Your data analysis solution
Name is required.
Email address is required.
Invalid email address
Answer is required.
Exceeding max length of 5KB

Residue analysis and DDL

By alain brauman | Feb 05, 2016 09:59AM CET

When I do Anova with XLSTAT, the results are sometime different from one to other analysis with the same data table. Sometime the analysis stop after residue analysis and sometimes its goes on with tukhey analysis. I wonder if this problem originated from the residue analysis. Why we just have a graphic visualisation of the residue test and not a shapiro test for normality of the residue and a barlett one for the homogeneity ?

Another point is about ddl, for the same table some variables have a ddl of 0 and some others has a ddl of 1 or 2 ? with the same number of repetitions and observations ? how could it be possible ?




By Efthalia | Feb 05, 2016 03:44PM CET | XLSTAT Agent


This is quite surprising. Perhaps you have modified some parameters of the anova dialog box. In that case, even with the same data you could get different results.
Can you please verify that there are no differences between the settings of the two analyses ? You could also send us one example.

Indeed, you can only visualise the residuals through the plots. If you wish to get a normality or a barlett test, you can use the functions Normality test and Comparison of variances respectively.

Regarding the last point, please send us the data set used and the output at support@xlstat so that we can have a closer look.


This question has received the maximum number of answers.

Contact Us
seconds ago
a minute ago
minutes ago
an hour ago
hours ago
a day ago
days ago
Invalid characters found