Importance of Questioning Statistics
People are accustomed to trusting statistical data because they are believed to be representative of the general population and to yield valid findings on various processes, economic developments, and social phenomena. However, both D’Ignazio and Klein (2020) and Wheelan (2014) illustrate how wrong it is to blindly perceive statistical data as authentic. These readings provide insights into the likelihood of errors in reporting channels and wrong assumptions based on statistical data. By including personal anecdotes and conducting thorough analyses of the information, the authors make a compelling argument for the questionable reliability of statistics.
Review of the Sources
Both readings emphasize the importance of treating statistical data with a pinch of salt. Wheelan’s book (2014) focuses on common mistakes in processes such as regression, statistical analysis, and program evaluation, among others. The author’s humorous, personal style of writing, his use of funny comparisons, and references to numerous research studies and examples make the book not only informative but also highly exciting to read.
Meanwhile, the focus of D’Ignazio and Klein’s (2020) article is on a specific type of statistical data errors, namely, the concerns related to data feminism. The latter is a concept asserting that there is no neutrality or objectivity in data. Instead, statistics are regarded as products of “unequal social relations,” resulting in unethical outcomes (D’Ignazio & Klein, 2020, p. 2). Although both pieces come to this conclusion in different ways, they agree on the issue that statistics “cannot prove anything with certainty” (Wheelan, 2014, pp. 107-108). I can relate to this opinion, as I have often witnessed the negative effects of people misinterpreting data or trusting statistics too much.
Indeed, all sorts of information should be taken with a pinch of salt. However, quite frequently, people do not doubt it when it comes from a source they consider reliable or scholarly. Meanwhile, as D’Ignazio and Klein (2020) demonstrate, statistical reports often reveal not only wrong but also dangerously twisted results.
It was quite surprising to me to find that a popular blog made the mistake of adding up the number of sources covering a news issue rather than the number of victims. As a result, they published the material that was very far from reality. Until the time they retracted their initial post, many readers had read and believed the article.
However, it is not even the magnitude of the error, but the article’s topic that irked me most. D’Ignazio and Klein (2020) speak of the level of patriarchy and cis-masculinity’s dominance over data capture and analysis. The authors have even coined a term for this process, “Big Dick Data” (D’Ignazio & Klein, 2020, p. 4). It is both funny and revealing, as it implies self-assuredness with which males dominate the realm of information
Wheelan’s (2014) book does not lag in humor, as reflected in the author’s personal experiences. Through numerous examples, both from his own life and from his findings, Wheelan (2014) explains how statistical data are, in fact, weak in certainty despite the prevailing opposite opinion. He notes that unless an unlikely pattern is substantiated by supplemental evidence, it is nothing more than an unlikely pattern. The writer argues that while there is no problem with the amount of data and the availability of methods for its analysis, people need to approach this process carefully and thoughtfully.
This applies to my personal worldview, as I am used to verifying facts whenever possible. At least, I never spread information I am not sure about, so as not to mislead others. Wheelan’s (2014) opinion resonates with D’Ignazio and Klein’s (2020) principle of the importance of considering context. As they assert, numbers “cannot speak for themselves” (D’Ignazio & Klein, 2020, p. 26). This quote is the core idea behind writing both pieces.
Strengths and Limitations
Both the book and the article are quite successful in presenting the facts and do not need improvement. D’Ignazio and Klein’s (2020) article contains sufficient charts, graphs, and other visuals, along with hyperlinks to references. Wheelan’s (2014) book contains few graphics, but there is no need for them, as the author’s writing is extremely vivid. His use of parentheses, italics, and other graphical devices helps the reader identify where to pause and where to look for the most interesting point in the paragraph or page. I cannot think of anything that these two readings lack in terms of exposing their respective topics.
Personal Reflection
The readings have reinforced me to be more critical of statistical reports and to always try to learn from the primary source rather than secondary data analysis. Of course, it is not always possible, yet whenever there is such an opportunity, it should be used. I have been quite cautious about such data analyses, but from now on I will be even more attentive to potential inaccuracies. With access to so much information, one needs to be careful not to allow quantity to overshadow quality.
References
D’Ignazio, C., & Klein, L. (2020). 6. The numbers don’t speak for themselves. Data Feminism.
Wheelan, C. (2014). Naked statistics: Stripping the dread from the data. W. W. Norton & Company.