print

Correlational findings

Study Alesina et al. (2004): study US 1981

Public:
Adults, general public, United States, 1981-1996
Sample:
Respondents:
N = 19895
Non Response:
not reported
Assessment:
Interview: face-to-face

Correlate

Authors's label
Inequality
Our Classification
Operationalization
Income inequality in US states in Gini coefficients (after-transfer, after-tax monetary income)

Observed Relation with Happiness

Happiness Measure Statistics Elaboration / Remarks O-HL-c-sq-v-3-aa = STATE inequality by INDIVIDUAL happiness O-HL-c-sq-v-3-aa OLRC = -2.9 p < .05 All individuals O-HL-c-sq-v-3-aa OLRC = -2.4 p < .05 Political left individuals O-HL-c-sq-v-3-aa OLRC = -3.1 ns Political right individuals O-HL-c-sq-v-3-aa OLRC = -2.8 ns Poor individuals O-HL-c-sq-v-3-aa OLRC = -2.9 p < .05 Rich individuals

OLRC's are controlled for personal characteristics:
- employment satus
- gender
- age
- educational level
- marital status
- number of children
- income
- race
O-HL-c-sq-v-3-aa OLRC = -2.1 p < .10 Additionaly controled for unemployment rate at state level, all individuals O-HL-c-sq-v-3-aa OLRC = -1.4 ns Political left individuals O-HL-c-sq-v-3-aa OLRC = -2.3 ns Political right individuals O-HL-c-sq-v-3-aa OLRC = -1.3 ns Poor individuals O-HL-c-sq-v-3-aa OLRC = -2.8 p < .05 Rich individuals

All OLRC's remain the same when additionally controlled for crime rate.

OLRC's cannot be interpreted as absolute effect sizes. Relative values denote that the negative effect is larger among rich and political right individuals. Comparisons of OLRC's are only possible if the same control variables are used in the regressions.