Patricia Cohen's selective use of economic statistics in May 2015

Economy Stats

The economic statistics cited by Patricia Cohen are meant to put a positive spin on an otherwise doubly-seasonally adjusted economy still limping under the Federal Reserve's guidance and policy of quantitative easing. Cohen calls it "strong job growth data," but there are scores of other stats that researchers use to indicate just the opposite, such as the correlative study of Labor Force Participation Rate and People not in the Labor Force by Durden (2015). Cohen's study is meant to put a good face on an otherwise grim economic outlook, and this paper will show how she does it.

Statistical Procedures Mentioned in the Study

There are no statistical procedures clearly delineated in the study by Cohen; however, it is evident that Cohen uses a number of procedures in order to present her findings. For example, Cohen uses correlations and ANOVA statistical analysis in order to arrive at her conclusion that, indeed, the economy is looking stronger. One correlation study that she uses is the relationship between higher wages being offered and an increase in jobs by 280,000 for May 2015. Cohen argues that the increase in wages may be what has led more people into the workplace: "Hourly wages…rose 0.3% last month, possibly helping to lure back some discouraged workers who had been staying on the sidelines" (Cohen, 2015). The argument is a dubious one that is based on nothing other than a supposed correlation and an apparent desire to see something positive in the latest round of economic data. Cohen ignores other possible correlation studies and even accepts the data generated by the Bureau of Labor Statistics (BLS) as authentic without questioning them or authenticating them on her own. This level of research is unacceptable for a serious level of analysis and therefore throws suspicion upon her entire study.

Her other approach, the ANOVA, relies on a combination of a number of different charts indicating, supposedly, that the current state of the economy is better than it has been over time. Again, Cohen relies on the BLS to show that the unemployment rate has decreased over the past year and that the labor force has increased. She juxtaposes these findings with demographics of unemployment, showing that for each demographic unemployment has decreased over the past year. She also includes percentage increases in type of work and average weekly earnings to support her argument that the economy does appear to be looking stronger. She also cites the positive opinions of persons like James Bullard of the St. Louis Fed in order to criticize the anxiety of the IMF, which has suggested that the Fed should not raise rates this year because the economy has not yet stabilized and sufficiently turned around.

Conclusions

Cohen concludes that while the economy may be looking better there are still some signs that need to be displayed in order for recovery to be clearly on course. One of those signs is when employers "convert their temporary hires into permanent workers" (Cohen, 2015). When that happens along with a much larger increase in labor participation rates, the economy will be safely on its way to prerecession times of plenty and prosperity.

Reasoning behind the Appropriateness

The reasoning behind this assessment of the appropriateness of the study is that Cohen's method of sampling is based on the BLS which does not provide an accurate assessment of the real state of the economy. The processes for gathering the data are imprecise and lack the "rigid method" that Snider calls for. Therefore, much of Cohen's argument is based on less than solid findings and data reports. The number of charts she includes in her study are helpful in effecting the positive image she wants to portray, but anyone looking at the actual numbers contrasted with those of the BLS would see what Snider sees, which is that the economy is not doing nearly so well as some would have us to believe.

Were the Findings Statistically Significant?

No, the findings were not statistically significant, mainly because they relied on data that comes from a club-rule source (the BLS), which has been criticized for inaccurately reporting statistics about the economy (Durden, 2015; Snider, 2015). The findings are more akin to a political gesture or stance than a dispassionate or objective analysis. Therefore, it would be remiss to assess them as statistically significant because they lack the character of scientific analysis.

Level of Significance

The significance of this study is at a level of >.10 because they do not rate as significant on any serious level, but rather only on the pseudo-political level. The process used to arrive at this determination was based on outside research, which sought to verify or deny the findings put forward by Cohen. Evidence from third party researchers indicates that Cohen's assessment is mired, at best, in wishful thinking and an eagerness to see strong data where none actually exists. For instance, just because more low quality jobs have been made available does not in any way indicate the economy is on its way to recovery, especially since the stagnating wage growth suggests just the opposite as well as the inevitably painful effects of several years' worth of quantitative easing on the part of the Fed. Cohen's goal is to convince the reader that it is time to end quantitative easing, and this may be true, but the method of persuasion she uses -- namely that the economy is in recovery-mode, lacks serious substantiation. The stats she uses and the analysis she employs are not up to scholarly par and are based on data that are questionable at best. The fact that Cohen describes the data as "strong job growth data" in her title is enough to implicate her because the data is anything but.