Data used for multilevel regression of quantity of insecticide and type of container, material of container, manufacturer of insecticide with district, block and measuring container (squad) as levels of clustering of data.

Published: 13-03-2018| Version 1 | DOI: 10.17632/f5rtkzy9s5.1
Suman Saurabh


This data describes the readings of quantity of insecticide measured across 46 blocks (sub-districts) of 12 districts of Bihar, India. The hierarchical nature of the data is also visible as it is organized across squads, blocks and districts. Type of measuring container, material of measuring container and manufacturer of insecticide (anonymized as A or B) is also provided. This data can be used to perform Multilevel linear regression with 'value' (quantity of insecticide measured) as independent variable and 'containercode' (type of container), 'cmatrlcode' (material of container), 'pmanfcode' (manufacturer of insecticide powder) as dependent variables. Distrct, block and squad can be assigned random effect (random intercept).


Steps to reproduce

1. Open STATA data analysis software (Stata Corp., College Station, Texas, USA) 2. Open the above STATA data file (.dta) 3. Type the command: xtmixed value containercode cmatrlcode pmanfcode || district: || block: || squad: 4. Press enter and visualize the output