Data ranking
Another common way to present small datasets is to rank the data, assigning ordinal labels (first
, second
, third
, and so on) to the data points. This can be done by sorting the data on the key field.
Figure 3-17 shows an Excel worksheet with data on students' grade-point averages (GPAs).

Figure 3-17. Excel data
To rank these, select Tools | Data Analysis | Rank and Percentile to bring up the Moving Average dialog box:

Figure 3-18. Ranking scores in Excel
Here, we have identified cells B1
to B17
as holding the data, with the first cell being a label. The output is to start in cell D1
. The results are shown in the following screenshot:

Figure 3-19. Results from Excel ranking
Column D contains the (relative) index of the record being ranked. For example, cells D3-G3 show the record of the third student in the original list (in column A) with name Cohen: that student ranks second, at the 86th percentile.