HPP Subsequent Consultation Manuscript Analytical Dataset Construction Do-file
Description
This Stata do-file contains the code used to construct the analytical dataset for the study from the original Tanzania National Panel Survey (NPS) data files. It documents the data-management procedures required to identify the study population, extract and harmonize relevant variables across survey waves, merge the required household- and individual-level information, recode variables, and generate the final variables used in the empirical analysis. The do-file produces the final analytical dataset used to generate the descriptive statistics and econometric results reported in the study. It therefore provides a reproducible record of how the original NPS survey data were transformed into the study-specific analytical dataset. The underlying NPS microdata are not redistributed in this repository because redistribution is not permitted under the data-access conditions of the Tanzania National Bureau of Statistics. Researchers wishing to reproduce the analytical dataset should obtain the original NPS data from the World Bank Microdata Library and execute this do-file using the corresponding source files. File type: Stata do-file (.do) Purpose: Construction of the study analytical dataset Source data: Tanzania National Panel Survey (NPS), Waves 2–4 Output: Study-specific analytical dataset used to produce the reported empirical results Software: Stata
Files
Steps to reproduce
The dataset used for the study was derived from the Tanzania National Panel Survey (NPS) data is available through the World Bank Microdata Library (https://microdata.worldbank.org/catalog/3814; https://microdata.worldbank.org/catalog/1050; https://microdata.worldbank.org/catalog/2252; https://microdata.worldbank.org/catalog/2862). Steps in obtaining g the analytical dataset: 1. Merge sections a, b, c, d, e, i1, j1, and k of the NPS Tanzania Universal Panel Dataset 2. Merge the resulting dataset from step 1 with HH_SEC_A.dta from wave 2 3. Merge the resulting dataset from step 2 with COMSEC_CB.dta from wave 2 4. Merge the resulting dataset from step 3 with HH_SEC_A.dta from wave 3 5. Merge the resulting dataset from step 4 with COMSEC_CB.dta from wave 3 6. Merge the resulting dataset from step 5 with hh_sec_a.dta from wave 4 7. Merge the resulting dataset from step 6 with com_sec_cb.dta from wave 4 8. Merge the resulting dataset from step 7 with TZY2.HH.Consumption.dta from wave 2 9. Merge the resulting dataset from step 8 with ConsumptionNPS3.dta from wave 3 10. Merge the resulting dataset from step 9 with consumptionNPS4.dta from wave 4