Source: Tables and graphs created by the author
In this tutorial, you're going to learn about line charts, which are a display that we can use their distribution for quantitative data. So we can display our quantitative data in a line chart, and the construction is very similar to a histogram. There's a separate tutorial on how to construct histograms.
So suppose that we have an elementary school class in Muncie, Indiana, say, keeping track of the high temperatures each day for the days that they're in school. In Indiana, it might get down to, say, zero degrees Fahrenheit, or highs of maybe 90 degrees at the beginning of the year in September or at the end of the year in May. And so suppose that their distribution looked something like this.
10 days of the year, the temperature was in the zeros, it was between zero and 10 degrees. It was single digits. Here, the number started with one, it was 10 to 19.9 degrees. Here, it started with two, et cetera.
The histogram for it looks like this. Between zero and 10, there were 10 days the did that. Between 10 and 20 there were 16 days that did that, et cetera.
To create a line chart, what we do is we take those heights of the bars, and instead of creating them as heights of bars, create dots instead. The heights are indicated with a dot. Then we get rid of the boxes, and the dots are connected.
This is a line graph, and it's almost the same visual display is the histogram showed. Also, like with histograms, binning makes a difference. If we bin differently than by tens, we're going to be going from this many dots, to more dots.
But the dots won't be so up high as they are right now. This is what it looks like when you bin by five degrees instead of 10 degrees. Finally, if you put the histogram and the line chart on the same set of axes, you can create something called a frequency polygon.
That's when you have the histogram and the line chart together, and you connect the tops of the midpoints-- the midpoint of the tops of each box. It's also possible to do multiple line charts on the same set of axes. This is really helpful, because it's not really all that possible to do that with a histogram.
So suppose that a school in Tucson, Arizona did the same project as the kids in Muncie, Indiana did, and then they share their information. Maybe the Arizona data looked like this. A lot more warmer days, days in the 70s, 80s, and 90s. And I suppose the '60s we're all very common, and not so much in Indiana.
The line charts, then, can be compared, and you see that on the line charts. That days in the 60s, 70s, 80s, 90s, and even 100s were very common. Whereas in Indiana, the days of our most common we're days in the 30s, and 40s, and 50s.
And so to recap, line charts are nice way to visualize quantitative data. It's much the same construction as a histogram, but the heights are determined by dots and set of boxes. The best use of a line chart is when you can compare a couple of different line charts on the same set of axes.
So the terms we used were line chart and frequency polygon, which is when it's used in the same location as a histogram. Good luck and we'll see you next time.
Source: STOCK GRAPH CREATED BY JOSEPH G.; WEATHER GRAPHS FROM GOVERNMENT DATA NATIONAL WEATHER SERVICE
In this tutorial, you're going to learn about time series diagrams. Now, time series diagrams are among the most common graphs that you're going to see in everyday life. You see them almost every day in the stock market, in the business section of the newspaper. This is a graph of the price of a stock over the course of several months-- October, November, December, January.
It looks like this stock price has been going up and down, but then during the first couple months of 2012, started getting high again. You've seen this on multiple graphs-- newspaper, online. They're all over the place. And all it shows is the different value that a variable takes over the course of time.
And it allows you to see certain values that you might not be able to see in some other kind of graphical display. For instance, take a look at this histogram of high temperatures over the course of the year. There are 365 different values here, binned by tens for Chanhassen, Minnesota. This is from the National Weather Service.
And this is that same data in a time series. What you can see is you can see all the same information that you would see on a histogram, but you can see it over time. So for instance, the histogram shows you that the most common temperature throughout the year in Chanhassen, Minnesota was in the 70s-- between 70 and 80 degrees. Happened about 80 times.
You'd have to look a little bit more closely, but you would see that that's also the case here on the time series. You'd need to look horizontally between 60 and 70 on this axis-- the vertical axis-- and make horizontal lines over to see how many points were within that band. And you would see that there were lots and lots of points within that band.
Additionally, this shows how the temperature tends to change over the course of the year. It gets high in the middle of the year-- unsurprising, because it's summer-- and then gets low at both the beginning and the end of the year, bottoming out in early January. This additional trend is something that you don't see on the histogram, and it's some nice additional information to have.
So to recap, a time series will show the change of a value over time. And if that's something that you're interested in, this is a good choice to use. Time series are not only useful in telling you what the data values are, but also when they occurred. They can show you some additional information that other plots might not. Vocab we used in this lesson-- time series. Good luck, and we'll see you next time.