Manipulative Statistics

Let me start with a quote from a web page for a reputable skin care brand:

“How does Olay Professional work?
Years of genomic research were focused on developing a comprehensive understanding of the gene activities which affect aging. A comparison of the gene activities in younger and older skin showed that skin aging is multi-factorial, and clearly suggested that a comprehensive approach is needed to truly affect the skin aging process. This genomics understanding could then inspire a range of formulations for Olay Professional; using ingredients trusted by world-leading skin experts and designed to help improve the appearance of the skin. The result of this breakthrough collaboration is a potent range of products” (Olay Professional, 2011)

Their explanation of how this product has been tested, by this statement, seems a little ambiguous. That’s not to say that they have not performed any sort of product research, but would you not want to advertise these incredible results that show your product working. However, because it sounds Scientific, we may be inclined to believe their claims, or just overlook the lack of supporting evidence. – Who knows what ‘Genomic Research’ is anyway? Or that ageing is ‘multi-factorial’? This confusion tactic is employed by many companies to help us to trust them in what they do; if it sounds trustworthy then it must be. This same tactic is often employed when releasing research statistics; particularly in quantitative research and correlative data. Personally my favourite example would be the infamous 0.1% of bacteria which cannot be killed by any soap on the market. However if I were to use a better example I would say: If a researcher were to be employed by a popular high street retailer to conduct a survey to show what percentage of the population chose to shop in their stores, they would most likely use a survey, or questionnaire method of collecting the results. For their company to get the best results they would chose to take this survey in the best location, maybe on the same street as which their store is located, or from outside their shop front. They would also tend towards general questions, such as ‘Have you ever shopped at…?’ This isn’t to say that they regularly shop there, will return or even enjoyed the experience, but for the goal of this survey this doesn’t matter. If they then questioned 100 people, to whom 80 answered ‘Yes’, then this would allow them to report that  80% of people they asked shop at their store. Now considering there are around 61 Million people living in the UK, (Rogers, 2010) that is a lot of people who shop with this retail chain; somewhat questionable as this information would have been gained from such a small sample.

However it’s not just large companies and high street retailers that use this method, Psychologists and other Scientists will all use this method of presenting their findings to push their thoughts. Maybe even unknowingly (Experimenter Bias), after all, everyone wants to be right the first time. A few statistically proven statements I found around the internet are: Homemakers are happier in their marriages than career women (Wilcox & Nock, 2006); Women are not advancing in science and technology fields because their brains aren’t wired for it (Nettle, 2007); And Neanderthals died off because men and women played equal roles as hunters (Premo, 2006). Each of these statements has their own research and findings to prove what they are claiming, but how reliable is it?

While they do offer some good research background in their papers, these hypotheses by themselves appear to be very drastic statements. So consider my final points when looking at research and advertisements; how controversial is the statement? Who is writing it? Is this person well known, or even who they say they are? What qualifies them to make such a statement? Are they willing to show you evidence of the research behind the statement? Does the research cover a good sample? I guess in short, take nothing for granted and trust no results that appear invalid or unreliable.

Before you go: A footnote…
I contacted the company behind my first quote to request information on the research to back up their statements. I received a reply shortly after explaining:

“…I am sorry to inform you that we are unable to help as we have no information relating to your query.
However, I can suggest that you revisit us at for information about P&G brands…” Make of that what you will, but please note, I am not accusing the company of misrepresentation.


Nettle, D. (2007). Empathizing and systemizing: What are they, and what do they contribute to our understanding of psychological sex differences? British Journal of Psychology, 237-255.

Olay Professional. (2011). Science of Olay Professional. Retrieved 03 09, 2012, from Olay Professional:

Premo, L. S. (2006). Patchiness and Prosociality: Modeling the Evolution and Archeology or Plio-Pleistocene Hominin Food Sharing. USA: The University of Arizona.

Rogers, S. (2010, 09 21). Data Blog. Retrieved 03 09, 2012, from The Guardian:

Wilcox, B. W., & Nock, L. S. (2006). What’s Love Got To Do With It? Equality, Equity,Commitment and Women’s Marital Quality. Virginia, US: The University of Virginia.


About DJPsych

... And now I find my self in my first year at Bangor University (Wales, United Kingdom), studying Psychology with Neuropsychology With the long term goal of gaining my Doctorate. In the shorter term however, I intend on sharing with you all, my thoughts and opinions across a wide variety of subjects within the vast realm that is 'Psychology'. However, if you want to hear (Or indeed read), my opinion on wether Psychology can be called a 'Science', then it would probably be best to email me (: View all posts by DJPsych

You must be logged in to post a comment.

%d bloggers like this: