Introduction
The U.S. Bureau of Labor Statistics (2018a) published a short article that reports the results of the analysis of the changes in the “nonfarm payroll employment” in metropolitan areas “for the year ending February 2018” (para. 1). The article does not include a detailed description of the study that produced the results, but it provides a link to the report by the U.S. Bureau of Labor Statistics and the U.S. Department of Labor (2018) with a technical note. The latter offers more details, showing that the methodology was prone to notable errors that might have affected the study’s accuracy and precision.
Main body
The variable of interest for the study was the “nonfarm payroll employment.” It was measured every month using ratio-level data: the numbers of people employed were used (Healey, 2015). The changes were also depicted with the help of a chart that used percentages (U.S. Bureau of Labor Statistics, 2018a). Technically, there was only one variable (non-farm employment), which is why the distinction between dependent and independent variables is not very applicable to this analysis (Levine, 2014). The study is descriptive (Healey, 2015; Levine, 2014); the investigators did not manipulate any variables, and no variables were introduced to explain the changes in employment.
The study incorporated the results of the Current Employment Statistics (CES) program, which produces employment data through estimation. For large samples, CES uses the “weighted link relative estimation technique,” in which a sample of establishments is studied to calculate the “ratio of current month weighted employment to that of the previous-month weighted employment” (U.S. Bureau of Labor Statistics & U.S. Department of Labor, 2018, p. 4). The latter is then used to calculate employment estimates. For small or highly variable samples, CES chooses models based on sample estimates and historical data (U.S. Bureau of Labor Statistics & U.S. Department of Labor, 2018, p. 5).
The U.S. Bureau of Labor Statistics and the U.S. Department of Labor (2018) report that the data of the study is subject to multiple limitations and errors that originate from its use of surveys and models. The source focuses on the sampling errors, which have been calculated and are available for review (U.S. Bureau of Labor Statistics, 2018b). However, non-sampling errors, including those related to data collection, are also mentioned for the study, even though no information about them is available for checking (U.S. Bureau of Labor Statistics & U.S. Department of Labor, 2018, p. 5). Thus, the analyses are subject to measurement errors (Levine, 2014).
Conclusion
Only the standard errors of the study can be reviewed (U.S. Bureau of Labor Statistics, 2018b). However, since the anticipated standard and random errors for the survey are notable, it can be assumed that the results are not highly accurate or precise (Ogundare, 2015). Still, the U.S. Bureau of Labor Statistics and U.S. Department of Labor (2018) report controlling the errors and provide proof to at least some of their efforts (U.S. Bureau of Labor Statistics, 2018b), which implies that the accuracy and preciseness of the results are adequate.
References
Healey, J. (2015). Statistics. New York, NY: Cengage Learning.
Levine, G. (2014). Introductory statistics for psychology. Cambridge, MA: Academic Press.
Ogundare, J. (2015). Precision surveying. New York, NY: John Wiley & Sons.
U.S. Bureau of Labor Statistics, & U.S. Department of Labor. Metropolitan area employment and unemployment — February 2018. Web.
U.S. Bureau of Labor Statistics. Nonfarm payroll employment increased in 313 metropolitan areas for year ending 2018. The Economics Daily. Web.
U.S. Bureau of Labor Statistics. (2018b). Reliability of state and area estimates. Web.