Econometrics – GradSchoolPapers.com

. In order to identify what are some of the household characteristics that account for differences in human and social capital in a community that faces high levels of violence, we run the
following regressions for children 7-12 of a country.
Human Capitali = β0 + β1 AGECi + β2 AGEPi + β3 SEXCi + β2 SEXPi +β3 RACEi +
+ β4 EDUCPi + β5 OCCUPi + β6 NUMJOBSi + β7 NUMJOBPi + β8 FAMDINi +
+ β9 FAMSTRUCi + β10 NUMCHILDi + β11 SEXHEADi + εt (1)
Social Capitali = β0 + β1 AGECi + β2 AGEPi + β3 SEXCi + β2 SEXPi +β3 RACEi +
+ β4 EDUCPi + β5 OCCUPi + β6 NUMJOBSi + β7 NUMJOBPi + β8 FAMDINi +
+ β9 FAMSTRUCi + β10 NUMCHILDi + β11 SEXHEADi + εt (2)
where Human and Social Capital are the child reported action that reflects the existence or absence of human and social capital. AGEC and AGEP is the age of the child and of the parent
respectively. SEXC and SEXP is the gender of the child and of the parent respectively (man=0 or woman=1), and RACE captures the ethnic group of child, typically the same than that of the parents
(Ladino=0 or indigenous=1). EDUCP is the educational level achieved by the parents; the higher the ordinal number, the more years of education and experience the parent has. OCCUP is the
occupation held by the head of the household. It is defined as an ordinal variable, where the higher the number the lower the skills required. NUMJOBS and NUMBJOBP are the number of jobs the
child or the parent hold, FAMDIN is the frequency with which family dinners take place in the household, FAMSTRUC is a dummy that captures married couples, including widows and unions, versus non-
married couples (divorced and single parents) (not married=0 or Married=1.) NUMCHILD is the number of children living in the household while SEXHEAD is the sex of the head of household.
Tables 1 and 2 presents the estimation outcomes (please note that some of the numbers have been made up, so there is a repetition of them):
Table 1, Human Capital
Saw or suffered bullying Sexually active Use drugs & alcohol Has problems with the Law
Is involved with gangs Protected siblings against violence Quality of life
Intercept -16.00*
(5.13) -18.80*
(13.75) -15.04**
(7.53) -1.55
(3.25) -152.16
(97.80) -4.41
(7.96) 0.46***
(0.16)
FAMSTRUC x
EDUCP -0.07*
(0.03) -0.14***
(0.03) -0.06* (0.001) -0.06*
(0.03) -2.18***
(0.91) 2.1
(1.98) 0.12***
(0.01)
FAMSTRUC -0.17***
(0.05) -0.09**
(0.06) -0.08**
(0.04) 0.02
(0.05) -0.11***
(0.06) 0.09*
(0.02) 0.09*
(0.02)
EDUCP 0.01
(0.01) 0.09*
(0.01) 0.12*
(0.07) 0.04
(0.05) 0.11***
(0.05) 0.08
(0.06) 0.08
(0.02)
RACE 3.2
(3.00) 2.11*
(1.21) 2.1
(1.98) 2.1
(1.98) 2.1
(1.98) 2.1
(1.98) 2.1
(1.98)
SEXC 1.39*
(0.51) -3.03**
(1.52) 2.1
(1.98) 1.37*
(0.67) 2.1
(1.98) -2.62*
(1.56)
AGEC 2.1
(1.98) 2.1
(1.98) 3.45**
(1.88) 2.1
(1.98) 2.1
(1.98) -2.26
(1.59) 0.16***
(0.05)
FAMDIN 3.2
(2.9) -2.45*
(0.97) -2.65*
(0.88) 2.1
(1.98) -37.78*
(22.24) -1.80*
(0.83) 0.17*
(0.07)
EDUCP 2.5
(2.98) -1.6
(1.95) 2.1
(1.98) 2.1
(1.98) 2.1
(1.98) 2.1
(1.98) 0.01***
(0.004)
AGEP 2.0
(1.96) -0.59***
(0.34) 2.1
(1.98) -0.58*
(0.14) 2.1
(1.98) -0.99*
(0.49) 2.1
(1.98)
NUMCHILD 1.99
(1.98) 2.2
(1.98) 2.1
(1.98) 0.88*
(0.13) 2.1
(1.98) 2.1
(1.98) 2.1
(1.98)
SEXHEAD -0.86*
(0.32) 2.1
(1.99) -1.14***
(0.70) 2.1
(1.98) 2.1
(1.98) -1.51*
(0.67) 2.1
(1.98)
OCCUP 1.01*
(0.25) 1.7
(1.68) 2.1
(1.98) 2.1
(1.98) 2.1
(1.98) 2.1
(1.98) 0.01***
(.001)
R2 72*** 82* 83* 88** 67*** 84* 81**
Notes: OLS estimates, ***, **, * indicate 1%, 5%, and 10% level of significance respectively.
Table 2, Social Capital
Family Violence Verbal Abuse Sexual Abuse Community violence Civic
Responsibility Social Responsibility
Intercept -10.03
(7.01) -3.36
(5.23) -1.54
(5.02) -1.60
(7.06) -0.22
(0.34) -0.38
(0.26)
FAMSTRUC x
EDUCP -2.41**
(1.25) 3.6
(3.5) -0.14***
(0.03) 3.6
(3.5) 0.11***
(0.01) 0.03**
(0.01)
FAMSTRUC -0.17***
(0.05) -0.09**
(0.06) -0.08**
(0.04) 0.02
(0.05) 0.11***
(0.06) 0.09*
(0.02)
EDUCP 0.01
(0.01) 0.09*
(0.01) 0.12*
(0.07) 0.04
(0.05) 0.11***
(0.05) 0.08
(0.06)
RACE 2.21**
(1.22) 3.6
(3.5) 2.10*
(1.21) -3.92***
(2.44) 3.6
(3.5) 0.11***
(0.06)
SEXC -2.79**
(1.32) -1.84*
(1.04) -3.03**
(1.51) 3.6
(3.5) 0.09***
(0.06) 3.6
(3.5)
AGEC 3.6
(3.5) 3.6
(3.5) 3.6
(3.5) 3.6
(3.5) 0.23*
(0.08) 0.20***
(0.06)
FAMDIN 3.6
(3.5) 3.6
(3.5) 3.6
(3.5) 1.88***
(1.64) 3.6
(3.5) 3.6
(3.5)
EDUCP 3.6
(3.5) 3.6
(3.5) 3.6
(3.5) 3.6
(3.5) 3.6
(3.5) 0.001**
(0.0001)
AGEP -3.6*
(0.35) -0.6
(0.55) -0.60*
(0.34) 3.6
(3.5) 0.17***
(0.11) 3.6*
(0.5)
NUMCHILD 3.6
(3.5) 3.6
(3.5) 3.6
(3.5) -2.25*
(0.9) 3.6*
(0.5) 3.6*
(0.5)
SEXHEAD -1.09**
(0.54) -0.98**
(0.38) -0.94***
(0.42) -2.96*
(1.25) 3.6
(3.5) 3.6
(3.5)
OCCUP 3.6
(3.5) 3.6
(3.5) 3.6
(3.5) 3.6
(3.5) -0.02**
(0.01) 3.6
(3.5)
LR 75** 60*** 74*** 79 77* 74*
a. analyze the findings. Is there support in the literature for the choice of variables
in the regression?
b. In order to examine whether family structure, specifically marriage, is a multiplier factor in the effectiveness of interventions for development, a follow up regression was estimated. It
compared the impact of the educational intervention alone vis a vis the impact of the intervention when the beneficiaries belonged to married households. To capture the intervention’s impact
alone, it used fixed effects and controlled for all other household characteristics but family structure, and it use two dummies, one dummy for the intervention (if participated in the educational
intervention =1, 0 otherwise) and one dummy to capture married households (married and unions = 1, 0 otherwise.) Table 3 presents the estimations. Please analyze the findings.
Based on all the findings, if you were to recommend some strategy to improve the educational intervention efforts, what would you recommend?