Scatterplot for Measles and Age with WAZ Score

Return to Previous Page

Try running this scatterplot in the following exercise and compare the results to these:

1.  Open keast4j.sav

2.  Click on Graphs, Scatter, Simple, and Define.

3.  Insert waz (weight for age std. deviations) for the Y-axis and hmeasyn (measles immunization) for the X-axis. [ Repeat using WAZ and age]

4.  Click OK.

5.  Double click on the Output graph to make it editable (hash marks around edges).

6.  Click on Chart, Options and select Fit Line, Total (put a check in the box).

7.  Click on OK.

wpeB.jpg (15729 bytes)

This scatterplot appears to say that those with measles immunization (1) have lower nutrition status than those who do not have measles immunization. Because this result is contrary to the logical framework, it is likely due to some other factor in the data that is interfering. It is likely that this might be a case of confounding between some other variable and measles. The first hunch is to look at the effect of age, since nutrition status declines during the first few years of age most often in developing countries. First, look at the scatterplot of age and WAZ score.

This scatterplot of WAZ std. deviations by age depicts the decreasing nutrition status as the child ages from 0 to 60 months. The fit line has a shallow decreasing slope, but if the focus is given to the first 6 months then there is a steeper decline in the WAZ score that is obvious. Considering the effect of age represented here, it will play a large role in exploring the relationship between measles immunization and nutrition status. Now a regression can show the outcome considering age in the equation.


Return to Top of Page