Versión de HTML Básico
Analíti a
k
9
Revista de Análisis Estadístico
Journal of Statistical Analysis
Analítika, Revista de análisis estadístico, (2015), Vol. 9
Impact of Fertility on Female Labor Supply
11
3 Data, Summary Statistics and Variables
This research uses data on labor supply, fertility and characteristics of household members
from the Ecuadorian Population Census of 2010 conducted by the National Institute of
Statistics and Census (Instituto Nacional de Estad´ısticas y Censos INEC).
The main motivation of this paper can be seen in Figure 1,where fertility (measured as
percentage of women with more than two children) and female labor supply move in opposite
directions. Within the group of women aged from 21 to 35 years old and with two or more
children, labor supply has increased by 13 percentage points (i.e. variation of 50% in paid
work) while the percentage of women in the group with more than two children decreased
by 14 percentage points (variation of 21%), all from 1990 to 2010.
Table 1 presents some labor force participation rates and probabilities of additional chil-
dren for women of different ages and different marital status for 1990, 2001 and 2010 Cen-
suses.
For the sex-mix instrument, the sample is limited to women between 21 and 35 years old
with at least two children and whose oldest child was, at most, 18 years old at the moment
of the census
3
. Following Angrist and Evans (1998), in order to match women with their
own children, the sample used is conformed by females who are “heads” or “spouses” in each
household, then the reported number of children alive is checked to coincide with the number
of children in the household matched to the women, restraining the sample to women for
whom both numbers were the same.
This is performed for two samples: all women and those married at the time of the census
455.125 observations for the first one and 404.795 observations for the last. Table 2 shows
some descriptive statistics and variable definitions for covariates, instruments and dependent
variables.
3
As Angrist and Evans argue census does not track children across households; therefore, the sample is
limited to mothers aged 21-35 whose oldest child was less than 18 years of age at the moment of the census.
5