The use of statistics in the media, and the responsibility of the publication of statistics from research and evaluation

Introduction

Newspapers, television, and other media outlets are accessible to the public, including laypersons, who oftentimes do not question the numbers in reports, articles, and other publications. The public assumes that researchers and the media outlets would have conducted some due diligence to ensure that the information it is passing across is accurate. It is important that researchers and evaluators conduct their research using methods and designs that ensure that their findings are credible, so that they do not intentionally or unintentionally mislead the public.

Why it is important to share statistics with journalists and the public

Making statistics available to journalists and the public is often expected, especially for research that has been publicly funded. The public indirectly paid for this research and should be able to access the findings. However, the statistics shared should be clear and understandable. The research design used and the justifications should also be provided in clear terms that a non-technical audience can understand.

In addition, researchers should not only share statistics, but they might also need to share their interpretation of the statistic and alternate interpretations possible. Showing where the information has come from, who produced and published the data, and their reasons for conducting the research also helps to clear doubts about any foul play (The Open University, 2005). Because society relies on informed judgment supported by these statistics, researchers have to work in a “professional, competent, respectful, and ethical manner” (AMSTAT, 2018) and show that their research is independent and credible.

Ways the media often use and misuse statistics

The media could use statistics wrongly by reporting data from a study with a flawed methodology. The media could also not mention the details of their methodology, hoping no one would be inquisitive to ask, so they are not subject to any scrutiny. Hiding details of the research design could also lead the public to assume that the methodology should be fine since the findings are publicly published. The media could also only cherry-pick the findings that tally with or corroborate the views they want to propagate.

Does the information in the articles (see references below) influence the way I think about my work in evaluation

It is also vital that I commit to “good statistical practice which includes having “transparent assumptions,” “reproducible results,” and “valid interpretations” (AMSTAT, 2018). Also, that I maintain professionalism by advancing knowledge while avoiding harm and not use statistics to advance unethical ends. It is crucial that I identify and avoid situations that constitute a conflict of interest and likely to predetermine or influence the results of my research (AMSTAT, 2018).

It is important that my research has adequate rigor and is not conducted with leading or misleading findings. For example, the research referenced in The Open Learning article with the skewed report that 70% of persons had tried to quit smoking, but no one succeeded. They did not state that the research was conducted among current smokers and not a random selection from the population (The Open University, 2005). It is also important that I employ sampling methods appropriate and valid for my research question (AMSTAT, 2018), interview enough people needed for the particular research to be statistically valid, ask the appropriate target groups and word questions appropriately (The Open University, 2005). The research should also be conducted such that results are generalizable under “reasonable assumptions” and with “minimal error” (AMSTAT, 2018).

References

American Statistical Association (AMSTAT) Board of Directors. (2018, April 18). Ethical Guidelines for Statistical Practice. Retrieved from https://www.amstat.org//asa/files/pdfs/EthicalGuidelines.pdf

The Open University. (2005, January 10). Statistics and the media. Retrieved March 05, 2021, from https://www.open.edu/openlearn/science-maths-technology/mathematics-and-statistics/statistics/statistics-and-the-media