Assessing Accuracy and Coverage of Housing Units Sampling Frame using the Post-Frame Update Survey (PFS) Method

Survey research using nationally representative samples is growing substantially in the rapidly developing countries of the Gulf Cooperation Council (GCC). However, in such countries the population and their demographic makeup change at some of the fastest rates in the world. In Qatar for instance, government censuses tend to be undertaken every five to six years, while the population and its demographic composition change very quickly (Alrouh, Ismail & Cheema, 2013). In 2004, the population of Qatar was just 798,000 and Qataris represented 24% of the country’s residents. The 2010 census measured the population at 1,699,000, only 17% of which were Qataris.In such countries therefore, sampling frames based on censuses may quickly become outdated as demographic compositions change considerably over short time periods, more so than in other countries. These contexts present a challenge for survey research, as the target population may differ significantly from the sampling frame shortly after it is constructed. Outdated frames may lead to coverage and content errors that not only affect the estimates produced by a survey but also operational processes associated with data collection. Frame error can negatively impact time spent in the field and survey costs. Periodic evaluation of the quality of a sampling frame used between censuses is required for estimating coverage bias and informing researchers about the quality of the frame data. Adjustments in field operations can then be made to improve the effectiveness and efficiency of data collection.This paper describes the Post-Frame update Survey method conducted by the Social and Economic Survey Research Institute (SESRI) at Qatar University. It aims to assess the accuracy and coverage of its housing units sampling frame and improve field operations. This method of evaluating sampling frames could be applicable to other research organizations that survey rapidly changing populations.