Questions tagged [post-hoc]
"post-hoc" refers to analyses that are decided upon after the data has been collected, as opposed to "a priori".
722 questions
1 vote
0 answers
23 views
Post hoc analysis in complex survey data analysis chi square test
I am running chi square tests on complex survey data of annual household income and education levels, and I want to determine which cells are contributing to the significant results. I am familiar ...
- 121
27 votes
1 answer
3k views
How to tell author that post hoc data manipulation is not okay
I’m a clinical/forensic psychologist with a PhD and some research experience, and often get asked to be an ad hoc reviewer for a journal. I recently recommended rejecting an article that had a lot of ...
- 353
2 votes
1 answer
70 views
Adjusting p-values for single post-hoc comparison
I have a linear model in R my_lm <- lm(Measurement ~ Group, my_df) where 'Group' has 3 levels (UHR, CC, HC). I use treatment contrasts to do the following ...
- 688
1 vote
1 answer
112 views
What statistical test to use here?
How to analyse this data? The percentage of wound closure was determined for 4 different substrates + control group over 4 different time points (18, 24, 48 72 h). Each substrate had 3-4 replicates, ...
- 11
0 votes
0 answers
81 views
2 (within) x 2 (within) x 2 (between) design: three-way interaction post-hoc test
I'm currently working on my master's thesis and would really appreciate some advice on the appropriate statistical analysis for my design. Study context: My research examines the stress reduction ...
- 3
0 votes
0 answers
124 views
Post-hoc test in R for lm() with significant interaction between numeric and multiple categorical variables
I am interested in how the relationship between two traits (trait1 and trait2) varies between groups (A and B) and treatments (C and T). Specifically, I want to know whether the relationship between ...
- 11
4 votes
2 answers
644 views
Which is better - FDR or Holm-Bonferroni? [closed]
I am conducting an auditory flanker task with monkeys. It is a within-subjects design with 4(6 soon) subjects and has 4 conditions. I want to run pairwise comparisons after running a one way RM ANOVA. ...
0 votes
0 answers
68 views
GLMM for count data with random effect of Subject in small sample size with risk of inflated type I error, how to infer factor effects and post-hoc
I am a newbie in GLMM. My data designed with repeated measures involved 9 subjects, each subject performed in duplicate, the count as a response. Since dispersion, the model of negative binomial was ...
- 1