Example 1 Suppose we want to examine the association between unemployment days (x) and self-rated health (y). Unemployment days are measured as the total number of days in unemployment during a year, and ranges from 0 to 365. Self-rated health has the values 1=Poor; 2=Fair; and 3=Good. Let us say that we get an OR that is 0.93. That would mean that we have a negative association: the higher the number of unemployment days, the lower the odds of having better health.
Example 2 In another example, we may examine the association between intelligence scores (x) and the number of books read per month (y). Intelligence scores are measured by a series of tests that render various amounts of points, and ranges between 20 and 160 points. Book reading has the values 1=0 books; 2=1-3 books; and 3=4 or more books. Here, we get an OR of 1.18. We can thus conclude that a higher intelligence score is associated with more reading of books.
Practical example
Dataset
StataData1.dta
Variable name
educ
Variable label
Educational level (Age 40, Year 2010)
Value labels
1=Compulsory 2=Upper secondary 3=University
Variable name
gpa
Variable label
Grade point average (Age 15, Year 1985)
Value labels
N/A
sum educ gpa if pop_ordinal==1
ologit educ gpa if pop_ordinal==1, or
When we look at the results for gpa, we see that the odds ratio (OR) is 4.63. In other words, a unit increase in gpa (e.g. going from a grade point average of 2 to 3, or from 4 to 5) is associated with higher educational level.
The association between gpa and educ is statistically significant, as reflected in the p-value (0.000) and the 95% confidence intervals (4.30-4.98).
Summary There is a positive association between grade point average at age 15 and educational level at age 40 (OR=4.63). The association is statistically significant (95% CI=4.30-4.98).
