These do files were used in the paper “Family structure and child development in Chile: A longitudinal analysis of household transitions involving fathers and grandparents” by Sarah A. Reynolds, Lia C.H. Fernald, Julianna Deardorff, & Jere R. Behrman All do files were written by Sarah Reynolds using Stata 14 Do-files’ Purposes Cleaning I: A general cleaning & harmonizing data sets from both years of the survey CASEN income: uses 2011 data to determine community level income for men & women Cleaning II: More specific data generation of household structure variables. Uses the data sets generated in Cleaning I & CASEN income Descriptive & Tables: Generates tables & outputs descriptive statistics, includes appendix tables Graphs: Figures 1 & 2 Appendix CASEN: generates figures for the national trends pictured in the appendix (Figure A-1) DATA SOURCES Apply the do files to the original data sets for ELPI 2010 & 2012. Information on the ELPI data is found here: http://www.microdatos.cl/Encuestas/PrimeraInfancia/LongitudinalPrimeraInfancia A translation of the introductory page (presentación) follows: The Microdata Center of the Department of Economics of the University of Chile conducted the Second Round of the Longitudinal Survey of Early Childhood (ELPI) in response to the need for information that would allow analysis of the national situation of children between 6 months and 7 years old. This was done in order to evaluate the effects of the programs oriented to this segment of the population, which had been implemented in previous years. The ELPI is a longitudinal survey, that is, designed for the periodic monitoring of the selected sample. In the second semester of 2009, a pilot study was applied to a sample of 650 children from the Metropolitan Region, which allowed the selection of instruments. The first round of ELPI was commissioned by the Ministry of Education and was carried out in 2010, between the months of March and June, when a sample of approximately 15,000 boys and girls from all regions of the country was surveyed. The second round was commissioned by the Ministry of Labor to be applied in 2012 to a sample of approximately 18,000 children throughout the country, 15,000 of which corresponded to the monitoring of the 2010 sample plus a sample of another 3,000 children from younger cohorts. The objective of ELPI is to collect data on the socio-demographic situation of each of the household participants, in addition to gathering information on the physical and socio-emotional development of the selected children. To this end, a complete set of instruments suitable for each age group was chosen. The state of socioemotional and physical development of the mother or the person in charge of the care of the selected child, who lived with him or her within the home, was also evaluated. For the administration of the tests and in order to maintain the objectivity of the data and to safeguard the use of psychological material, psychologists evaluated the socio-emotional, cognitive and physical development of both the children and the adult in charge. In this way, the ELPI represents an instrument that is not only fundamental, but also unique to study the reality of early childhood in Chile. The data for both 2010 & 2012 can be acquired by contacting the Centro Microdatos (bd@microdatos.cl). I have included the publicly available questionnaires, which serve as codebooks. The codes before the questions correspond to the variable names. The psychological variables, however, are not part of this codebook/questionnaire. The variable names for the Peabody Picture Vocabulary Tests, the Child Behavior Checklist, and the Wais intelligence test of numeracy & vocabulary are tvip_pb, cbcl1_pb_t, wais_pb_num, wais_pb_vo in both 2010 and 2012. In our study, we use all the children in the 2010 sample who were followed up in 2012 (12,898) for some descriptive statistics (Table 1, for example), and we use smaller samples in the longitudinal analysis. The selection and sample size of these smaller samples are described in section 2.2 of the paper. A few variables (e.g. birthdate & cluster) are not available for public use. Data sets from CASEN (La Encuesta de Caracterización Socioeconómica Nacional – Survey of National Socio-economic Characterization) are included in this replication file package, but I recommend downloading the original: http://observatorio.ministeriodesarrollosocial.gob.cl/casen/basededatos_historico.php