Estimates of sampling errors
---> Sampling Errors
Data of this survey affected by sampling errors due to use of the sample and not a complete enumeration. Therefore, certain differences are expected in comparison with the real values obtained through censuses. Variance were calculated for the most important indicators, the variance table is attached with the final report. There is no problem to disseminate results at the national level and government level.
---> Non-Sampling Errors
Non-statistical errors are probable in all stages of the project, during data collection or processing. This is referred to as non-response errors, response errors, interviewing errors, and data entry errors. To avoid errors and reduce their effects, great efforts were made to train the fieldworkers intensively. They were trained on how to carry out the interview, what to discuss and what to avoid, carrying out a pilot survey, as well as practical and theoretical training during the training course.
Also data entry staff were trained on the data entry program that was examined before starting the data entry process. To stay in contact with progress of fieldwork activities and to limit obstacles, there was continuous contact with the fieldwork team through regular visits to the field and regular meetings with them during the different field visits. Problems faced by fieldworkers were discussed to clarify any issues.
Non-sampling errors can occur at the various stages of survey implementation whether in data collection or in data processing. They are generally difficult to be evaluated statistically. They cover a wide range of errors, including errors resulting from non-response, sampling frame coverage, coding and classification, data processing, and survey response (both respondent and interviewer-related). The use of effective training and supervision and the careful design of questions have direct bearing on limiting the magnitude of non-sampling errors, and hence enhancing the quality of the resulting data.
The implementation of the survey encountered non-response where the case (household was not present at home) during the fieldwork visit and the case (housing unit is vacant) become the high percentage of the non response cases. The total non-response rate reached 4.7% which is very low once compared to the household surveys conducted by PCBS, The refusal rate reached 0.8% which is very low percentage compared to the household surveys conducted by PCBS, and the reason is the short and clear questionnaire.
Data appraisal
Errors in data processing, such as coding and punching. The data underwent checking and completion of missing information in the office and logical checks were computerized as well as manually, including call-backs if needed.
Response errors which resulted from misunderstanding of the questions, interviewers’ bias in asking the questions and in probing. Thorough training, supervision, and various quality control checks were used to minimize bias resulting from these kinds of errors.