Analytics Lessons from Penises, Professors & Prohibitions

Nobody expects to read about statistics in the comedy section of Huffington Post, or see a title like: "That's Not Normal! The Statistics of Penis Size." An even bigger surprise was that this piece by Chicagoan Floyd Elliott included legitimate statistics based on research from Ansell, manufacturer of LifeStyles condoms.

Elliott, a humorist, said he believes 'guys will talk about math," as long as the math centers on a topic that interests them. He said he's still thinking about a strategy to get women talking about math.

Can we get more than a giggle from this story? We can.

It’s obvious that personal, sexual, or embarrassing topics pose special challenges. It’s less obvious that the same challenges may impact even mundane data analysis. A look at tough analytics cases can give us clues about flaws we may have overlooked in our own work.

Ansell explained some details of its research:“Four hundred one males over the age of 18 were measured, one at a time, by qualified medical staff…” at a nightclub in Cancun, Mexico. “Effective measurements” were recorded for 300 men. 101 men were, let’s say, unsuccessful.

So many questions come to mind! Were the men paid? Are there any laws regarding paying people to let you measure their genitals? Had the men been drinking in the nightclub? How would drinking affect the measurements? How would being measured affect the measurements? Where, exactly, do you measure, and how? Do you use a ruler, a tape measure…?

While the details of this project may seem unique, we can draw lessons from it that apply to all sorts of research.

Lesson 1) Know what laws apply to your work, and comply with them.

Lesson 2) The presence of an observer may affect outcomes. Outcomes also may be affected by other factors that are difficult to control when conducting research.

Lesson 3) Clearly define what information is to be collected. Provide detailed, written instructions, including diagrams if applicable. Some data collection tasks require trained professional staff.

Another research area that poses special challenges is the study of sexual behavior. This requires self-reporting, which is affected by memory, embarrassment, and ego, nor are respondents the only concern.

A professor once told me that she had conducted interviews for a major sexual behavior study. But she did not like to interview men, so when it was her turn to do so, she traded duty with a male research partner, and lied about it on the interview notes.

Lesson 4) Staff members don’t always do what you they are supposed to do. Monitor.

Even topics that may seem innocent at first can be layered with personal issues. One of the things that students often discover when exploring real data from the US General Social Survey is that, when asked how many children they have, women report having more children than men. This has been the starting point of some lively class discussions, and leads to one more lesson.

Lesson 5) Sometimes, people cannot, or will not, tell you what you want to know.

Have you learned lessons from challenging research? Please share your stories.

Meta S. Brown, Business Analytics Consultant

Meta S. Brown is a consultant, speaker, and writer who promotes the use of business analytics. A hands-on analyst who has tackled projects with up to $900 million at stake, she is a recognized expert in cutting-edge business analytics. She has conducted more than 4,000 hours of presentations about business analytics, and written guides on neural networks, quality improvement, statistical process control, and many other statistical methods. Meta's seminars have attracted thousands of attendees from across the US and Canada, from novices to professors.

Tell Me a Story: Why Data Analysts Must Be Storytellers, Too

Data alone won't make an analyst's work memorable or actionable in the eyes of a business executive. A story puts it into perspective.

It's the Data, Stupid

When it comes to acquiring the data that will feed your analytics initiative, "free" isn't always the best approach.

Great Blog!
  • 11/26/2013 4:41:48 PM

This is a great blog and who better to write it than Meta!  Doubt not her qualifications.  You know your culture is advancing when a woman can write on any topic without fear of being shot (Malala Yousafzai) or having an Ayatollah issue a fatwa (Salman Rushdie).  This blog exposes studies that do not measure up.  It lays out the usual lessons, yet I think we should place more emphasis on one larger issue--leadership.  Corporations are less proficient at specialized tasks because the team manager usually does not understand what the team does or can do. 

Re: Reporting Bias
  • 11/25/2013 4:51:32 PM

Hi Michael,

I agree. Those weren't my issues though.

Re: Reporting Bias
  • 11/25/2013 4:38:15 PM

We are absolutely free, within the constraints of civil discourse, to voice our opinions. However, I feel that emphasizing the vehicle of the salient points, rather than the salient points themselves, is only worthwhile if it's used to humorous effect and in furtherance of the conversation.  

Re: Reporting Bias
  • 11/25/2013 4:34:33 PM

Hi Michael, why would you redact yourself? I thought we were all free to voice our opinions here.

Re: Alternate headlines
  • 11/25/2013 4:30:46 PM

I seem to have misconstrued your meaning, then, Heurist. Regardless, I don't think it's productive to belabor the point. 

Re: Discussion
  • 11/25/2013 4:30:38 PM

Whoah Michael, now that's stretching it. How do my comments compare with trolling?

Re: Alternate headlines
  • 11/25/2013 4:27:40 PM

Hi Michael,

Thank for the great picture. I had something else in mind. That sure does make good 'art' though. I wasn't talking about words by the way, I was talking about something completely different.

If you have more examples to bring to bear, I'm sure I will find them interesting.

Re: Reporting Bias
  • 11/25/2013 4:08:11 PM

See, I agree with that, kq4ym, but I also think that there's some psychological benefit to posting the results without judgement or editorializing. There's an interesting article about this on this week, but I won't summarize its findings here for three reasons:

1. It's not germane to analytics, per se.

2. The source is notoriously liberal, politically.

3. I'd be accused of advancing the erosion of popular culture.

Re: Reporting Bias
  • 11/25/2013 4:03:45 PM

I think you're painting the Huffington Posts of the world as advancing an agenda that runs counter to your culture. Which culture is that?

Re: Discussion
  • 11/25/2013 4:01:59 PM

Nature of the beasts. Or trolls, in this case. But your rebuttal stands as a great example of patient and reasoned argumentation.

Page 1 / 5   >   >>