Extending the communicative ecology model of successful aging using talk about careers and retirement

ABSTRACT

Given the emphasis on work in American lives, this study extended the communicative ecology model of successful aging (CEMSA) by testing two new domains of communication about aging: career meaning and preparing for retirement (N = 340 older adults). Findings supported many of the proposed paths including negative environmental chatter to negative affect, uncertainty, and both domains of communication. Positive environmental chatter was associated with positive affect, efficacy, and both domains. In both models, uncertainty was positively associated with negative affect and inversely with positive affect. Efficacy was consistently related to successful aging. Finally, the effect of positive environmental chatter on successful aging was serially mediated by career meaning, negative affect, and efficacy. Results offer theoretical and practical implications.

KEYWORDS:

• Aging
• careers
• communicative ecology model of successful aging
• retirement

Notes

1 Although explicit positive messages have also been theorized (Gasiorek et al., 2016), we are not aware of a reliable, multi-item scale for this variable (see Gasiorek & Fowler, 2020) and therefore did not assess it in this study.

2 Given the focus of this study on career-related issues, education is a theoretically relevant consideration as, in general, higher levels of education are associated with higher wages, lower levels of unemployment, and other key outcomes (U.S. Bureau of Labor Statistics, 2020). Empirically, level of education has been identified as an important covariate in CEMSA scholarship (e.g., Fowler et al., 2015; Gasiorek & Fowler, 2020) and is treated as a covariate in all hypotheses (Becker et al., 2016).

3 As the CEMSA is an evolving framework, varying numbers of domains have been tested. For example, Fowler et al. (2015) referenced seven domains of communication whereas Gasiorek and Fowler (2020) identified six. We used the items associated with this more recent measure, which the authors referred to as “Revised Scales” (p. 118), as the starting point for our measures (described in Measures).

4 Organizational communication scholars, in particular, have been at the forefront of investigating career meaning/the meaning of work (including meaningful work) and how these are communicatively constituted (see, for example, Zorn & Townsley’s, 2008 forum in Management Communication Quarterly). However, scholarship from a communicative perspective that explores the intersections of meaning(s) of work and aging is limited. Hence, we draw from a range of disciplinary literature (e.g., psychology) to advance our argument. Also see Online Supplement Table S1.

5 A Kaiser Meyer-Olkin measure of sampling adequacy of .934 confirmed the sample was suitable for factor analysis, as did a significant Bartlett’s test of sphericity, χ2(1431) = 14,851.10, p < .001 (Kaiser, 1970). Items were assigned to factors if they (a) had primary loadings of at least .5, (b) had no cross-loadings within .2 of the primary loading, and (c) cohered thematically with other items loading on that factor. Eight factors with Eigenvalues exceeding one were extracted (range 1.366 to 16.570), explaining 69.60% of the variance. Six of these factors were nearly identical to those from Gasiorek and Fowler (2020) with a few exceptions (e.g., the item “I challenge ageism or age-based stereotypes that I encounter” from Expressing Optimism had a high cross-loading with Resisting Anti-Aging Media messages and so was dropped). In addition, two new career-related factors were identified. The first focused on the idea of one’s work/career being important and meaningful (Eigenvalue = 6.722; 10 items) and the second on preparing for retired life, which featured a primary emphasis on financial issues (Eigenvalue = 1.625; 6 items). See Online Appendix A for all retained items.

6 The SEs are larger in magnitude than the indirect effects in some cases, possibly as a result of the skewed distribution of the bootstrapped estimates and SEs.