Who studies math on weekends?

In a recent IntMath Poll, I asked readers about their weekend math habits. The question was:

During weekends, how many hours do you spend doing math homework (or study)?

I asked this question because of the consistent pattern that I've observed in the traffic to my site, Interactive Mathematics. The majority of visitors (probably 90%) visit the site when they are looking for an answer to a specific math question. They visit maybe 1 or 2 pages, find the answer to their question, then leave. In other words, they are doing math homework or study. (Common search queries are "how to draw tan graph", "z table", "derivative of sec".) Only a minority of visitors start reading at the beginning of a chapter and work their way through each page in that chapter.

Either way, it is safe to say that anyone visiting the site is either "doing math homework (or study)" or perhaps is a teacher preparing a lesson.

Out of 2100 people who responded to the poll, 32% said "4 or more hours" of weekend math study and 22% said "None" while the remainder said 1, 2, or 3 hours, in that order.

Now to the evidence (which more likely indicates what people really do, rather than what they say they do). Here is the IntMath visitor traffic information provided by Google Analytics for the period August to October 2009. Each vertical gray line is a Monday.

The traffic during August is relatively low because that's the summer holiday period in the Northern hemisphere. By September school has started and there is a lot more activity.

There is a considerable drop in traffic every weekend, indicating a lack of interest in math. Most of the time it is less than half of the traffic during the week. So it was a surprise that the most common answer by poll respondents was "4 or more hours" for weekend math study. I expected it to be "none".

Let's think about this some more. The type of visitor is different on weekends. They tend to look at more pages (during the weekend it tends to be 25% more pages per visitor) and they stay longer on each page (about 20% longer on each page). The graphs for "number of pages" and "time on page" tend to be the opposite of the traffic graph, peaking on weekends.

So while there are less people doing homework on weekends than during the week, the ones who do it on weekends are more serious about it. No real surprises there.

Another issue is that "weekend" is different in different countries. Mostly it means Saturday and Sunday, but in some countries Friday and Saturday are days off.

Also, it is worth considering the people who will actually respond to the poll question. They are more likely to see the poll if they spend quite a bit of time on weekends looking through the site.

So who studies math on weekends? We can conclude that it is those who are especially keen (or in a state of panic 🙂 ), and they represent less than half of the "normal" population. They are also more likely to do well in math!

Trends over time: People Searching for "math"

Let's now look more broadly at the trends in searches for the term "math" to see what we can find out. Google Trends for the term "math" give us some insights into when people are interested in math, and when they are not. Here is the chart for Google searches for "math" for the period Jan 2005 to Nov 2009.

The top trend line is "search volume index". This is scaled so that the average for the period is "1". When the blue curve is above 1, it means there is heightened interest compared to normal, and below 1 means people are less interested in that topic.

The second line indicates "news reference volume", which means the number of times the word "math" appears in the (online) news. The letters (A, B, etc) on the top graph refer to news items. The first 2 refer to:

A. Gains in math, mixed results in reading [CNN International - Oct 19 2005]

B. Bush says math, science economic tools [Akron Beacon Journal - Feb 1 2006]

The graphs show us that in June, July and August each year, math takes a holiday, and also that very few people are interested in algebra or logarithms on Christmas Day. Fair enough! There is a dip also for Thanksgiving in the US, which falls late November.

You can see more on how they calculate Google Trends. There's some interesting statistics involved. As they say, Google Trends is still a work in progress and subject to sampling bias, but it sure is interesting. At the bottom of each Google Trends page it has:

Google Trends provides insights into broad search patterns. Please keep in mind that several approximations are used when computing these results.

Here's another math-related trend graph. Pi Day occurs each year on March 14th (3.14 are the first 3 digits in the decimal representation of pi). Many schools have Pi Day activities and I imagine students need to do assignments on the topic, hence the surge in search.

As you know, North Americans use the term "math", whereas the English, Australians and the Indians use "maths" to refer to "mathematics". In Google Trends you can compare terms (by separating them with a comma in the search box). Here is the result when comparing math, maths, mathematics. It seems we are all getting lazier about typing the full "mathematics". Are we seeing the slow death of the longer word?

More Google Trends

Check out the graphs for these interesting topics on Google Trends:

• Michael Jackson (and see at the bottom that this was a huge story in Indonesia!)
• Iran
• Climate change (which shows that Australians regard this as a very important topic, compared to other people in the world)

Spotting patterns and trends is an important aspect of "mathematical thinking". Analyzing such data and making recommendations for action is what (some) mathematicians do.

I'd be interested in your responses, especially about weekend math study.

Be the first to comment below.