Issues in the measurement of change in reading achievement over time

Year: 1994

Author: Lietz, Petra

Type of paper: Abstract refereed

In the literature, assumptions regarding the structures involved in the reading process vary. While previously the discussion focused on different skills involved in reading and their possible hierarchical ordering, the scholarly debate has shifted to distinguish between processes associated with different types of reading materials.

These assumptions, in turn, have an impact on the way in which student test scores are calculated. If, for example, skills are considered to be distinct, the calculation of a total score involving items that assess different skills may appear inappropriate. If, on the other hand, the assumption is that different skills exist but that they share a substantial common trait, the calculation of a total score may be appropriate. In this paper, the assumptions underlying the tests employed in two studies of reading achievement at the 14-year-old level undertaken by the International Association for the Evaluation of Educational Achievement (IEA), namely the 1970/71 Reading Comprehension Study and the 1990/91 Reading Literacy Study, are examined using confirmatory factor analysis (CAF) and item response theory (IRT).

In addition, an attempt is made to investigate the changes in student reading performance over the 20-year period for those eight countries that participated on both occasions (Belgium French, Finland, Hungary, Italy, Netherlands, New Zealand, Sweden, USA) using the items that were common to the two reading tests.