Skip to main navigationSkip to main content
The University of Southampton
Practical Applications of Statistics in the Social SciencesResearch Question 1: Confidence in the police

Bivariate analysis

One of the first steps in conducting research is to look at the relationship between your outcome variable – the aspect you are interested in – and potential explanatory variables – those that you think influence your outcome. This is called bivariate analysis – looking at the relationship between two (‘bi’) variables (‘variates’).

In this example analysis, we are interested in finding out what factors influence CSEW respondents’ police confidence, which, you’ll recall, is a continuous variable in our dataset. One of the hallmarks of a continuous variable is that it allows us to calculate and compare means, as the data in these variables are numerical and can take on any values within a set range (for example: height in inches or scores on GCSE exams).

We can use SPSS to calculate and compare the mean police confidence scores between different groups, and to test the statistical significance of these differences. A calculation and comparison of means can be done using the Compare Means function. We can extend these mean findings to the general population from which the survey sample was taken by investigating Confidence Intervals. And, we can test the significance of this mean difference by using either the Independent-Samples T Test or the One-Way ANOVA functions.

Privacy Settings