Michael J. Swart

This post is a story about my daughter’s school project with a plug for Wolfram Alpha and maybe Excel.

My daughter came to me the other day and asked me to help her with some homework… She said she needed data. She was learning about plotting graphs in math class and she was asked to bring in some data that she could plot. Learning, Math, Data. I felt pretty proud (Just as any parent might when kids take an interest in their profession.)

Personally, I thought this was great! My kid is asking me about data! At the very least, I could explain to her a little better what it is that I do at work. At best maybe she could “catch” what that I like about the topic.

The project itself was pretty straightforward. The students were asked to bring in a list of numbers that they could use in class to create a chart. The data could be about anything and there were extra marks for plotting two series instead of one.

What We Did

So I asked my daughter what she wanted to bring… it could be anything she was curious about. Eventually we:

• We started at Wolfram Alpha
• And we picked the local temperature for 2012.
• For the extra series, I picked temperature for a location that I thought would contrast well with the local temperature. That location was Sydney Australia.
• This led me to search Wolfram Alpha for 2012 temperature for Kitchener, Canada and Sydney, Australia
• But we needed the data points, this led me to use the most beautiful and useful feature that Wolfram Alpha provides, Data Download:


Pretty much the only feature analysts need

• It was the work of a minute (using Excel) to summarize the average monthly temperatures for both cities. I got the data into the right shape, created a pivot table, and adjusted some of the filters.

After everything was done, the chart itself was interesting. As I expected, the average temperature (here) reaches its maximum in July and we saw the opposite trend in Australia. One surprising thing was that the temperatures varies very little in Sydney. It looks like 10°C (50 °F) is a very cold day in Australia. So the numbers confirmed what we guessed and showed us something new… Not bad.

Keeping away from SQL Server (For Now)

Not once did I even mention or use SQL Server or tables or databases. And I didn’t dwell on the Excel features. I just explained what we were looking at. In this case, I thought that it was better to show what the data tools could do before showing how they do it.

Missed Opportunitiy?

But I feel like I must have missed out on something here. Don’t you think so? Tracking temperature is kind of boring…  I felt like a carpenter who was trying to show off what could be done with a router but ending up with a birdhouse. What kind of data would you have picked? I tried to think of something that’s actually interesting or useful but came up empty and went with average temperature. If you have any better ideas, let me know.