Topics
- Completely Randomized Design (CRD)
- Randomized Complete Block Design (RCBD)
- Split-Plot Design
- Latin Squares Design
- 2^k Factorial Design
- Taguchi DOE
References
Penn State (2025) Design of Experiments
https://online.stat.psu.edu/stat503/
Montgomery, D. C. (2013) Design and analysis of experiments, 8th ed. Wiley
https://faculty.ksu.edu.sa/sites/default/files/douglas_c._montgomery-design_and_analysis_of_experiments-wiley_2012_edition_8.pdf
Winer, B. J. (1962) Statistical principles in experimental design. McGraw-Hill
Toutenburg, H., Shalabh (2009) Statistical analysis of designed experiments, 3rd ed. Springer
https://perpus.unimus.ac.id/wp-content/uploads/2012/05/ebooksclub.org__Statistical_Analysis_of_Designed_Experiments__Third_Edition__Springer_Texts_in_Statistics_.pdf
Kaps, M., Lamberson, W. R. (2004) Biostatistics for animal science. CABI
Lawson, J. (2015) Design and Analysis of Experiments with R. CRC Press
https://books.google.it/books/about/Design_and_Analysis_of_Experiments_with.html?id=TOxMBgAAQBAJ&redir_esc=y
Hi,
I have simulated 100ns of 4 protein-complexes and have three triplicate runs simulated for each.
I want to see how the reflection of 100ns on each unit of the protein differs in each of the four complexes, such that I can say for sure that in 100ns this unit is different in complex C compared to the parent complex A, and sister complexes B and C are the same but C and D are different. Although time is not a visible factor, it is accounted into the Root mean square fluctuation value of each unit of the protein. So how can I account for violation of sphericity due to this factor: The units are physically bonded to each other so one unit may affect the other unit. Yet knowing this I still want to highlight which units of the protein are different in the four complexes, and how the sister variants are different from each other and how they are different from the parent variant.
Hi,
In general the following are two ways to deal with violations of sphericity:
1. Use MANOVA instead: See https://real-statistics.com/multivariate-statistics/multivariate-repeated-measures-tests/
2. Use a sphericity correction: See https://real-statistics.com/anova-repeated-measures/sphericity/
Charles
Hello
How can I perform the multiple range test LSD (least significant difference) in two-way ANOVA?
Hello Hugo,
I don’t know how often anyone uses the LSD test after a significant two-way ANOVA, I don’t have a webpage about this subject, but I will explain how you could perform such a test.
1. The LSD after one-way ANOVA is explained at LSD Post-hoc Test
2. The following webpage described how to conduct a HSD test after a significant two-way ANOVA.
https://real-statistics.com/two-way-anova/follow-up-analyses-for-two-factor-anova/tukey-hsd-after-two-factor-anova/
You can now make the type of modifications to make an HSD test into an LSD test as described in #1.
Charles
Hello Charles,
my research question is to measure the reliability of a tool to evaluate muscle strength across three sessions (S1, S2, S3) in two groups (people with CLBP and healthy controls). However, I’m interested in checking if there are any differences in strength levels between the groups and also exploring if there is any learning effect between sessions. I thought about exploring this through the Two way analysis of variance tests ANOVA. Is what I’m planing to do correct?
Hi Shouq,
From your description, it does seem like Two way analysis of variance is the way to go.
Charles
Dear Charles,
Thank you so much for your help,
I have one more thing that concerns me and I hope you can answer me,
I measured the reliability between sessions for each group using the interclass correlation coefficient (ICC) for relative reliability and using the SEM and the LOA for absolute reliability.
My question is: Are the results I’ll get from the Two-way analysis of variance tests (ANOVA) to explore the learning effect between sessions will be similar to the ones I’ll get from the LOA?
Regards
Shouq
Hello Shouq,
What sort of Two-way ANOVA do you have in mind?
I am assuming that you are using Bland-Altman for LOA. See
https://real-statistics.com/reliability/interrater-reliability/bland-altman-analysis/limits-of-agreement/
In this case, I don’t see the connection between LOA and Two-way ANOVA.
Charles
Who needs Minitab when we have you?
Thank you