Each row in the raw data set represents how a single person performed in the experiment. The first column is an identifier for the participant. The second column identifies the participant's sex.  The third column is the participant's age.  The next thirty columns indicate whether a given participant recalled the word at each of the thirty serial positions when they had to count backwards by 3. A zero indicates that they did not recall the word at the given serial position, while a one indicated that they did recall the word at the given serial position.  The next thirty columns give similar information, but for when the participant did not have to count backwards by three.

1. Select the raw data set by dragging your mouse over it.
2. Copy the data set into Microsoft Excel.
3. At the bottom of each column of data, enter a formula similar to this:
   
   =average(d4:d44)
   
   The "d" indicates the column of data for which you want the average. The 4 indicates the first row for which you want the average. The 44 indicates the last row for which you want the average. That is, we are asking for the average of the scores in cells d4 through d44. You should replace the "d", "4", and the "44" with values that are appropriate for the column for which you are taking the average and the first and last row of data in your data set.
4. Copy the formula into the cells at the bottom of the remaining columns. These averages represent the probability that the item in a given serial position was recalled.
    
5. Select the averages you just created for columns D through AG (that is, the counting data.)
6. In the ribbon, click on Insert | Line | Line with Markers
7. In the ribbon, click on Chart Tools Design | Select Data
8. Click the Add button
9. In the dialog box that appears, type "No Counting" for the series name
10. Select the no counting averages for the Series Values.
11. Click OK
12. Select Series1 and click the Edit button
13. Type "With Counting" for the series name
14. Click OK
15. Click OK
16. The graph shows the results of the serial position effect for the counting and no counting data. Theoretically, the first several serial positions should have a high probability of being recalled. This should taper off as the serial position increases. This is called the primacy effect. The middle items should have a low probability of being recalled.  The end items should be recalled better for the no counting data than for the counting data.
     
17. Make sure that your graph is still selected.  Click on Chart Tools | Layout
18. Click on Axis Titles | Primary Horizontal Axis Title | Title Below Axis
19. Type "Serial Position" (without the quotes) and press Enter
20. Click on Axis Titles | Primary Vertical Axis Title | Rotated Title
21. Type "Proportion Recalled" (without the quotes) and press Enter
     
22. In an empty cell enter a formula similar to:
    
    =countif(b4:b44,"F")
    
    The "b" represents the column that has the sex data.  The 4 indicates the first row for which you want the count. The 44 indicates the last row for which you want the count. That is, we are asking for the number of the scores in cells b4 through b44 which have the value "F" (e.g. the number of females who participated.)
23. In an empty cell enter a formula similar to:
    
    =countif(b4:b44,"M")
    
    The "b" represents the column that has the sex data.  The 4 indicates the first row for which you want the count. The 44 indicates the last row for which you want the count. That is, we are asking for the number of the scores in cells b43 through b44 which have the value "M" (e.g. the number of males who participated.)
24. In an empty cell enter a formula similar to:
    
    =average(c4:c44)
    
    The "c" represents the column that has the age data.  The 4 indicates the first row for which you want the count. The 44 indicates the last row for which you want the count. That is, we are asking for the mean age in cells c4 through c44 (e.g. the mean age of the participants.)
     
25. Optional: Clean up the graph so it is closer to APA style for figures.  For example:
    1. Remove the border and gray background from the graph and chart area
    2. Make the lines and markers black for both data series
    3. Increase all font sizes to 12 points and remove the bold
    4. Remove the border from the legend
    5. Make the graph sufficiently large
    6. Remove the grid lines

Reload the raw data set