Suchergebnisse

Reducing the response burden and widening available statistical information necessitate new approaches in the National Statistical Institutes production process. Our article focuses on longitudinal data needs. Two approaches for building business longitudinal data in a context of cross‐section surveys and administrative sources information are considered. The article describes construction approaches and evaluates the quality of two data bases obtained through multisources integration. The computed databases aim to represent the target population of Italian firms with 20 persons employed and over. The similarity of the distribution of the main economic variables between the target population and the computed databases is considered a basic criterion in evaluating the quality of the created databases. To this end, rank correlation, and the Fligner–Policello test are applied. In addition, representativeness R indicators are computed. No differences are found between distributions.

AbstractThe effects on employment growth in firms, grouped by size class, of the economic crisis that began in 2008 are analysed using multifactor partitioning (MFP). Italy's employment growth is decomposed into four explanatory factors: the stage in the business cycle; the effect of firm size; industry composition; and regional distribution; together with the interactions among these four effects. The interpretation of these effects is facilitated by the introduction within the MFP framework of a new decomposition of several key elements. The results show that the adopted approach and the suggested decompositions are useful to study the effect of size on employment change. This effect is found to be negative only for micro units (with less than 10 employees). For the other classes, it is positive. The observed negative changes in these classes are mainly due to the business cycle and an unfavourable industrial composition.

Handbook of Web Surveys presents a self-contained guide to methodological issues related to web surveys, covering both theoretical and practical aspects of the topic. The book first approaches the subject of web survey design, addressing the main question: How can an electronic questionnaire be designed such that respondents can complete it on the Internet without introducing (too many) errors? The second portion of the book is devoted to sampling issues and helping readers understanding how proper samples for web surveys can be selected to allow for unbiased estimation of population characteristics. The authors also explore the topics of under-coverage, self-selection, adjustment weighting, and propensity scores as they pertain to effective web survey development. Each chapter follows the same easy-to-follow format. Following an introduction, a description of theory is presented along with key formulae. Next, the discussed theory is applied to a real data set and accompanied with illustrative descriptions. Example sets and exercises are spread throughout each chapter, and a summary provides a brief overview of main points and concepts. Real data is used throughout the book, and a related Web site features additional data sets, interactive simulations, and solutions to exercises

Abstract
Probability sample (PS) surveys are considered the gold standard for population-based inference but face many challenges due to decreasing response rates, relatively small sample sizes, and increasing costs. In contrast, the use of nonprobability sample (NPS) surveys has increased significantly due to their convenience, large sample sizes, and relatively low costs, but they are susceptible to large selection biases and unknown selection mechanisms. Integrating both sample types in a way that exploits their strengths and overcomes their weaknesses is an ongoing area of methodological research. We build on previous work by proposing a method of supplementing PSs with NPSs to improve analytic inference for logistic regression coefficients and potentially reduce survey costs. Specifically, we use a Bayesian framework for inference. Inference relies on a probability survey with a small sample size, and through the prior structure we incorporate supplementary auxiliary information from a less-expensive (but potentially biased) NPS survey fielded in parallel. The performance of several strongly informative priors constructed from the NPS information is evaluated through a simulation study and real-data application. Overall, the proposed priors reduce the mean-squared error (MSE) of regression coefficients or, in the worst case, perform similarly to a weakly informative (baseline) prior that does not utilize any nonprobability information. Potential cost savings (of up to 68 percent) are evident compared to a probability-only sampling design with the same MSE for different informative priors under different sample sizes and cost scenarios. The algorithm, detailed results, and interactive cost analysis are provided through a Shiny web app as guidance for survey practitioners.