Multi-group invariance of the conceptions of assessment scale among university faculty and students

Melanie Anne DiLoreto, University of Southern Mississippi


Conceptions are contextual. In the realm of education, conceptions of various constituent groups are often shaped over a period of a number of years during which time these groups have participated in educational endeavors. Specifically, conceptions of assessment are influenced by beliefs, actions, attitudes, understandings, and past experiences. These conceptions can impact both teaching and learning, and ultimately student achievement. Based on the past work of Gavin T. L. Brown (2004, 2006, 2008, 2010, & 2011) and Fletcher, Meyer, Anderson, Johnston, and Rees (2011) concerning conceptions of assessment held by educators and students in environments with a low-stakes assessment culture, this study re-examines the factor structure of the Conceptions of Assessment III (CoA-III) for faculty and students of higher education within a high-stakes assessment culture. Five models were initially considered based on past research by the above-mentioned researchers. Upon examining model fit of these five models, results indicated an acceptable model fit to the data collected from faculty and students within the United States. Furthermore, invariance testing elicited differences in how faculty ( N = 159) and students ( N = 404) of higher education conceptualize the purpose of assessment. Specifically, faculty members report that a primary purpose of assessment is for improvement of both teaching and learning. This group also suggested that assessment is useful for ensuring student accountability. However, results indicate that students associate assessment with accountability - both at the institutional and student levels. Furthermore, the data also suggest that a strong relationship between accountability and improvement exists. These results can be interpreted to mean that as accountability measures increase, there is a concomitant rise in the use of assessment for improvement purposes. Additional results of this study and implications of these findings for educational settings with high-stakes assessment cultures are discussed. iii