Data Science – what does it look like in the classroom?

Data Scientist, the sexy new word for a statistician with programming skills?

Data scientists are highly sort after in industry and even though it may sound like some sort of a mystery science, data science is really just good old statistics with a touch of programming.  Most data scientists use some Python coding, however for students in the classroom, using an Excel spreadsheet is a sufficient enough introduction.  If you are an educator, especially a teacher of primary or secondary mathematics and your not quite sure what a data scientist is, I suggest you watch Data Science in 60 seconds.

So how do we get our students analysing big data in schools?

CENSUS AT SCHOOLS

The Government Statistics Departments of  Australia, Canada, New Zealand, United Kingdom and the United States all have there own version of Census at School. Census at School is a free resource that allows students to download other student’s data on an Excel spreadsheet. You can find links to each of these pages at the end of this article.  On the Australian site you can download a sample of up to 200, from the U.S site I was able to download a sample of 1000 students, 200 from each country. Samples of this size are   big data for students.

If you are a user of Google, Facebook, Apple or any one of these big companies then you are aware of how valuable a resource your data is. When you tick I agree to use that new app, website, search engine or server then your personal  information has been signed over to that company. By starting this blog I have agreed that my personal details  are owned by whatever platform you are reading this on.

BIG DATA IS GOLD

In Western  Australia the Kalgoorlie Gold Rush  transformed our Australian colony. Big data is here, ready to be analysed. We just need the people to  mine it

If we are to analysis and communicate statistics well, then we should be using technology.  I have found no possible reason for continuing to draw a pie chart by hand – its like long division, why?

From slide rule to calculator, it’s time for mathematics educators to accept that big data and technology are here to stay. Teaching children how to understand, sort, analyse and communicate data is much more important.

Why draw a pie chart by hand when an infographic is the most popular way to communicate statistics today?

So how does the Western Australian  Curriculum suggest we apply the statistical investigation process to real world tasks.

  1. Clarify the problem and pose one or more questions that can be answered with data                                                            

    The wonderful thing about working as a data scientist is that your skills are so transferable. Your students can choose a problem, a question that is relevant to them. Encourage collaboration between your learning areas by asking your science and humanities colleagues what content theme they are currently doing. The Australian Census at Schools website has a vast array of age appropriate ideas and resources that are lined up to learning areas across the Australian Curriculum.  I do suggest though that if you leave it open to students to choose a problem, look at it carefully to make sure that the question they choose is appropriate for your learning intentions. For example in the Yr 7 investigation I use I want students to recognise outliers and find the mean, mode, median and range. So I ensure there is a categorical and numerical data component to their question. 

  2.  Design and implement a plan to collect or obtain appropriate data                

    Students choose a sample that represents their problem. I tell my students “the bigger the better and repeated samples helps remove bias”. Working in a team or as a whole class,  each student can collect their own sample to analyse. The data sample is downloaded using  RANDOM SAMPLER on the Census at Schools website.  It is very easy to use and you just need to tick I agree that the data will be used for the sole purpose of teaching and learning.  If the students don’t tick the questions they want in their sample they end up with a very large spreadsheet of data. Personally I love that experience, but when students first see big data in this form it is overwhelming for them. This is when you refer them back to stage one – what is their problem and then delete any information that is not relevant to that. Saving the original, just in case there is a change.

  3. Select and apply appropriate graphical or numerical techniques to analyse the data 

    Now comes the the fun part –  lets start programming. I am going to assume as teachers you have all had some experience with spreadsheets using it as a tool for recording students results. Excel is a powerful application and is a great way to introduce programming language to students.  Get your students to first sort the data, they can find averages, measures of spread, counts using the predetermined functions for example =average(B1:B200) or =countif(A3:A1000,”yes”). I can guarantee that their first graph will be dodgy as students tend to just type insert chart without much thought to the variables they use. Your role as the teacher is to assist them in choosing appropriate graphical and numerical analysis for their data set. Youtube comes in handy if you need help using Excel.

  4. Interpret the results of this analysis and relate the interpretation to the original question

    So now the analysis is complete, what does the data say? Did it answer the original questions? Encourage your students to critique their analyses and conclusions. get them to identify any potential weaknesses in the analysis and make suggestions for  improvements. Over analysis is not always best, and it is OK to say there is no conclusion from the data they analysed.

  5. Communicate findings in a systematic and concise manner.

    Students write a report or display in a PPT to communicate their findings. They don’t include all of the raw data. I have had a little play on  www.canva.com and I really love how easy it is to create a really cool infographic. A little too easy as mathematicians know that 72% of statistics are made up on the spot.  So next year when I do this investigation again with my Year Sevens I am going to STEAM it up a little more by asking the Technology and Arts Media teachers if they can help create the Infographics.

I hope that you have found this article useful. As a secondary mathematics teacher,  I use this blog to share how my school has implemented STEAM education across the already existing curriculum. Unfortunately, due to funding cuts by previous Australian Government Administration the last Census at School survey was carried out in 2014 and the two administrators removed. If you feel that this resource could be useful and relevant to you implementing STEM Education practice in your school, could you please join me in lobbying the Australian Federal Government to reinstate the program.

  Resources

Census at School OZ

Census at School Can

Census at School NZ

Census at School UK

Census at School US

John Curtin College Statistics Investigation

Author: steamdonnab

Mathematics teacher in a secondary school of the Arts. Passionate about STEAM learning and finding real ways to make it work in the classroom

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