Methodology of the Informal Sector Enterprise Survey

The Informal Sector Enterprise Surveys use a two-stage sampling process.

The first stage uses a methodology called Adaptive Cluster Sampling (ACS). ACS requires a well-defined geographical area; for these surveys, it is typically an urban center. On a map of this delineated area, the Enterprise Analysis Unit creates a grid of evenly sized squares, called Block Areas or BAs, on the map and selects an initial sample of BAs at random (often within strata) without replacement. Within an enumerated BA, some basic information is listed on all encountered informal businesses, including through observation for those that refuse the exercise or those that are unavailable at the time of fieldwork. ACS is ‘adaptive’ in the sense that, using a pre-defined threshold number of informal businesses, all neighboring BAs of a BA that meets the threshold are subsequently sampled until there are no BAs that meet this expansion requirement. This process allows for first-stage weights to be calculated, allowing the Informal Sector Enterprise Surveys to be geographically representative of the urban areas where it is implemented. Full details of the methodology are provided in Aga et. al (2023).

The second stage of Informal Sector Enterprise Surveys involves the random selection of informal businesses within a BA to participate in an interview that lasts approximately 20–25 minutes. This interview consists of a standardized and tested questionnaire, properly designed for the circumstances of interviewing informal businesses. Aberra et al. (2022) provide further information on how these interviews are conducted. The selection for these longer interviews is done randomly and in real time. . As a result, a second-stage weight can be applied. Aga et. al (2022) provide some additional information on the selection process for the interviews, including adjustments to these weights based on assumptions about refusals and unavailable businesses.

The topics in the questionnaire generally include:

  • Business characterstics                                                 
  • Business owner’s background                                                    
  • Usage of electricity and water                                  
  • Linkages with suppliers and customers                             
  • Business practice 
  • Access to finance     
  • Workforce 
  • Sales and assets

Note that prior to 2015, the first stage sampling was done through a convenience sample approach, where a block is chosen randomly in a neighborhood of interest; enumerators then followed a path according to a predetermined set of instructions, enumerating every informal business encountered. Unlike the currently used two-stage sampling process, a convenience sample approach does not render a representative sample of informal businesses.

All the datasets are available through the portal, including documentation on the survey implementation details. These Implementation Reports contain information on the fieldwork, sampling procedures, dataset structure (including variable names), as well as information specific to the survey’s implementation.


Image

PAPER: An Application of Adaptive Cluster Sampling to Surveying Informal Businesses

Gemechu Aga, David C Francis, Filip Jolevski, Jorge Rodriguez Meza, Joshua Seth Wimpey

Informal business activity is ubiquitous globally, but it is nearly always uncaptured by administrative data, registries, or commercial sources. This article applies a well-established sampling method for rare and/or clustered populations, Adaptive Cluster Sampling (ACS), to a novel population of informal businesses. Generally, it shows that efficiency gains through the application of ACS, when compared to Simple Random Sampling (SRS), are large, particularly at higher levels of fieldwork effort. 

Image

WORKING PAPER: Understanding Informality : Comprehensive Business-Level Data and Descriptive Findings

Adam Aberra, Gemechu Aga; Filip Jolevski, Nona Karalashvili

This paper introduces and provides a descriptive analysis of data from more than 15,000 detailed interviews of representative samples of informal businesses operating in 24 cities across seven countries: India, Iraq, the Lao People’s Democratic Republic, Mozambique, Somalia, Zambia, and Zimbabwe. The paper is a companion paper to a study that presents the methodological underpinnings of the informal business data collection. It is an innovative application of area-based adaptive cluster sampling, rendering a representative sample of these businesses. The paper presents salient descriptive results of the data to motivate further research. 


Subscribe To Our Newsletter