Data Screening

Data screening is important to ensure no unintended errors such as wrong values and outliers remain in the data before performing further statistical analysis.
It is encouraged to do data screening after completed the task of data entry.
If there is an error but we skip data screening, it will cause more problems as the further analysis involve the previous data entry too.

Step 1

• On the Menu Bar, select "Analyze" --> "Descriptive Statistics" --> "Frequencies".
• A pop-up window will appear.

Step 2

• Insert the relevant variable(s) that you would like to screen in the "Variable(s)" box.
• You may click "Statistics" to obtain more options of statistical screening.

 

 Below is the example of my data screening. I clicked "Statistics" and checked "Range" to view whether my data entry is within its range or not. (1-5)

 

Step 3

• Assess the result of data screening by looking at the table. 

 

• For instance, as the items of Scale B is a 5-point Likert scale, the range of score supposed to be from 1 to 5 only.
• However, there is a score of "50" in the table which can be identified as a wrong value.
• Thus, the data in B1 column on the Data View tab should be reviewed.

Step 4

• On the Menu Bar, select "Data" --> "Sort Cases".
• Insert the column that you would like to sort, specifically that one that contains wrong value or outlier in the "Sort by" box.
• Choose whether you would like to sort that particular column in ascending or descending order.

 

 As I choose "Descending" order, thus the highest score in "B1" column will the at the topmost.

As can be observed, the other columns such as "ID" and "Age" are also been sorted.

By that, we can identify which participant's data was entered wrongly. From the example, the participant with "9" as the ID is the one who has a data of "50".

Step 5

• Refer back to the affected scale answered by the participant with related ID number. (from the example above, participant with ID of "9")
• Look again at the score answered by the participant.
• Modify the data in SPSS' Data View tab to match the exact value from the original scale or item.
• To make the Data View more organized and neat after sorting, rearrange the data by sorting the "ID" column in ascending order.