The nature and correlates of self-esteem trajectories in late life

Jenny Wagner, Denis Gerstorf, Christiane Hoppmann, Mary A. Luszcz

Research output: Contribution to journalArticlepeer-review

58 Scopus citations

Abstract

Is it possible to maintain a positive perspective on the self into very old age? Empirical research so far is rather inconclusive, with some studies reporting substantial declines in self-esteem late in life, whereas others report relative stability into old age. In this article, we examine long-term change trajectories in self-esteem in old age and very old age and link them to key correlates in the health, cognitive, self-regulatory, and social domains. To do so, we estimated growth curve models over chronological age and time-to-death using 18-year longitudinal data from the Australian Longitudinal Study of Ageing (N 1,215; age 65-103 years at first occasion; M 78.8 years, SD 5.9; women: 45% of sample). Results revealed that self-esteem was, on average, fairly stable with minor declines only emerging in advanced ages and at the very end of life. Examination of the vast between-person differences revealed that lower cognitive abilities and lower perceived control independently related to lower self-esteem. Also, lower cognitive abilities were associated with steeper age-related and mortality-related self-esteem decrements. In our discussion, we consider a variety of challenges that potentially shape self-esteem late in life and highlight the need for more mechanism-oriented research to better understand the pathways underlying stability and change in self-esteem.

Original languageEnglish (US)
Pages (from-to)139-153
Number of pages15
JournalJournal of Personality and Social Psychology
Volume105
Issue number1
DOIs
StatePublished - Jul 2013

All Science Journal Classification (ASJC) codes

  • Social Psychology
  • Sociology and Political Science

Fingerprint

Dive into the research topics of 'The nature and correlates of self-esteem trajectories in late life'. Together they form a unique fingerprint.

Cite this