Skip to main content

Table 1 Testing for balanced covariates—mean difference test

From: The effects of youth training on risk behavior: the role of non-cognitive skills

  Control group Treatment group Difference SE
Sociodemographics
Female 0.1 0.1 0.0 (0.0)
Age 23.2 23.1 0.1 (0.4)
Single 0.7 0.8 −0.1 (0.0)
White 0.2 0.2 −0.0 (0.0)
Unemployed 0.2 0.2 −0.0 (0.0)
Ever worked 1.0 0.9 0.0 (0.0)
N. of jobs 4.8 5.1 −0.4 (0.4)
Employed (last week) 0.6 0.5 0.1 (0.1)
Income (unconditional) 498.6 416.1 82.5 (49.4)
Years of education 11.71 11.65 0.06 (0.18)
Household size 3.5 3.9 −0.3 (0.2)
Household income 1554.2 1729.5 −175.3 (127.2)
N. children 0.5 0.6 −0.1 (0.1)
Household’s members literate 1.0 1.0 −0.0 (0.0)
N. rooms 4.7 4.8 −0.2 (0.1)
Rent 348.3 338.1 10.2 (32.5)
Has TV 1.0 1.0 −0.0 (0.0)
Has DVD 0.9 0.9 −0.0 (0.0)
Has fridge 1.0 1.0 −0.0 (0.0)
Has freezer 0.2 0.3 −0.1 (0.0)
Has computer 0.5 0.5 −0.0 (0.1)
Has car 0.2 0.2 −0.0 (0.0)
Has motorcycle 0.1 0.1 −0.0 (0.0)
Bolsa Carioca beneficiary 0.0 0.0 −0.0 (0.0)
Bolsa Familia beneficiary 0.1 0.1 −0.0 (0.0)
Socio-emotional skills
CPS: total −0.00 −0.21 0.21 (0.11)
Grit: total −0.00 −0.04 0.04 (0.10)
Risk behavior
Smoked ever? 0.26 0.21 0.05 (0.05)
Casual alcohol last week? 0.31 0.38 −0.07 (0.05)
High alcohol last week (>5 drinks)? 0.57 0.53 0.03 (0.09)
Consumed marijuana last week? 0.05 0.03 0.02 (0.02)
Ever consumed any substance? a 0.03 0.02 0.01 (0.02)
Fight last month? 0.03 0.02 0.00 (0.02)
Witness any crime (last year) 0.53 0.54 −0.01 (0.05)
Victim any crime (last year) 0.31 0.26 0.04 (0.05)
Summary index 0.08 0.05 0.03 (0.04)
Observations 178 150   
  1. The CPS and Grit Scales are presented in z-scores to ease interpretation
  2. aThe question asks whether the individual has ever consumed any substance such as cocaine, heroin, ecstasy, or sniffed glue; it excludes marijuana consumption. The summary index of risk behavior is an equally weighted average of the z-scores of its components (i.e., the eight measures of risk behavior). The z-scores are calculated by subtracting the control group mean and dividing by the control group standard deviation