Application of sampling to statistics of industrial establishments

Abstract

This document examines the methodologies employed in conducting industrial establishment surveys across various African countries, with a focus on their application in collecting manpower statistics. These surveys are crucial for manpower planning, providing detailed data on employment, wages, occupational distribution, and educational standards. While industrial surveys often yield manpower data as a by-product, specialized manpower surveys can also be conducted through direct interviews with establishment heads or employees. The statistical unit typically used is the "establishment," distinguished from "enterprises," with stratification based on criteria such as geographical location, economic activity, and establishment size. Sampling methods vary, with single-stage stratified sampling being the most common. The document highlights specific examples from countries like Gabon, Libya, Tunisia, Ghana, and Cameroon, illustrating diverse approaches to survey design, sampling frames, and data collection. Despite the importance of these surveys, non-sampling errors often exceed sampling errors, and the accuracy of estimates can be compromised by methodological inconsistencies.