validity (content, structure, generalizability, external, substantive, consequences), as well as exploring teachers’ practices in using student’s portfolio for assessing students' achievement. The research was applied in two secondary schools in Bahrain, one for males and one for females. A random sample of 30 teachers from 130 teachers participated in the study. Two instruments were used: a 20-item questionnaire pertaining to teachers’ classroom assessment practices; and a 22-item form about the content of students' portfolio, with five dimensions. The results revealed that most teachers' assessment practices of the portfolio focused on students' activities and projects. Also, assigning grades to the students was another priority of teachers. Evidence of validity of students' portfolio in assessment according to Messick’s model was found. Inter-correlations ranged between 0.63 to 0.85; and reliability coefficients ranged from 0.84 to 0.92 among the 5 dimensions of the student portfolio. Regarding the generalizability validity, 47% of the variance of students’ scores in the portfolio is attributed to raters’ judgments about the relevance of portfolio components. Moreover, the results indicated evidence of the generalizability validity of the portfolio as depicted in the relationship between students' portfolio scores and students' GPA (r = 0.75) and between students' portfolio and students' scores in the national exam (r = 0.68). The researcher provided suggestions for further research