Correlational finding on Happiness and Compulsory retirement (vs free choice)
Subject code: R14ae02

StudyThompson et al. (1960): study US 1952
TitleThe Effect of Retirement on Personal Adjustment: A Panel Analysis.
SourceJournal of Gerontology, 1960, Vol. 15, 165 - 169
Public60+ aged, employed at start, followed 2 years, USA, 1952-54
SampleNon-probability purposive sample
Respondents N =1559

Author's labelCompulsory retirement.
Page in Source 168
Our classificationCompulsory retirement (vs free choice), code R14ae02
Retirement between T1 and T2
0 Willing
1 Reluctant
T1-T2 difference 1,5 years

Observed Relation with Happiness
T1-T2 CHANGE in happiness   happy    unhappy
                            to       to
                            unhappy  happy
Gainfull employed thoughout 34%(550) 22%(532)
Retired                     44%(238) 26%(239)
- willing voluntary         31%(75)  29%(65)
- willing administrative    39%(74)  42%(84)
- reluctant voluntary       64%(22)  26%(23)
- reluctant administrative  57%(67)  19%(67)
OR -1,04 
Unaffected by  T1 orientation to retirement

Appendix 1: Happiness measures used
CodeFull Text
M-AO---mq-v*-10-aSelfreport on 3 questions:

A All in all, how much happiness would you say you find in life today......?
3 a great deal
2 some but not very much
1 almost none

B In general, how would you say you feel most of the time, in good spirits or in low spirits.....?
1 I am usually in low spirits
2 I am sometimes in good spirits, sometimes in low spirits
3 I am mostly in good spirits

C On the whole, how satisfied are you with your way of life today.....?
4 very satisfied
3 fairly satisfied
2 not very satisfied
1 not satisfied at all

Items selected by Guttman scaling.

Appendix 2: Statistics used
Type: descriptive statistic only.
Measurement level: Correlate level: dichotomous, but nominal or ordinal theoretically possible as well. Happiness level: dichotomous
Range: [-100; +100]

Meaning: the difference of the percentages happy people at two correlate levels.
Type: descriptive statistic only.
Measurement level: Correlate: dichotomous, Happiness level: dichotomous
Range: nonnegative unlimited

OR = 1 no association at all;
OR = 0 or infinite at least one level of the correlate allows an error-free prediction of the happiness.
Ruut Veenhoven, World Database of Happiness, Collection of Correlational Findings, Erasmus University Rotterdam.