Apply


Why Data Still Matter When Nothing Else Does

LinkedInFacebookTwitterPrintFriendlyShare

Reading naval history, one of my pastimes, provokes in me a range of different emotions. Perhaps the most common one, as the late historian Barbara Tuchman so eloquently described in her summation of fruitless human combat The March of Folly, (Random House, 1984) is a sense of resignation and a recognition that George Santayana was way off base—even if we do remember the past, we persist in repeating unforced errors. As much as we praise reason as the one factor that distinguishes us from other sentient beings, the lack of reason, as measured by greed, hubris, egotism and untrammeled anger is responsible for most human caused disasters of any scale at all.

But then a sense of optimism, however unwarranted, sneaks in when I realize that as bad as things seem now, we have been through similar rough patches and emerged relatively unscathed as individuals and as a society. Ian Toll’s latest work describing the end of World War II (The Twilight of the Gods, Norton, 2020) serves as an exemplar. Throughout that conflict, we were not, as our current lazy recollection has it, a unified people fighting against a monolithic enemy. Political battles were as fierce then as today, with segments of the press controlled largely by William Randolph Hearst and his allies loudly and continuously denouncing anything Franklin Delano Roosevelt did. Roosevelt and his allies in turn withheld important national news from the press, indiscriminately employing the excuse of “wartime secrets” whether or not the news posed a real threat to national security. The war effort was inescapably caught up in politics, with Douglas MacArthur, even while serving as military Commander of the Southwest Pacific Area, running as a dark horse candidate for the Republican nomination in 1944. In the period before the war, an adamant and powerful anti-immigrant movement was ascendant, called, then as now, America First. Though universally xenophobic, their principal targets were not immigrants and refugees on our southern border but instead Jewish refugees from Europe. Whatever group was singled out, their motivation and actions were essentially indistinguishable from those of current xenophobes. The bad things in life stay that way.

Regardless of the ends to which we twist them, however, data still matter. In times of turmoil when the temptation is to retreat into our own belief systems and resolutely ignore other points of view, data can be our salvation. The data are often not comforting, and often they directly challenge our thinking, but we ignore them at our peril.

Why is this important to our profession? Primarily because our training has taught us to be careful and accurate consumers of data and it is at least in part our responsibility to model to others how to interpret data. Whether such data emerge from psychological investigations or from other fields, we know what a sound research design looks like and—most importantly—we know the limitations of the data and how this affects the authors’ conclusions. Data, if properly collected, don’t lie. But they also are known to stretch the truth a bit, and researchers are infamous for ignoring the limitations of their own data. I recently reviewed a paper for a credible scholarly journal. After noting that the groups under investigation had fewer than 15 observations per cell yet the authors blithely persisted in performing t-tests, I began to suspect that they had submitted the paper as an exemplar of what not to do in conducting a proper analysis. But even with ineligible data, the authors continued their analysis. When I later found that the results of their analyses had yielded no statistically significant findings and the authors had then highlighted a secondary finding not included in their original research questions because this observation had approached significance, my suspicion grew stronger. Alas, it seemed that the authors had submitted the paper not to illustrate what not to do, but in the hopes that all their hard work would somehow get published. I am sure that somewhere it will.

Experimental data tend to be relatively straightforward to interpret, but much of our attention these days is focused on epidemiological data, which present a different set of challenges in determining causality. Data that rates of COVID-19 deaths are higher in nursing homes that have primarily African-American residents are compelling. Comparing those homes with those having low numbers of African-American patients is relatively simple presuming other factors are adequately controlled for. An association also seems apparent between local rates of COVID-19 in college towns where schools offer in-person classes. Universities may or may not adopt rules against in-person learning, but they cannot ignore the data. Similarly, an association between infection rates and large gatherings of people not observing mask precautions or social distancing seems clear, but tracing the origin of infections to such gatherings is difficult. Here it is legitimate to argue about how compelling the data are, but again, what we cannot do is ignore them. We know that social distancing and mask-wearing are protective against the virus. We can argue whether or not government mandates prescribing mask wearing are an unconstitutional infringement of individual liberties. At the same time, the data concerning viral transmission are well established. We ignore those data at our peril.

Another example that has struck uncomfortably close to home arises from the wildfires in the Western US. For me, being close to the epicenter of the Almeda fire has instilled some urgency in how we assess the data regarding climate change. At this point, the data seem clear that declines in annual rainfall and increases in regional surface temperatures are associated with higher incidence of wildfires in the Western US. What we do with these data is of course a political, not a scientific, decision. But we cannot ignore the science if we hope to make any progress in mitigating the effects of climate change. As I’ve noted before, psychology has a role in helping our society address this complex problem. As psychologists, we can insist on the appropriate role of data in decision making, and we can help shape public response and avoid the problem of attentional fatigue. By assisting others in focusing on components of the problem that are amenable to solution we can assist in counteracting a sense of futility. These processes are, after all, not dissimilar to how we direct individuals through the process of therapeutic change. International bodies have recognized the role that our profession might play in addressing this global problem. World meteorological organizations have recognized the value of psychology in their deliberations. Psychological science has contributed to understanding of how the effects of global warming affect individuals and societies, a newer focus is on how we can directly contribute to ameliorating the negatives associated with climate change. It is our responsibility as a profession to continue to provide such input. It’s our responsibility as individuals to apply our education and skills towards helping others to understand the essential role of data in decision making.

Copyright © 2020 National Register of Health Service Psychologists. All Rights Reserved.

1200 New York Ave NW, Ste 800

Washington DC 20005

p: 202.783.7663

f: 202.347.0550

Endorsed by the National Register