The Elementary Reading Attitude Survey: Factor Invariance Across Gender and Race

The Elementary Reading Attitude Survey (ERAS) is purported to measure attitude toward recreational and academic reading. The present study examined the validity of this two-factor model across gender and race. Since measurement results are often group specific, factor invariance cannot be assumed, but should be tested empirically. No statistically significant difference in the factor structures for male and female respondents was found, indicating factor invariance across gender. However, there was a statistically significant difference (p < .05) in the factor structures of the African American and European American respondents. Follow-up analyses indicated that the sources of the invariance were three of the recreational reading attitude items (p < .05) and the recreational reading attitude factor variance (p < .005), with the variance being greater for African American than for European American students.