Correlational findings

Study Miller et al. (1991): study US 1956

White married mothers with children under age 13, USA 1956
N = 356
Non Response:
Interview: face-to-face
and questionnaire.


Authors's label
Social integration
Our Classification
Number of roles beyond that of wife and mother:
- paid worker
- club member
- religious member
- friend
- neighbor
- relative
Rated: yes/no

Observed Relation with Happiness

Happiness Measure Statistics Elaboration / Remarks M-CO---mq-v-3-a Beta = +.20 p < .001 Betas controlled for:
- age
- education
- husband's occupation (professional or not)
- life-cycle stage (chlid < 6 years or not)
- number of children
M-CO---mq-v-3-a Beta = +.24 p < .001 Additional control for employment
(paid work omitted from list of roles)
M-CO---mq-v-3-a Beta = ns Unaffected by interactions between
- roles/employment
- roles/ husband's occupation
- employment/husbands occupation
M-CO---mq-v-3-a Beta = ns Squared roles: To test for curvi-linearity life satisfaction was regressed on the square of the number of roles, controlling for individual and structural variables and number of roles M-CO---mq-v-3-a Beta = ns To assess whether the number of roles or one role in particular mattered most, each role was entered separately into a regression equation along with with the control variables and a summary measure of other roles. No single role proved statistically significant, while the contribution of the number of roles remained significant