Human Development Cash Transfer Program Impact Evaluation - Midline Household Survey 2018
Producers and sponsors
The following was provided by CAETIC Développement to summarize the methodology used and constraints faced in collecting data for the midline survey.
As with the baseline survey, Ideas42 designed the mid-term survey methodology. It consists in a representative survey which includes two components: a household survey and an administration of a test, called MDAT, for children aged from 24 to 71 months old. The sampling method, the different parts of the household questionnaire, the MDAT test, and the different concepts of the mid-term survey were kept identical to those of the baseline survey, except some minor changes on the household questionnaire in order to keep a certain harmonization with the FIAVOTA questionnaire.
Sample size for the mid-term survey was 6006 households, comprising 1522 additional households, distributed among 925 beneficiary households and 597 control households. It is important to stress that these additional households were drawn from the same fokontany (primary unit) as for the baseline survey. For the MDAT test, sampled children were randomly drawn from an exhaustive list of eligible children (aged from 24 to 71 months old) identified within the sample of household.
Deviations from the Sample Design
Replacement rate for the TMDH household survey was about 13.5% (816/ 6006 households). Among these 816 households, 651 households were replaced because of their absence during the survey, 114 households were not found, and 45 households refused to participate in the survey (no consent).
From the 6006 surveyed households, 3748 were eligible for the MDAT test, representing 62.4% of the sample, from which 4903 children could be observed. For the MDAT test in particular, 591 children were replaced because of diverse reasons. Thus, in sum only 3376 children were tested, at a rate of 1 child per household. Indeed, 342 eligible households for the MDAT test were not surveyed because of refusal or unavailability of the child. Besides, no households’ nor mothers’ refusal were recorded during the administration of the MDAT test.
Dates of Data Collection
Data Collection Mode
Computer Assisted Personal Interview [capi]
Data collection mobilized a total of 144 individuals including 6 supervisors, 16 controllers, 122 investigators. Teams were endowed of all the necessary tools and equipment to carry out properly the survey (4x4 cars, laptops, generators, converters, GPS systems, and kits for children tests). Data were collected at rather a month after the last transfer, from 23rd May until 30th June 2018, while respecting the duration fixed by the terms of reference, namely 30 days after the last transfer payment.
Data Collection Notes
As the main task of CAETIC Développement was to implement the established survey protocol and to make all the necessary effort to ensure high data quality, the overall approach adopted by CAETIC Développement was established upon the following aspects:
1. Investigators’ recruitment and training: Investigators were preselected on the basis of various criteria, including but not limited to intellectual, physical, and moral quality. Preselected investigators followed a 12 days training. For the MDAT test, preselected investigators were assessed on their ability to perform high quality child observations, and were certified later on. Only certified investigators were recruited for the data collection. It should be stressed that even the investigators who participated for the baseline survey were trained and assessed on the same way as the new candidates;
2. Use of CAPI (Computer Assisted Personal Intrerview) approach for data collection: this means using an electronic questionnaire during the interview. The CAPI approach allows data entry directly on the computer during the interview and exhaustive control of the questionnaire, with a special focus to completeness, jumps, coherences, and likelihoods;
3. Implementation of an operational plan to insure optimal control in the field: The survey was carried out by 22 teams. Each team was formed by two or three pairs of investigators, under the supervision of a team leader, whom could be a controller or a supervisor. For the MDAT test, investigators worked in a team formed by an examiner and an assistant. Fieldwork was organized in such a way that each team of investigators doesn’t work out of reach of their supervisors.
Use of the dataset must be acknowledged using a citation which would include:
- the Identification of the Primary Investigator
- the title of the survey (including country, acronym and year of implementation)
- the survey reference number
- the source and date of download
World Bank. Human Development Cash Transfer Program Impact Evaluation - Midline Household Survey 2018 (HDCT 2018). Ref: MDG_2018_HDCT-ML_v01_M. Downloaded from [uri] on [date].
Disclaimer and copyrights
The user of the data acknowledges that the original collector of the data, the authorized distributor of the data, and the relevant funding agency bear no responsibility for use of the data or for interpretations or inferences based upon such uses.