From the frequency distribution (Output 4.1) we know that most people reported having either no sexual partners or one sexual partner last year. Regardless, this histogram does summarise the information in Figure 4.1 quite well. For this reason, histograms are best used with data where nonintegers are actually possible. Now, this might seem somewhat silly given that number of sexual partners must be an integer (i.e., a discrete variable). So, for example, you can see that 16 respondents in the data set reported having between 3.75 and 6.25 sexual partners last year (i.e., 4, 5, or 6). The "Y" axis (also called the "ordinate") displays the frequency or number of times a particular piece of data in the data set falls into that interval.
So by convention, any score in the data set equal to or greater than 3.75 and LESS THAN 6.25 gets assigned to the "5" bar. The upper real limit is between 5 and 7.5, or 6.25.
The lower real limit is half way between 2.5 and 5, or 3.75. For example, notice the third bar in this histogram. Halfway between adjacent intervals are the real limits of the interval, which determine where a particular data point gets "counted" in the histogram. The numbers on the X axis (also called the "abscissa") correspond to the midpoints of the interval. In Figure 4.2 you will find a histogram produced by SPSS of the sexual behaviour data in Figure 4.1.įigure 4.2 A histogram produced by SPSS for the "number of sex partners in the past year" variable.įirst of all, notice that in this histogram, there are 7 intervals. If you donÕt tell the computer how many intervals to use, it will make the decision based on the data it has. Most computer programs that construct histograms will allow you to select the number of intervals, as well as their width. The choice is between reducing the information sufficiently while still providing enough variability to picture the shape of the distribution. It is constructed by first selecting a number of "intervals" to be used. A histogram is a graphical way of presenting a frequency distribution.