To Örebro University

Editing for measurement errors is always part of data processing. In traditional editing, all data records are checked for errors and inconsistencies. In a new way of editing, only the subset with the most important erroneous responses is considered for editing. This approach is applied in selective editing procedures, which have been shown to save resources considerably. However, selective editing lacks a probabilistic basis and the properties of estimators cannot be established using standard methods. In particular, bias properties of the estimator are unknown except for level estimates based on historical data. This paper proposes combining selective editing with an editing procedure based on the traditional probability-sampling framework. The variance of a bias-corrected Horvitz-Thompson estimator is derived and a variance estimator is proposed. The results of a simulation study support the use of the combined editing procedure.

The purpose of editing is to correct erroneous entries in the dataset and assure the quality of data. It takes a lot of resources to correct all errors, so editing procedures where only a subset of errors to be corrected are sought after. Correcting only a subset of all errors will influence the final estimates, and tools evaluating the properties of the estimates like bias and variance need to be available. This paper introduces a probabilistic editing procedure where the responses are selected for editing through Poisson Mixture (PoMix) sampling and a bias adjusted GREG estimator is used for estimation. An expression for the variance of the bias adjusted GREG estimator is derived, and variance estimator is proposed. The effectiveness of the proposed editing procedure and the GREG estimator is illustrated using empirical data from Statistics Sweden.

Probabilistic editing has been introduced to enable valid inference using established survey sampling theory in situations when some of the collected data points may have measurement errors and are therefore submitted to an editing process. To reduce the editing effort anavoid over-editing, in current practice selective editing is most often used, which is a form of editing that limits the edit checks to those potential errors that, if indeed in error, are likely to have the biggest impact on estimates to be produced. However, selective editing is not grounded in probability theory associated with survey sampling, and cannot provide expressions for point and variance estimates that account for the uncertainties introduced by selective editing.

In the spirit of the total survey error paradigm, this paper extends the previous work on probabilistic editing by proposing an estimation procedure that provides valid inference when two kinds of nonsampling error are simultaneously present, in addition to the sampling error: the measurement error, requiring an editing step, and the practically unavoidable nonresponse error which also needs to be taken into account when producing unbiased estimates.

In a three-phase selection setup, bias due to measurement error is estimated through probabilistic editing while weight adjustment employing auxiliary information is used to deal with nonresponse. An estimator based on calibration for nonresponse and corrected for bias due to measurement error is introduced. Its theoretical variance and an estimator of the variance are derived. A simulation study illustrates the three-phase selection setup and the practical performance of the derived point and variance estimators.