Statistics on a daily basis

“Statistics: that’s a subject that you could and should use on a daily basis. It’s risk, it’s reward, it’s randomness, it’s understanding data. I think if… all American citizens knew about probability and statistics, we wouldn’t be in the economic mess we’re in today.”

-Arthur Benjamin [TED talk]

Educating for the future

“Children starting school this year will be retiring in [2070]. Nobody has a clue what the world will look like in even 5 years time, and yet we’re meant to be educating them for it. The unpredictability, I think, is extraordinary.”

 – Ken Robinson [TED talk]

Advice from Steve Jobs

I thought this was worth coming back to. From an article titled “Heart Before Head: The Legacy of Steve Jobs.”

Jobs: “Don’t let the noise of others’ opinions drown out your own inner voice. And most important, have the courage to follow your heart and intuition.”

You: What do my inner-voice and heart want most for me to do with my life?

Jobs: “Believing that the dots will connect down the road will give you the confidence to follow your heart, even when it leads you off the well-worn path, and that will make all the difference.”

You: If I trusted that eventually the dots will connect when I follow my heart, what would be my vision and what would I do next?

Jobs: “Remembering that I’ll be dead soon is the most important tool I’ve ever encountered to help me make the big choices in life.”

You: What big choices would I make if I only had a short while left to live?

More fun with high school class projects

In high school, I made a website on ancient Roman architecture as a class project in world history. That was over ten years ago. A few days ago I got an email:

I recently came across your resource website at and I found it to be extremely helpful in some personal academic research I’m doing — I just wanted to say thank you.  As a student and an educator, it is a rare treat to come across such thoughtful and concise online resources like this, especially the older ones. In my humble opinion, this is one of the timeless treasures of the Internet.

I forwarded that along to the teacher of the history class, with the subject line “History project website still useful more than ten years later…” His response:

Hi Robin.  Wow.  That is really kind of awesome.  I wouldn’t have bet money that the site was still on-line. Bravo all over again.

I guess this is why I keep things around on my website.

Data manipulation

To plan a statistical hypothesis test, specify the model you will use to test the null hypothesis and the parameter of interest. Of course, all models require assumptions, so you will need to state them and check any corresponding conditions.

When the conditions fail, you might proceed with caution, explicitly stating your concerns. Or you may need to do the analysis with and without an outlier, or on different subgroups, or after re-expressing the response variable. Or you may not be able to proceed at all.

Intro Stats (De Veaux, Velleman, Bock)

Science is not intuitive

From the trenches of science, it’s easy to label evolution deniers and global warming skeptics as dumb or at least uninformed. But we should remember that the scientific method is not an intuitive process, and takes a lot of time to learn.

For one thing, the scientific method is based on probabilities, which humans famously misunderstand. We play the lottery because we have a chance of winning, even though that chance is vanishingly small. We tend to think of very large or small probabilities as less extreme than they actually are, so when scientists say there is a 95% chance (or even a 99.9% chance) that, say, smoking causes cancer, it is easy for us to simply hear “they’re not sure.” Intuitive science is based on stories and anecdotes, rather than quantifiable statistics. We might pray and then find out that our friend returned to health. So there’s “a good chance” that the one caused the other. Isn’t that the same as scientists saying “there’s a good chance” that antibiotics were the cause of her recovery?

For another thing, it’s not immediately obvious that we need controls and placebos. If we want to find out whether an intervention works, why would we do an experiment that doesn’t actually use the intervention? Worse, isn’t it morally wrong in many cases to not give the intervention? In order to prove that an educational or healthcare strategy is effective, we have to not use the strategy in some classrooms and hospitals. Scientists are monsters!

These are just the first few tricky aspects of science that come to mind. As I re-read the Intro Stats textbook, I’m reminded how subtle a lot of this is, and how even professional scientists can sometimes get it wrong. That doesn’t mean we should be skeptical of science as a whole, though science itself relies on a healthy skepticism of individual results (another subtle distinction!). It does mean that scientists and journalists should try to explain the scientific process, over and over, day after day, when presenting results to the public. Because if you’re not doing science every day, it’s easy to forget how it works.

I hope to do what I can to help everyone do more science every day.