Associations

A lot of quantitative research is about examining relationships or (associations) between variables (see X, y, and z for a more detailed discussion about those issues).

Effect

Assuming that all is done correctly, data analysis will give us information about the change in the outcome per unit increase in the exposure. Another word for this is (statistical) effect.

Direction

It will also provide information about the direction of the relationship (i.e. whether the relationship is negative or positive).

Statistical significance

Effect and direction are the two most important outcomes of data analysis, but it is not uncommon that research inquiry also focuses on a third point: statistical significance.

Statistical significance can be seen as an indicator of the reliability of the results – although that is important, it is not what exclusively should guide which findings we focus on and which we discard.

Practical importance

A fourth issue that needs to be considered is whether the findings have any practical or clinical importance – in order words; do they matter?

Priority list

We therefore suggest the following priority list when it comes to how results from data analysis should be interpreted and valued:

Effect	How much does the outcome change per unit increase in the exposure?
Direction	Is the relationship positive or negative?
Statistical significance	Is the relationship reliable?
Practical importance	Is the relationship relevant?