Representation of Numerical Variables: Histograms

00:01 The most common graph used to represent numerical data is the histogram.

00:05 First, we'll learn how to create it, and then we'll provide a description of the way the data is represented.

00:11 We are going to use the frequency distribution table from our previous example to help us out.

00:16 Here it is. As you can see, it looks like a bar chart, but actually conveys very different information.

00:25 As in the bar chart, the vertical axis is of numerical type and shows the absolute frequency. This time, though, the horizontal axis is numerical to.

00:35 All right, so each bar has width equal to the interval and height equal to the frequency. Notice how the different bars are touching.

00:44 This is the show that there is continuation between the intervals.

00:48 Each interval ends where the next one starts.

00:51 In the bar chart, different bars represented different categories.

00:54 So the bars were completely separate.

00:58 Okay. Sometimes it is useful to plot the intervals against the relative rather than the absolute frequency.

01:05 As you can see, the histogram looks the same visually but gives different information to the audience. Remember, relative frequency is made up of percentages.

01:14 There is no way to do that in Excel but is a useful piece of information.

01:20 There is one last thing to note here.

01:21 We may create a histogram with unequal intervals.

01:25 An example is designing age groups.

01:28 You've likely completed some survey where you were asked about your age and the possible answers were 18 to 25, then 26 to 33, 31 to 35 and so on until 60 plus.

01:40 Clearly the interval widths vary and reflect different focus groups for the experiment at hand. An explanation for the choice may be young adults under 25 cannot afford the product, while adults over 60 have no interest in the product.

01:54 In any case, you should be quite experienced to accurately design and interpret such groups. It is highly recommended that you stick with the equal with intervals until you gain enough experience.

02:05 Okay, great.

02:08 This is how we can build a histogram in order to represent numerical data.

02:13 Thanks for watching.