Graphs and charts

Bar charts are used to visually represent data from experimental studies, in which averages are compared. Usually the two (or more) conditions or groups are plotted on the x axis and the value on the y axis. For example you might plot a bar chart to show cultural differences in the percentage of attachment types measured using the Strange Situation procedure (see figure 1). The x axis would be culture and the bar labels would be given the location of each study and the researchers name(s). The y axis should extend to 100% (the maximum value that could be achieved and the labelled (e.g. percentage of mother-infant pairs. You will also need a key (legend) to explain the colour/texture coding of your bars, e.g. red/dots for Type A, blue/stripes for Type B and yellow/zig-zags for Type C.

Figure 1: A bar chart to show the percentage of attachment types (A, B, C) in two cultures 

Histograms are used to consider the distribution of a full set of scores. For example, you might plot a histogram to show the range of scores achieved on a reading test, where the minimum score is 0 and the maximum is 30 (see Figure 2). The y axis would represent the scores on the test divided into ‘bins’ meaning scores ranges e.g. 11-15, 16-20 etc. To calculate the required number of bins, find the square root of the number of participants, e.g if you had 25 participants, you would plot five bins. To work out the width of the bins, take the range of scores, e.g. maximum value minus minimum value and divide by the number of bins. The x axis should be labelled according to the variable that is being measured. Notice that each bin or class interval in figure 2 is labelled in the middle and with the value that is the midpoint is the range for that bin, e.g. in the bin for scores between 11 and 15 the bin is labelled 13. The height of the bars reflects the number of data points (scores) that fall into the range for each bin. Bars are always touching as the data plotted on the x axis is continuous. Histograms can be useful to check whether the data set has a positive or negative skew or whether they are normally distributed (see below).

Figure 2: A histogram to show the scores on a reading assessment (graded from 0-30)