print

Correlational findings

Study Selim (2008): study TR 1990

Public:
18+ aged, general public, Turkey, 1990-2001
Sample:
Respondents:
N = 6338
Non Response:
not reported
Assessment:
Interview: face-to-face

Correlate

Authors's label
Health status
Our Classification
Operationalization
Self report on single question:
1 very bad (reference)
2 poor
3 fair
4 good
5 very good

Observed Relation with Happiness

Happiness Measure Statistics Elaboration / Remarks O-SLW-c-sq-n-10-a OLRC = +1.1 p < .01 poor (compared to very bad) O-SLW-c-sq-n-10-a OLRC = +1.6 p < .01 fair (compared to very bad) O-SLW-c-sq-n-10-a OLRC = +2.2 p < .01 good (compared to very bad) O-SLW-c-sq-n-10-a OLRC = +2.8 p < .01 very good (compared to very bad) O-HL-u-sq-v-4-a OLRC = -.03 ns poor (compared to very bad) O-HL-u-sq-v-4-a OLRC = +.55 ns fair (compared to very bad) O-HL-u-sq-v-4-a OLRC = +1.0 p < .01 goog (compared to very bad) O-HL-u-sq-v-4-a OLRC = +1.7 p < .01 very good (compared to very bad)

OLRC's are controled for:
- year dummies
- gender
- age
- marital status
- number of childeren
- education level
- type of employment
- income scales
- importance of life domains
- political orientations and trust

OLRC's cannot be interpreted as absolute effect sizes. Coefficients denote relative differences, e.g. greater difference in happiness to a very good health. Comparison is only possible across associations with the same happiness measure.