Using Statistical Regression Methods in Education Research





There are different types of validity, but in quantitative research the concept generally refers to the extent to which a data collection instrument accurately measures the concept or construct which it sets out to measure.


A variable is a measurable characteristic or attribute of cases or units. Unsurprisingly a variable is expected to vary between cases or over time. For example gender is a variable that for different cases can take the values male or female, which may be coded 0 and 1 respectively. In regression analysis we need to distinguisg between explanatory and outcome variables, with the latter being the phenomena we want to account for or explain.

Variable View

The SPSS variable view is the other 'tab' of the spreadsheet. It is here where you can edit the properties of your variables. For example, you can label a variable, code its values and specify whether it is continuous, ordinal or nominal.

Variable inflation factor (VIF)

The variable inflation factor (VIF) provides a statistical indication of multicollinearity in regression analysis. It tests the explanatory variables to see if there are any strong linear relationships between them. If the value of the VIF is 10 or more your model may well have problems with multicollinearity.


The variance estimates how spread out around the mean data for a particular variable is. The distance between each data point and the mean is calculated and squared (to remove minus signs for cases where the value is less than the mean). These squared differences are then summed to produce a 'sum of squares'. The sum of squares is averaged (divided by N-1) to calculate the variance.

Page contact: Feedback to ReStore team Last revised: Thu 28 Jul 2011
Back to top of page