Date of Award

Fall 2004

Project Type


Program or Major


Degree Name

Doctor of Philosophy

First Advisor

John Limber


Recent research in social cognition suggests one's self-efficacy beliefs regarding one's cognitive abilities can influence the effort expended on cognitive tasks (Bandura, 1989; 1997; Cavanaugh & Greene, 1990; Dunlosky & Hertzog, 1998), which may affect performance. This project was conducted to examine the relationship between age, self-efficacy beliefs, text difficulty, resource allocation to text comprehension processes, and memory for text. 82 younger adults and 74 older adults completed the Metamemory in Adulthood Questionnaire (MIA; Dixon, Hultsch & Hertzog, 1988), Reading Self-Efficacy Questionnaire (RSEQ), and Media Consumption Habits Questionnaire. Using the on-line word-by-word moving window method, participants read 24 two-sentence passages for immediate recall after reading either comparatively easier or more difficult texts. Younger adults reported higher Memory Self-Efficacy (MSE) and higher Reading Self-Efficacy Strength (RSE) than older adults; there were no age differences in RSE Level. Groups were split into high reading self-efficacy (HRSE) and low reading self-efficacy (LRSE) based on RSEQ scores. Analyses of reading times indicated that HRSE individuals allocated more time to processing target texts after reading difficult texts than LRSE individuals, suggesting that SE may influence effort and persistence following difficulty. HRSE individuals recalled more of the text than LRSE individuals overall. A marginal interaction of Self-Efficacy and Age was found; older adults with HRSE recalled more from the text than older adults with LRSE, whereas there were no differences in performance among younger adults. RSE was also more strongly related to recall performance among the old than among the young. HRSE individuals recalled more from target texts following difficult texts than those with LRSE. While HRSE individuals overpredicted recall performance, LRSE individuals were relatively accurate. Regression analyses indicate that working memory span, verbal ability, age, and reading self-efficacy make independent contributions to recall performance. The data also suggest that reading self-efficacy, as measured by the domain-specific RSEQ, may be a better predictor of memory for text than memory self-efficacy, as measured by the domain-general MIA. Collectively, the data support Bandura's (1997) self-efficacy theory in that self-efficacy beliefs influence both effort to reading and performance. Limitations and avenues for future research are discussed.