MDG_2022_MES_v01_M
Micro Enterprise Survey 2022
MES 2022
Name | Country code |
---|---|
Madagascar | MDG |
Enterprise Survey [en/oth]
The Micro Enterprise Surveys (MES) implemented in Madagascar between March and December 2022. The MES collect both objective data based on firms’ day-to-day experiences, and perceptions of enterprises regarding the business environment in which they operate.
Sample survey data [ssd]
The 2022 Madagascar Micro Enterprise Survey covered the following topics:
National coverage
The universe of inference includes all formal (i.e., registered) private sector businesses (with at least 1% private ownership) and with at least five employees. In terms of sectoral criteria, all manufacturing businesses (ISIC Rev 3.1. codes 15-37) are eligible; for services businesses, those corresponding to the ISIC Rev 3.1 codes 45, 50-52, 55, 60-64, and 72 are included in the Micro Enterprise Surveys. Cooperatives and collectives are excluded from the Micro Enterprise Surveys. All eligible establishments must be registered with the registration agency. In the case of Madagascar, the definition of registration with INSTAT or EDBM was used.
Name |
---|
World Bank Group (WBG) |
Name | Abbreviation |
---|---|
World Bank Group | WBG |
The MES use stratified random sampling, where the population of establishments is first separated into non-overlapping groups, called strata, and then respondents are selected through simple random sampling from each stratum. The detailed methodology is provided in the Sampling Note.1 Stratified random sampling has several advantages over simple random sampling. In particular, it:
The MES typically use two levels of stratification: industry classification, and subnational region (used in combination). Starting in 2022, the MES bases the industry classification on ISIC Rev. 4 (with earlier surveys using ISIC Rev. 3.1). For regional coverage within a country, the MES has national coverage.
Note: Refer to Sampling Structure section in "The Madagascar 2022 Micro Enterprise Survey Implementation Report" for detailed methodology on sampling.
Overall survey response rate was 31.9%.
Since the MES uses stratified random sampling, individual observations should be properly weighted when making inferences about the population, since unweighted estimates are biased unless sample sizes are proportional to the size of each stratum. For each MES, special care is given to correctly compute sampling weights. Whenever the Universe is used to draw the sample (i.e., Sampling Frame is the same as the Universe) it is imperative to accurately adjust the universe within each stratum to account for the presence of ineligible establishments (e.g., the firm discontinued businesses, or is deemed ineligible due to its business activity or having fewer than five employees). Proper treatment of panel establishments is also crucial. Details about how the WBES sampling weights are calculated are given in the Sampling Note section of "The Madagascar 2022 Micro Enterprise Survey Implementation Report".
Three versions of sampling weights are calculated, depending on the assumptions that determine eligibility of establishments to be counted towards the WBES universe.
The questionnaire implemented in the Madagascar 2022 WBES included additional questions covering management practices, worker skills, and technology adoption. These questions were selected in collaboration with the members of the WB local country team.
The MES datafiles are organized in ways that reflect the corresponding questionnaire. The variables that are standard across countries have the first letter in their name correspond to the questionnaire section where the variable belongs in the questionnaire, i.e., a1 denotes section A. The variables that are country or region specific are generally named using a prefix in capital letters. These country-specific variables are generally not available beyond the specific country or region where they were implemented. For the Madagascar 2022 MES, these variables use prefixes MG. All variables are numeric with the exception of those variables with an “x” at the end of their names. The suffix “x” denotes that the variable is alpha-numeric.
The MES datafiles contain two establishment identifiers, idstd and id. The former is a global unique identifier of each establishment, while the latter is unique identifier within each survey. The variable idstd can be used to match the WBES establishment one-to-one across databases. The variables wweak, wmedian, and wstrong corresponding to sampling weights based on, respectively, weak, median, and strong assumptions about eligibility (see Section III.6 of "The Madagascar 2022 Micro Enterprise Survey Implementation Report"). The variable strata corresponds to the stratum of each observation.
Additionally, the MES datafiles contain many standard variables. The variable d1a2_v4 denotes the main activity of the establishment, as obtained during the WBES interview, coded in the four-digit ISIC Rev. 4. Users should note that this activity may differ from the industry classification given in the Sampling Frame, a4a. Users are generally advised to use industry categories based on the realized information in d1a2_v4. Additional sampling information is contained in variables a2 (region) and a6a (size). The variable panel identifies panel establishments, there are no panel establishments in Madagascar 2022. The combination of a4a, a2, a6a, and panel forms stratum of each establishment, which is contained in variable strata.
The last complete fiscal year for each establishment is contained in variables a20m (last month of last complete fiscal year) and a20y (last complete fiscal year).
Note that when an entry in the MES database is empty, this means that the question was not asked to the corresponding respondent. This happens when the question is deemed inapplicable, due to skip patterns or other reasons. In cases when the question was posed, some entry is provided, including don't know, which is an example of item non-response.
Start | End |
---|---|
2022-03 | 2022-12 |
Detailed information on the WBES methodology and data collection is provided in the Enterprise Surveys Manual and Guide (https://www.enterprisesurveys.org/content/dam/enterprisesurveys/documents/methodology/Enterprise-Surveys-Manual-and-Guide.pdf). The interviews for the Madagascar 2022 MES were conducted between March and December 2022. The interviews were conducted in the following languages: English, Malagasy, and French . For monetary variables, the currency was Malagasy Ariary.
The fieldwork for the Madagascar 2022 MES was implemented by Dahlberg Research.
The World Bank Group. Madagascar - Micro Enterprise Survey (MES) 2022, MDG_2022_MES_v01_M. Dataset downloaded from https://www.enterprisesurveys.org/en/data on [date].
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.
Name | Affiliation | |
---|---|---|
Enterprise Analysis Unit | World Bank Group | enterprisesurveys@worldbank.org |
DDI_MDG_2022_MES_v01_M_WB
Name | Abbreviation | Affiliation | Role |
---|---|---|---|
Development Data Group | DECDG | World Bank | Documentation of the study |
2025-02-05
Version 01 (February 2025)