Skip to main content
Social Sci LibreTexts

14.7: Using Statistical Data

  • Page ID
    13897
  • \( \newcommand{\vecs}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \) \( \newcommand{\vecd}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash {#1}}} \)\(\newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\) \( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\) \( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\) \( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\) \( \newcommand{\Span}{\mathrm{span}}\) \(\newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\) \( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\) \( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\) \( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\) \( \newcommand{\Span}{\mathrm{span}}\)\(\newcommand{\AA}{\unicode[.8,0]{x212B}}\)

    We have no intention to turn this lesson into a mini-seminar on statistics, but you need to apply some fundamental skills when making decisions about using numbers, data and statistics in your messages.

    There are some basic types of statistical interpretation problems to look out for:

    Everything is going up
    More people are employed, more people are getting certain diseases, etc.. This type of statistic is usually right, but also meaningless, because the total number of people is also going up all the time. A more useful statistic is the RATE or PROPORTION of the population that is affected by something or doing something. A front page news report with the headline “Playing in pain” trumpeted the alarming “statistic” that more than 3.5 million children under the age of 15 required medical treatment for sports injuries, and said that total had quadrupled since 1995 when it was 775,000. But the article failed to point out that the overall number of children under the age of 15 participating in sports had also increased over that same time period. A much more accurate way to report the injury statistic would have been to compare the rate of injuries per 1,000 children under the age of 15 playing sports across the two time periods. Using absolute numbers in comparisons across time is almost never an accurate way to interpret a statistic.

    Best foot forward
    Statistics that use the number that best supports a case. This is the figure that is chosen by the person making the argument. Someone may choose the worst year of a recession and compare current economic conditions to that worst year in order to make the current climate look good. Or someone may choose to use the mean family income figure to distort what’s typical when the median figure might more accurately reflect whether incomes have risen or fallen.

    “Gee-whiz” or half-truth

    Use of statistics where the number tells just part of the story. If the overall unemployment figure is not dramatic enough to make the point, then the person providing the number focuses on the unemployment rate for teens, or for specific industrial states, or some other smaller or unrepresentative group. Groups that are trying to influence public debate are notorious for these types of “half-truth” statistics.

    Anecdote
    The one-in-a-million exception that supposedly proves the rule. It is always crucial for you to ask how many instances or people are represented by the statistic.

    Everyone is average
    Use of statistics that tend to characterize individuals by group characteristics. For example, many argue that women can’t be soldiers or firefighters because the average man can lift more weight than the average woman. But people are individuals, not averages, and many women can and do lift more than many men.

    Coincidence
    When is something actually related to something else, and when is it just a coincidence? The corollary is that just because two things are correlated doesn’t mean that one thing CAUSED the other. For instance, a study may invite us to infer a causal connection with the following title: “Bottled Water Linked to Healthier Babies.” Without further investigation, this study should be rejected. Why? Affluent parents are more likely both to drink bottled water and to have healthy children; they have the stability and wherewithal to offer good food, clothing, shelter, health care and amenities. Families that own cappuccino makers are more likely to have healthy babies for the same reason, but we wouldn’t give a second thought to a “study” that was titled “Cappuccino Linked to Healthier Babies.”

    Suspect calculation
    Determining just how the statistics are arrived at requires careful attention. Playing with the numbers can get a company into trouble as seen in this case between Proctor & Gamble and Pfizer. P&G claimed in their advertising that 4 out of 5 dentists recommended Crest oral rinse to their patients. As reported on the Trademark Blog, March 7, 2006, “Proctor & Gamble sent the Crest product to 344 dentists who were asked to use the product for one week. The dentists were paid $75 to participate in a survey. 269 dentists participated in a phone survey where they were asked ‘Based on your experience using this oral rinse, which of the following statements best describes your most likely recommendation of this oral rinse to your patients?’ According to the complaint, P&G arrived at the 4 out of 5 number by combining those who responded that they ‘would recommend’ the product with those who responded that they ‘would recommend only if their patients asked about it.’ Pfizer alleges that this hypothetical recommendation does not constitute proper substantiation that health professionals recommend the product in their actual practice.”

    In sum, communicators are confronted with much information that is based on numbers, statistics, and numerical claims. It is no longer acceptable to simply rely on a source’s interpretation of those numbers. You have to be able to independently evaluate and critique this material, on deadline, and with confidence. It is a skill that is as basic as knowing how to ask a good interview question or run a tape recorder. Take whatever steps you require to get your numbers skills in shape, and keep them sharp.


    14.7: Using Statistical Data is shared under a CC BY-NC-SA license and was authored, remixed, and/or curated by LibreTexts.

    • Was this article helpful?