Christopher J. Cronin, David K. Guilkey, and Ilene S. Speizer, 
"Measurement Error in Health Facility Choice Models: An Example from 
Urban Senegal", Journal of Applied Econometrics, Vol. 34, No. 7, 2019, 
pp. 1102-1120.

The data used in this paper are the property of the Carolina
Population Center (CPC) at the University of North Carolina. The
survey instruments used to generate the data are publicly available.
Please see https://dataverse.unc.edu/dataverse/mle_senegal.

Our main data file combines three features of the Measurement,
Learning, and Evaluation (MLE) data collected in Senegal: (i)
individual-level survey data, (ii) facility-level survey data, and
(iii) a link file that matches individuals to the health facilities
that they visit for various health services. These data files may be
requested though the Carolina Population Center Data Portal for the
Measurement, Learning & Evaluation (MLE Senegal Baseline) project:

https://data.cpc.unc.edu/projects/14/view

Interested researchers should note that files (i) and (ii) are
publicly available subject to the signing of a data use agreement.
File (iii) has not been released publicly; the CPC considers access to
this file on a case-by-case basis.

Upon acquiring these data files, the main data file used in estimation
can be reproduced using the included Stata file "Clean_Data.do".
Estimation results can be reproduced using the included Stata file
"Results.do".

Both .do files are zipped in the file cgs-files.zip. They are ASCII 
files in DOS format.