Average Monthly Wage Means Difference Between Administrative Sources Vs. Business Survey, Case in Albania

Abstract

This paper examines the differences resulting from calculating the means of Pay Roll records and personal revenues used as secondary data with results from business survey in an empirical study taking a panel. The use of “secondary data” as primary source for producing the official indicators is a challenge worldwide. In the past decades has also been considered as the way forward for raising productivity and reducing burden on businesses. If the Short Term Survey is sample survey the Pay Roll records are administrative data. The purpose for what they are gathered is different. But both could be used for providing statistical indicators. In this paper the panel not weighted data are taken into consideration where the same business is analyzed from two related sources. The paired t-test is used to compare the values of means from two related sources. In those conditions the difference between the means of the two sources is unlikely to be equal to zero. In this study the hypothesis test is designed to answer the question "Is the observed difference sufficiently large enough to indicate that the alternative hypothesis is true?" What does it mean in our case study the answer which comes in the form of a probability - the p-value? The paper shows some interesting findings about the means difference between the two sources within a year. The differences resulting from the conducted analysis come as a result of the definition used in both sources for the same indicator, errors in reporting and treatment of non-response in the survey and administrative data source, coding errors.