I'm going to be giving a five minute "elevator pitch" presentation to undergraduate math majors in order to advertise an introduction to statistics class that I'll be teaching in the spring. It's an upper division course and the students will have already taken an upper division probability course. Here are a few objectives for the elevator pitch:

  • Briefly explain what statistics is. I'd like to give a quick, big picture summary of what statistics is that captures the essence of the subject.
  • Explain, in a compelling way, why statistics is so useful.
  • Explain what makes statistics beautiful or exciting. I'd like to mention some "gems" of statistics that could be covered or touched on in the course.

The presentation should also just give an overview of the specific topics that will be covered in the course, ideally in a way that sounds interesting. I have some ideas about what to say but I'd like to get your suggestions for how to make a statistics class sound exciting and interesting in a five minute presentation.

  • 1
    $\begingroup$ Emphasize the commercial importance of statistics. It is at the heart of new drug development and oil exploration. And many other industries. It also comes into play in many social activities (e.g. education). I would also mention that you will cover a lot of formulas and such (and be honest that it can become confusing) but that part of the reason to get through them is so you develop statistical "insights" and intuition. And that it makes looking up formulas or working problems easier in the future for having done them once. $\endgroup$
    – guest
    Commented Oct 4, 2018 at 22:18
  • $\begingroup$ Climate science (long term) and meteorology (short term) also use a lot of statistics. $\endgroup$
    – guest
    Commented Oct 4, 2018 at 22:33
  • $\begingroup$ @guest Why not write that as an answer? $\endgroup$
    – Tommi
    Commented Oct 5, 2018 at 9:07

3 Answers 3


I suggest you tie the "undergraduate statistics class" to data science.

(1) For example, Target’s pregnancy prediction algorithm: How do you explain data science to non computer science people?

(2) Or data analytics success stories.

(3) Machine learning vs. Statistics: "Machine learning may emphasize prediction, and statistics may focus more on estimation and inference, but both focus on using mathematical techniques to answer questions."


Here is the course description I wrote for a (high school) class on Statistics and Probability. I think you could modify it ever so slightly if you are covering Statistics, specifically, and that it could serve as a real elevator pitch.

What do the words probability and statistics mean? How are probability and statistics used or not used, correctly or incorrectly, in research journals, popular media (newspapers, television), and social media (blogs, Twitter, Facebook)?

How is it possible that the same areas of mathematics can be applied to meteorology (e.g., forecasting), sports (e.g., oddsmaking), and elections (e.g., polling)?

In this class, we will investigate topics of contemporary interest, and position ourselves better to be analytical and skeptical readers by using statistical and probabilistic tools to inform our critical consumption of information and data.

I think that the second paragraph above is something that can be especially exciting, as many know this to be true about statistics, but (for whatever reason) may not have paused to marvel at the wondrous nature of analytical tools that can be applied to such disparate areas of the world.

I also like to give one real world example about causation versus correlation, which (as far as I know/as best I remember) is hypothetical - but may have appeared somewhere else. (I'd appreciate a source if you know one!)

Sensible Gum Laws: Consider the question of whether chewing gum leads to cancer. Specifically, whether frequently chewing gum makes one more likely to suffer from mouth cancers later in life. Looking through the data, one may observe that, indeed, those who are chewing gum with a greater frequency are also being diagnosed with mouth cancers later on at a greater frequency. From here, one can argue that chewing gum is a bad idea, or look to pass laws about the ingredients in gum, and so on and so forth. But, a deeper look at the data can reveal that those who frequently chew gum are also much more likely to be cigarette smokers, who are looking to mask the smell of smoke throughout the day. And smoking cigarettes definitely causes mouth cancers.

I am deeply concerned that this gum example (with its pithy title!) is but one of many common conflations of causation and correlation. Considering the frequency with which articles are retracted after faulty statistical methods are discovered, and some questionable practices within certain disciplines around the use of statistics (I am thinking of p-hacking, for example) it is very important that we work towards a level of quantitative literacy that will enable us to be better consumers of information.


You can give the example of election polling.

Many popular explanations of election polling give a false impression of what statisticians can tell us about an upcoming election. For example, the so-called "probability clocks" on the front pages of major newspapers during the 2016 election gave the false impression that what statisticians do is to estimate probability (leading to false confidence of newpaper readers).

Instead, of course, what statisticians do is to compute the confidence of a prediction, based on polling data. You can explain this very clearly in a statistics course.

  • $\begingroup$ Thank you, great example. Feel free to mention more examples like this. $\endgroup$ Commented Oct 7, 2018 at 12:03
  • $\begingroup$ I would beware of too subtle distinctions of Bayesian confidence versus CI versus p values when you are doing a 5 minute sales pitch for the overall topic. $\endgroup$
    – guest
    Commented Oct 17, 2018 at 15:21
  • $\begingroup$ That's actually my central point. The 5 minute sales pitch which invites confusion between probability values and confidence values is very distorted from reality. So distorted as to be deceptive, sometimes in a really damaging way. $\endgroup$
    – Lee Mosher
    Commented Oct 17, 2018 at 16:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.