Skip to main content
. 2019 Feb 12;10:178. doi: 10.3389/fpsyg.2019.00178

Table 3.

Regression models: giving in dictator stage II.

DV: Giving to Recipient OLS LPM gamma-GLM
Two-part model
Intercept 40.806*** 0.738*** 4.008***
(3.742) (0.043) (0.066)
WEAK DEFAULT 2.379 −0.042 0.085
(5.051) (0.057) (0.091)
STRONG DEFAULT −4.361 −0.199*** 0.215**
(5.413) (0.062) (0.104)
CONTROL INCOME 2.709 −0.069 0.168**
(5.150) (0.058) (0.087)
CONTROL PASSIVE GIVING −0.051 −0.065 0.096
(5.091) (0.059) (0.090)
Income before DG II −13.600*** −0.218*** −0.052
(4.972) (0.059) (0.102)
WEAK DEFAULT × Income before DG II −6.239 −0.036 −0.067
(6.376) (0.075) (0.126)
STRONG DEFAULT × Income before DG II 4.523 0.074 0.070
(5.819) (0.069) (0.120)
CONTROL INCOME × Income before DG II 13.736** 0.240*** 0.023
(6.119) (0.071) (0.114)
CONTROL PASSIVE GIVING × Income before DG II 13.342** 0.196*** 0.074
(6.079) (0.071) (0.118)
Observations 678 678 443
R2 0.059 0.085
F(9, 668) / F(9, 668)/ χ2(9) 4.706 6.899 7.299

+p < 0.10;

* p < 0.05;

**

p < 0.01;

***

p < 0.0001.

Robust standard errors are in parentheses. The dependent variable is giving to the recipient in Dictator Stage II. NO DEFAULT is the omitted treatment captured by the intercepts. “Income before DG II” represents the (mean-centered) monetary income a participant had earned in the experiment when arriving at Dictator Stage II (partly endogenously determined in NO DEFAULT, WEAK DEFAULT, and STRONG DEFAULT, exogenously assigned in control treatments). Gamma-GLM estimates are on a log-scale. The two-part model fits the data better than the OLS specification subsuming the complete data. The combined log-likelihood of the two-part model is –2628.207 compared to –3454.219 of the OLS. The table was compiled using the “stargazer” tool by Hlavac (2018).