Statistical Research Consultants BD (SRCBD) Missing Value Management. Entering missing data in SPSS: web:

Size: px
Start display at page:

Download "Statistical Research Consultants BD (SRCBD) Missing Value Management. Entering missing data in SPSS: web:"

Transcription

1 Missing Value Management Entering missing data in SPSS: It s likely that your data set will contain some missing values, where participants didn t answer some items on a questionnaire or didn t complete some trails in an experiment. 1. When you initially enter your data, leave any missing values as blank cells. 2. To get SPSS to fill in all the empty cells, go to Transform Recode into Same Variables. 3. Move all your variables into the right hand box and click on Old and New Values. Mobile: , , FB page: Page 1

2 4. On the left select System- or user-missing and on the right enter a number that will not otherwise occur in your data set (eg ) in the New Value box. Click on Add, then Continue. Click on OK. 5. All blank cells will now be replaced with the value you entered in the previous step. Mobile: , , FB page: Page 2

3 6. So that SPSS doesn t include these numbers in any calculations you must complete one final step. Go to the Variable View. 7. The eighth column from the left is called Missing. Click on the first cell under this column, and click on the blue box that appears in the cell. 8. Select Discrete Missing Values and enter in the box the number that you chose in step 4. Click OK. 9. Repeat steps 7 and 8 for every row in the variable view (You can copy the first Missing cell and paste into all cells below to save time). N.B. If you are computing total scores please consider the impact missing values will have on this calculation. It might be more suitable to calculate mean scores instead based on the number of answers you have for each participant. Alternatively, some questionnaire manuals advise replacing missing values with the participants mean score before calculating a total score. Mobile: , , FB page: Page 3

4 Replace Missing Value in SPSS: Using ExampleData002' replace missing values on Age variable with the mean of Age. 1.Open the data ExampleData002'. Click Transform -> Replace Missing Values 2. Now, you should see a dialog box similar to the one on the left, below. Select and use the arrow to move Age [Age] from the available variables list to the "New Variable(s):" box. By default, "Series mean" is selected as the "Method"; however, it is not the only option and sometimes not the best option. So, click on the drop-down arrow to the right of "Series mean" and review the choices you have available. Mobile: , , FB page: Page 4

5 For the current example, we will use the default (series mean) for illustrative purposes. It is fairly widely accepted that mean imputation should not be used unless the percentage of missing data is very, very low (i.e. well below 5% missing). Click the "OK" button to complete the imputation and creation of the new variable: Age_1. We'll notice a new variable will be created during this procedure, called Age_1, which will include all the existing values and the mean of Age imputed in place of missing values or blank cells. Series mean: Missing values are replaced by the mean (or average) value of all other values for that variable. Mean of nearby points: Missing values are replaced by the mean of surrounding values (that is, values whose SPSS case numbers are close to the case with a missing value). We may designate how many values to use under Span of nearby points. Median by nearby points: Missing values are replaced by the median of surrounding values. We may designate how many surrounding values to use. Linear interpolation: Missing values are replaced by the value midway between the surrounding two values. Linear trend at point: If values of the variable tend to increase or decrease from the first to the last case, then missing values will be replaced by a value consistent with the trend. Mobile: , , FB page: Page 5