One-Way MANOVA Test: How to Assess If Mean Differences Exist Between the Samples Using SPSS?

It is extended version an ANOVA with two or more dependent variables. ANOVA test is used for evaluating the difference in means between two or more related groups, while a MANOVA test is used for evaluating the difference in two or more vectors of factors. 

ASSUMPTIONS:

1. All the observations should be statistically independent.
2. We should have an adequate sample size. As a larger sample size, the better it is. We should be having more cases than the number of variables in each group.
3. In ANOVA, the Dependent variables are normally distributed within the group. whereas in MANOVA,the Dependent variables have multivariate normality within the groups.
4. There are no univariate or multivariate outliers. Univariate outliers are often just called outliers.In one-way MANOVA, we see how to:

        (1) detect univariate outliers using box plots using SPSS statistics in order to check outliers

        (2) check for multivariate outliers using Mahalanobis distance, which we can do in SPSS 

5. The dependent variables should be ratio or interval level.

EXAMPLE

A company wants to determine user acceptance with their new online travel booking portal. The team conducts a study where they assign 40 random people into three groups. The first group needs to book their travel through an automated online-portal; second one registers over the telephone; the third one sends a request via the online-portal and will get a callback in return. The purpose of the study is to measure the rate of user acceptance to use the system under following conditions The three major factors to be examined are effort and ease of use and perceived usefulness.

EXAMPLE : USING SPSS

An educator was intrigued to learn if there was a scholarly contrast in understudies from various states. She arbitrarily chose 20 understudies from every one of three states; California, Africa, and Canada who were a piece of the entering rookie class at the University. She evaluated them in view of their English and math position tests. The free factor is the state, what’s more, the reliant factors are the scores on the two arrangement tests. The outcomes are as follows

Step 1: Enter the information in SPSS as first is state, second Maths score, English score respectively

Step 2: Click on Analyze from toolbar. 

Then click on  General Linear Model 

Then use multivariate 

Step 3: In Multivariate Window move-both Math & English to the Dependent Variable box and make it Fixed Factor. Then select Options from the dialog box. 

Step 4: Move State to the Display and put a check next to Descriptive Statistics, Then Click on Continue

Step 5: Select descriptive statistics, observed powers, and estimates of effect size. Then click continue in the dialog box.  

Step 6: Analyze the output from the test performed. 

The DS table provides the mean and standard deviation for both the dependent variables & independent variables

The p-value = .223 implies that there is no statistically significant academics from different states. 

Now check between- subjects effects to examine the effect of dependent and dependent variables. 

P value= 0.499& 0.110 >0.05 this implies that state does not have any significant impact on English and maths subject results.