. 2015 Nov 5;9:287. doi: 10.3389/fnbeh.2015.00287

Table 2.

Non-egalitarian choice (option B) as a function of CRT and Raven (Study 1).

	Decision 1	Decision 2	Decision 3	Decision 4
Dep var:	β ≤ 0 (vs ≥ 0)	β ≤ 0.5 (vs ≥ 0.5)	α ≤ 0 (vs ≥ 0)	α ≤ 0.5 (vs ≥ 0.5)
	(1a)	(2a)	(3a)	(4a)
CRT	−0.136^**	−0.054	0.249^***	0.236^***
	(0.065)	(0.054)	(0.055)	(0.059)
	[−0.029^**]	[−0.016]	[0.083^***]	[0.073^***]
Female	−0.573^*	−0.329	−0.205	−0.219
	(0.306)	(0.236)	(0.224)	(0.233)
	[−0.120^**]	[−0.098]	[−0.069]	[−0.068]
Cons	−0.465^*	1.077^***	−0.438^*	−0.070
	(0.278)	(0.252)	(0.230)	(0.233)
	[0.321]	[0.859]	[0.331]	[0.472]
ll	−57.017	−79.185	−88.038	−81.547
Wald χ²	5.27^*	2.43	23.72^***	19.25^***
Pseudo R²	0.061	0.014	0.137	0.126
	(1b)	(2b)	(3b)	(4b)
Raven	−0.061	0.034	0.093^*	0.121^**
	(0.052)	(0.055)	(0.050)	(0.050)
	[−0.013]	[0.010]	[0.035^*]	[0.040^**]
Female	−0.438	−0.242	−0.468^**	−0.446^**
	(0.273)	(0.228)	(0.211)	(0.219)
	[−0.095]	[−0.072]	[−0.175^*]	[−0.149^**]
Cons	−0.024	0.373	−0.945	−1.051
	(0.852)	(0.811)	(0.745)	(0.742)
	[0.490]	[0.645]	[0.172]	[0.147]
ll	−58.909	−79.473	−97.676	−88.273
Wald χ²	3.25	1.45	8.33^**	9.50^***
Pseudo R²	0.030	0.010	0.043	0.054
	(1c)	(2c)	(3c)	(4c)
CRT	−0.134^*	−0.090	0.253^***	0.221^***
	(0.075)	(0.057)	(0.061)	(0.064)
	[−0.028^*]	[−0.026]	[0.084^***]	[0.068^***]
Raven	−0.007	0.074	−0.009	0.036
	(0.070)	(0.059)	(0.055)	(0.055)
	[−0.001]	[0.022]	[−0.003]	[0.011]
Female	−0.571^*	−0.361	−0.202	−0.233
	(0.309)	(0.234)	(0.226)	(0.236)
	[−0.120^*]	[−0.106]	[−0.067]	[−0.072]
Cons	−0.376	0.131	−0.319	−0.540
	(0.936)	(0.817)	(0.734)	(0.749)
	[0.353]	[0.552]	[0.375]	[0.304]
ll	−57.012	−78.420	−88.026	−81.365
Wald χ²	5.41	4.57	23.72^***	19.30^***
Pseudo R²	0.061	0.023	0.137	0.128
N	150	150	150	150

Probit estimates. The α and β parameters associated with the dependent variable are displayed on top of each column. In “a” regressions, the main explanatory variable is CRT score. In “b” regressions, the main explanatory variable is Raven score. In “c” regressions, both CRT and Raven scores are included as explanatory variables. Robust standard errors clustered on individuals are shown in parentheses and average marginal effects of the explanatory variables are shown in square brackets (for the constant, this value represents the probability obtained from normal transformation of the Probit coefficient).

^{*, **, ***}

denote p-values lower than 0.10, 0.05, and 0.01, respectively.