Cohort Differences in Verbal Ability: Testing the Word Obsolescence Hypothesis

• In Event: 207 - Section on Sociology of Education Roundtable Session (one-hour).
  In Refereed Roundtable: Table 22. Media and Literacy

Abstract

Using data from the General Social Survey’s test of vocabulary knowledge (WORDSUM), included in twenty surveys between 1974 and 2012, this paper examines one persistent hypothesis introduced to account for intercohort trends in vocabulary knowledge—the word obsolescence hypothesis. We investigate this hypothesis by, first, employing a 3-parameter IRT model to examine the psychometric properties of WORDSUM and its components over time and between cohorts. This analysis asks whether the formal properties of the test have changed in such a way as to make the test items more difficult over time in the population as a whole. Second, we introduce a new measure of “exposure to word frequency” based on the Google Ngrams facility to measure intercohort differences in exposure to word frequency in published textual material. Using a number of measurement strategies, we define a “window of exposure” for each birth cohort—the frequencies of the WORDSUM words during the critical years of educational exposure—and examine the effects of these patterns on word knowledge. Our results establish a strong basis for the conclusion that intercohort differences in WORDSUM, across all levels of conceptual difficulty, can be explained by variations over time in cohort-specific exposures to the test items, thereby pointing to a “cohort experience” interpretation of intercohort trends in WORDSUM, and casting doubt on other interpretations (such as cognitive aging) that have been advanced for intercohort declines in vocabulary knowledge.

Authors

• Duane F. Alwin, Pennsylvania State University

• Shawn Dorius, Iowa State University

• Julianna Pacheco, University of Iowa