In order to run a one-way MANOVA, you require the following:

» One independent variable that is categorical with two or more groups (e.g., transport type: bus/car/tram/train).

» Two or more dependent variables that are continuous (e.g., satisfaction score, motivation score). (Laerd, 2013)

The 7 assumptions that are associated with a MANOVA test are;

Assumption #1: You should have independence of observations, which means that there is no relationship between the observations in each group or between the groups themselves. For example, there must be different participants in each group with no participant being in more than one group. This is more of a study design issue than something you can test for, but it is an important assumption of the one-way MANOVA.

Assumption #2: You should have an adequate sample size. Although the larger your sample size, the better; for MANOVA, you need to have more cases in each group than the number of dependent variables you are analysing.

Assumption #3: There are no univariate or multivariate outliers. There can be no (univariate) outliers in each group of the independent variable for any of the dependent variables. This is a similar assumption to the one-way ANOVA, but for each dependent variable that you have in your MANOVA analysis. Multivariate outliers are cases which have an unusual combination of scores on the dependent variables. Both of these assumptions are dealt with later in this guide.

Assumption #4: There is multivariate normality. This is an assumption that cannot be directly tested in SPSS. Instead, the normality in each group of the independent variable for each dependent variable is assessed. This is the closest you can get to testing this assumption and is the procedure that is demonstrated later in this guide.

Assumption #5: There is a linear relationship between the dependent variables for each group of the independent variable. If the relationship is not linear, it can lead to a loss of power to detect differences. This assumption can be tested with scatterplot matrices and is demonstrated later in this guide.

Assumption #6: There is homogeneity of variance-covariance matrices. This assumption can be tested using Box's M test of equality of covariance and is demonstrated later in this guide.

Assumption #7: There is no multicollinearity. There are many different methods to test for this assumption, but one of the most straightforward is to run correlations between the dependent variables to see if there are any relationships that are too strongly correlated. This assumption is demonstrated later in this guide also. (Laerd, 2013).

The following is an incredible resource:

https://statistics.laerd.com/spss-tutorials/multiple-regression-using-spss-statistics.php

In order to run a one-way MANOVA, you require the following:

» One independent variable that is categorical with two or more groups (e.g., transport type: bus/car/tram/train).

» Two or more dependent variables that are continuous (e.g., satisfaction score, motivation score). (Laerd, 2013)

The 7 assumptions that are associated with a MANOVA test are;

Assumption #1: You should have independence of observations, which means that there is no relationship between the observations in each group or between the groups themselves. For example, there must be different participants in each group with no participant being in more than one group. This is more of a study design issue than something you can test for, but it is an important assumption of the one-way MANOVA.

Assumption #2: You should have an adequate sample size. Although the larger your sample size, the better; for MANOVA, you need to have more cases in each group than the number of dependent variables you are analysing.

Assumption #3: There are no univariate or multivariate outliers. There can be no (univariate) outliers in each group of the independent variable for any of the dependent variables. This is a similar assumption to the one-way ANOVA, but for each dependent variable that you have in your MANOVA analysis. Multivariate outliers are cases which have an unusual combination of scores on the dependent variables. Both of these assumptions are dealt with later in this guide.

Assumption #4: There is multivariate normality. This is an assumption that cannot be directly tested in SPSS. Instead, the normality in each group of the independent variable for each dependent variable is assessed. This is the closest you can get to testing this assumption and is the procedure that is demonstrated later in this guide.

Assumption #5: There is a linear relationship between the dependent variables for each group of the independent variable. If the relationship is not linear, it can lead to a loss of power to detect differences. This assumption can be tested with scatterplot matrices and is demonstrated later in this guide.

Assumption #6: There is homogeneity of variance-covariance matrices. This assumption can be tested using Box's M test of equality of covariance and is demonstrated later in this guide.

Assumption #7: There is no multicollinearity. There are many different methods to test for this assumption, but one of the most straightforward is to run correlations between the dependent variables to see if there are any relationships that are too strongly correlated. This assumption is demonstrated later in this guide also. (Laerd, 2013).

The following is an incredible resource:

https://statistics.laerd.com/spss-tutorials/multiple-regression-using-spss-statistics.php

MANOVA is useful when you have multiple responses for a single data point. If you have say males and females and you test them for their english, reading, math and science skills, you would use a MANOVA test for this.

If you have test scores from the same student on say one exam after say 6 mos, 1 yr, 2yr,5yr, etc, you have repeated measures analysis (panel data).

If you have one response and multiple inputs, this is multiple (linear) regression test. If you have males vs females, time of day, and model physical exertion (calories burned), you will use a multiple regression method.

How many responses do you have from each person? What types of responses are they?

MANOVA can be used when we are interested in more than one dependent variable. MANOVA is designed to look at several dependent variables (outcomes) simultaneously and so is a multivariate test, it has the power to detect whether groups differ along a combination of dimensions. For example, we can distinguish drivers, non-drivers and drunk drivers only by the number of pedestrians that they kill, but they might be distinguished by a combination of the number of pedestrians they kill, the number of lamp posts they hit, and the number of cars they crash into.

Thank you very much for your answers. Males and females were measured (by accelerometry) at the same interval of time (7 days). We measured the physical activity in different bands of the day, but this bands were the same in males and females. When we get the results, we have sedentary, light, moderate and vigorous activity in males and females for each of the bands of the day.

Thus, for each subject, we have sedentary, light, moderate and vigorous activity.

My problem is that if for example in differents stripes (for example school, afternoon and night) we have different measurement time between boys and girls. This circumstance can alter the results. For example maybe kids do more physical activity because they have more time measurement in some stripes and not because they actually made ​​more physical activity.

While analyzing two way Manova in spss either we have to put all dependent variables  like in my case i have 6 dependent variable do i need to put all at once. on i have to analyze in pair of  two dependent variable.

You can analyze all the dependent variable at a time .The two way means you should have 2 independent variable with you,it is nothing with dependent variable

But please see below to make sure that you are really addressing the question you intend with MANOVA. If you are interested in individual DVs, it is not the technique for your question.

Huberty, C. J, & Morris, J. D. (1989). Multivariate analysis versus multiple univariate analyses. Psychological Bulletin, 105, 302-308.