Can the function 'anovan' handle unbalanced designs?

1 次查看（过去 30 天）
Zwithouta
Zwithouta 2017-8-10
编辑： Zwithouta 2017-8-11
Hello Guys,
imagine the following experimental setup for an unbalanced, mixed-ANOVA:
Unbalanced, non-repeated ANOVA with REPLICATION using
continous & categorical predictors
Entity tested: Washing powder
Outcome variable [continous]: Amount of living mites after one washing cycle (integer)
1st Predictor [continuous]: Amount of washing powder in gram [0:20:80]
2nd Predictor [categorical]: Branch of washing powder [Lenor, Spee]
3rd Predictor [categorical]: Branch of washing machine [Bosch, Siemens]
Number of possible predictor combinations: 5 levels * 2 Levels * 2 levels = 20
REPLICATION part of design: experiment is replicated 50 times for
each predictor combination.
-> e.g.
50 cycles, 0 gram, Lenor, Bosch
50 cycles, 0 gram, Lenor, Siemens
50 cycles, 0 gram, Spee, Bosch
50 cycles, 0 gram, Spee, Simens
same for 20, 40, 60, 80 gram
UNBALANCED part of design:
1 cycle missing for 20g, Lenor, Bosch
1 cycle missing for 60g, Lenor, Bosch
-> 18 predictor combinations have 50 cycles, 2 have only 49 cycles
-> Total washing cycles: 50cycles*5*2*2 - 2 = 998
Non-repeated part of design:
Different entities are used to test all predictor-level combinations
(the same washing powder unit is not/can not be used in two different washing cycles)
Is the function 'anovan' able to handle a situation like that, in which not all predictor-combinations have the same number of repetitions?
I would use the following input arguments and Name-Value-Pairs to run the mixed-ANOVA. Would this be adequate given the upper example?
[mixedANO.p, mixedANO.tbl, mixedANO.stats, mixedANO.terms] = anovan(y, groups, ...
'model', 'full', ...
'continuous', [1], ...
'varnames', {'WP used', 'WPbranch', 'WMbranch'});
Thanks a lot!

请先登录，再回答此问题。

回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Analysis of Variance and Covariance 的更多信息

标签

产品

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by