Better questions for statistical thinking

MTBoS blogging initiative, week 3!  This week’s topic is better questions.

betterquestions.PNG

Kiri Dillon (@DillonK_Chch) and I have spent some time looking at ways to develop our students statistical insight – that is, how to encourage them to be deeper statistical thinkers.

Briefly what I mean…  Here’s a sample of weekly income by gender.

surf incomesurf income summary

Most of our students have got reasonably good at describing what they SEE in the samples, the median weekly income of these males is about $300 higher than the median weekly income of these females etc.  But what questions or prompts can we give students to encourage them to think beyond just what they are seeing?  Why might we be seeing this big difference in gender? Were we expecting this? What other variables might help explain this?  What else might be going on? …  Which might lead them to exploring things like:

surf income 2

which shows that the hours worked per week is also important.

And the best questions we have found to get the students thinking further are the simple ones…

  • Why?

  • So what?

And then just repeating them to students (you know, like your annoying five year old that just wants to know more and more and more and more…)  It works surprisingly well.

Advertisements

5 thoughts on “Better questions for statistical thinking

  1. I’m a statistics teacher and this is something I need to keep revisiting. Thank you for the reminder! And a question – what software are you using for your graphs?

    Like

    1. Its been great being part of this blogging initiative to get all those reminders of bits and pieces that we maybe haven’t thought of for a while! Great for me timing-wise, as we start our new school year next week. I used iNZight (link) to make the graphs – it also has an online version called iNZight-lite. Its really just a user-friendly interface for the stats software package R, but has been specifically designed for use in NZ schools with some really neat visualisations for randomisation and bootstrapping available. Students find it very easy to use once they have imported the data (usually an excel file will be fine, just sometimes it needs to be converted to a .csv) as its all just dragging variables down to play. The best bit… its all FREE 🙂 Have fun with it. There’s a MOOC too – Data to insight – run through FutureLearn. I’m not sure when that’s due to start up again. Have fun 🙂

      Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s