One-way ANCOVA in SPSS
One Way ANCOVA in SPSS
One-Way ANCOVA is a statistical test used to determine if three or more groups of variables of interest are significantly different from each other, while accounting for the effect of another variable termed as a covariate.
Dependent variable and covariate variable(s) should be measured on a continuous scale.
Independent variable should be a categorical variable.
Independence of observations is necessary, meaning that there is no relationship among observations in each group or among the groups.
There should be no significant outliers.
Residuals should be approximately normally distributed for each category of the independent variable.
Homogeneity of variances is required.
The covariate should be linearly related to the dependent variable at each level of the independent variable.
Homoscedasticity is required.
Homogeneity of regression slopes is required, which means that there is no interaction between the covariate and the independent variable.
Here, Depression Score is considered as dependent variable, Body Mass Index is considered as independent variable and Diastolic Blood Pressure is considered as covariate.
In Body Mass Index, 1 is labelled as healthy, 2 is labelled as overweight and 3 is labelled as obese.
In SPSS, One way ANCOVA is found in Analyze > General Linear Model > Univariate
Then, we will get ‘Univariate’ dialog box.
Add Depression Score in the ‘Dependent Variable’ box, add Body Mass Index in the ‘Fixed Factors(S)’ box and add Diastolic Blood Pressure in the ‘Covariate(s)’ box.
In the Model dialog box, add independent variable and covariate variable in the ‘Model’ box by selecting ‘Main effects’ in ‘Build term(s)’ – ‘Type’ and then click on ‘Continue’.
In the options dialog box, add independent variable in the ‘Display Means for’ box and select ‘Descriptive statistics’ under ‘Display’ group and then click on Continue and Ok
Output of One-way ANCOVA
Consequently, ‘Between-Subjects Factors’ table would be obtained. This table gives the N for categories of independent variable (Body Mass Index).
Then, we will get a ‘Descriptive Statistics’ table that presents the descriptive statistics (mean, standard deviation, number of participants) on the dependent variable (Depression Score) for the different levels of the independent variable (Body Mass Index)
Then, we will get ‘Tests of Between-Subjects Effects’ table.
This table demonstrates the overall statistical difference in the Depression Scores among the different Body Mass Index groups after their means are adjusted for Diastolic Blood Pressure.
Here, the p value for (Sig) group is equal to 0.000. Thus, a statistical difference can be confirmed in Depression Scores among the BMI groups after adjusting for Diastolic Blood Pressure.
Then, the table of ‘Estimated Marginal Means’ would be obtained. This table illustrates the manner in which the covariate has adjusted the original depression scores for different BMI groups. These new values represent the adjusted means.