Statistics Prove Who is Dumber than a Rock

Rocks in Santa Monica mountains




Anyone who has ever taught has probably had this experience … you are reading a student paper and think,

“I really don’t understand what he/ she is trying to say here.”

Like most professors, you probably assume that in these situations, the students aren’t very clear on the concepts themselves. You’re probably right on that, too. If you spent enough years in school you probably caught yourself at least once facing  a question you had know idea what it was about so you tried to just write down whatever you did know and hope the professor was kind enough to find something in there somewhere to give you points. Sometimes this works.

Odd thing I’ve found though is that when it is the professor him/herself who is saying something that is not understood, again it is assumed that it is the students who do not really understand it.  I would say the assumption is questionable.

When you’re presenting, whether it be giving a lecture to a class, a conference presentation or a meeting, and the majority of your audience is confused, there are a few possibilities:

  1. You don’t understand the material.
  2. You don’t understand how to communicate your knowledge.
  3. You did not spend enough time preparing for your presentation.
  4. Your audience is not very bright.
  5. Your audience did not spend enough time preparing for your presentation.

Out of those possibilities, 60% are you. One of my pet peeves that I see at every point from high school classrooms to prestigious conferences is people who seem to have a disrespect for their audience. I often hear colleagues say that they don’t practice their presentations,

“I do better when I’m more spontaneous, otherwise it just sounds like I’m reading my notes.”

Or some bullshit like that.  No. You don’t. It is damn near impossible to always accurately estimate how long it will take you to give a talk. If you are teaching a class, that isn’t a big problem. You can pick up where you left off the next time.  In a meeting or conference, where people had to be scheduled to come together, juggle schedules, that isn’t always possible.  Sometimes, when your audience doesn’t understand it was because you had to leave out a lot of information they needed because you were rushed. Sometimes you assumed they had pre-requisite information and they didn’t.

How to fix that: I put my powerpoint up before every class. Students can read it ahead of time. I also often put up supplemental materials so that those students who have trouble can go read that prerequisite information. Conferences often have an option to post papers after the conference that go into more detail. I email my powerpoint to anyone who asks. I put them up on our company website when I have time and I end many presentations with references for additional information. I actually read my presentation out loud before a meeting and time it. Then I allow for a few minutes for introductions, people to get settled in, etc.  For great tips on doing good presentations,  I highly recommend the book, Rockstar Presentations.
Then there are the first two problems, where I have noted a recent, disturbing trend.  I’ve been a statistical consultant for nearly thirty years and I have had very, very few clients who were students who wanted help with their course work or dissertation. Our rates are not cheap – $100 an hour if I don’t have to leave my house. If I have to put on a suit, go anywhere further than my office downstairs or get up before 10 a.m., the rates go up sharply.

For consulting work, it is a good deal but it is not in the budget of most students. Yet, I am seeing more and more who are willing to pay it.  When I meet with the students and review the material from their courses, their assignments and the feedback they are receiving from their professors, I see why.
First, there are the professors who don’t really know all that much. I attended a university that required a minimum of four courses in statistics or research methods for every doctoral student. Three of those four courses also included a three-hour weekly computer lab.  Students who specialized in statistics took another five or six courses. One of those had a computer lab. After that, you were expected to do it on your own. Many of the programs producing Ph.D.s now require one, or at most, two, courses and no computer labs. The students might go to the lab two or three times during their program. That’s it.  Students receiving Ph.D.’s in statistics and mathematics tend to focus much more on theory than programming and very, very little on application to real problems.
These professors may know a lot about a lot of things. Unfortunately, one of those things is not what they are teaching, how to apply statistics, both conceptually and using statistical software, to conduct research. How do they publish? Most of them don’t. Some publish articles on topics that require little or no statistics, publish terrible articles in lower-tier journals or pay people like me to do the statistical analysis section for them. Just like the students, they have a hard time articulating ideas in statistics because they really don’t understand very well themselves what a planned orthogonal contrast, dummy variable or logit actually is.
Second are the professors who don’t know how to communicate the information. Sometimes it is because English is their second language. This has been a problem since I was a student and I frankly do not understand it. If you have a job where part of the requirement is to communicate information in English, why do you hire people who cannot speak English fluently? Why don’t the students object? Actually, they’ve been objecting since I was an undergraduate and no administration ever seems to give a damn. I don’t get it. There are also professors that English is the only language they speak and they just don’t communicate in it very well. Sometimes I think this problem is just plain arrogance, they don’t think it is worth their time to bother with communicating well. This has been explained to me as everything from that the university puts much more emphasis on research (often true) to the students just aren’t “college material” and should drop out or pick any easier major where said professor doesn’t have to be bothered with them.
While the students are certainly pleasant and interesting to work with, and I don’t mind the additional business, I remain puzzled by the whole picture.
Why do students put up with paying $40,000 a year for an education that leaves them needing to pay me $100 an hour to explain what their professor said in class?  Why do universities hire people who don’t really know their subject very well or who are not fluent speakers of the language they are supposed to be using to teach? Why don’t more universities go to the model of hiring people who can and want to teach as clinical or teaching professors and hiring the people who don’t want to be bothered with teaching as researchers?
These aren’t rhetorical questions. I don’t have the answers and I really am interested, so if anyone knows, please enlighten me.

Similar Posts


  1. In some ways this post made me laugh out loud. I BY FAR took the most statistics courses of anyone I in my PhD program; 5 (intro, regression, anova, multivariate, hierarchical), plus qualitative and tests and measurements. I am regularly called an expert. Yet what I mostly know is what I DON’T know.

    (I sat at my desk friday banging my head into a digital wall when my (very large n) logistic regression model had an ok r-squared (.22) but a lousy fit. No one ever told me what to do next, and no one I know knows enough more than I do to help me figure it out.)

    My point here is that taking a pile of coursework doesn’t add up to squat without the 20 years of experience OR without a mentor you can turn to when you are stuck. All I have going for me right now is stubbornness that says I can GET to the point of having that experience and being the expert people think I am. Without that you can force people to take as many courses as you want but they will have the same questions.

  2. I worked as a SAS developer at an Ivy League university for seven years. The prevailing attitude regarding many of the points that you brought up was that there was more to learn than could possibly be taught in a classroom and that those students that would be the most successful would take it upon themselves to fill in their own gaps. Resources were available in the form of a data library and a statistics helpdesk as well as teaching assistants, study groups and of course the professor. Perhaps to some degree it depends on the specific environments and the talent those environments attract. Certainly I learned those things throughout my academic career and later, when I became a teaching assistant, I gave the guidance on the points that you mentioned to the degree that students chose to engage me.

Leave a Reply

Your email address will not be published. Required fields are marked *