THE MODIFIED ATTITUDES TOWARD SCIENCE INVENTORY: DEVELOPING AN INSTRUMENT TO BE USED WITH FIFTH GRADE URBAN STUDENTS

Abstract

The purpose of this study was to develop an attitudinal instrument that could be used over the course of a funded local systemic change project to measure the changes in attitudes toward science of African-American fifth grade students. After examining many instruments, we modified an existing one so that it retained acceptable reliability for fifth grade students, both males and females, in an urban school district. The modified instrument was short enough so that young students who often have short attention spans would not become inattentive. We shortened the Attitudes Toward Science Inventory: version A (ATSI:a), a 48-item Likert-scale instrument to 25 items, and designated it the Modified Attitudes Toward Science Inventory (mATSI). We selected the ATSI:a because of its multidimensional nature and our belief that attitudes toward science cannot be measured on a single scale. Two phases were used for the modification: (a) after establishing that the ATSI:a was reliable for males and females and African-American middle grades students, we conducted a factor analysis to select an appropriate subset of items to form the shortened version, and (b) we administered this shortened version to a pilot group of African-American fifth grade students. Reliability was established using Cronbach's alpha. The new version, mATSI, cut administration time from 60 to 40 minutes. It also retained alpha coefficients on all five scales above the 0.50 minimum level set by McMillan and Schmacher (1993) and Nunnally (1967) for attitudinal instruments when sorted by ethnicity and gender for a sample of fifth grade students.