New answers tagged anova
1
vote
Anova model assumptions. How to go by?
Why do you think you are "going wrong" at all? Sometimes, there isn't homongeneity of variances. What you should do about it depends on various things. You could look at some plots.
Yes, ...
- 109k
2
votes
Can Tukey HSD groups be of slightly different sizes?
Tukey HSD test for unequal sample sizes is Tukey-Kramer test, I propose you check this one out! But yeah, just removing data for no 'good' reason, is almost always bad practice.
4
votes
Different results between ANOVA and Linear Mixed Effects
First, apart from Greg's excellent note about sequential vs. simultaneous tests, I wonder why you think the results should be more or less the same.
And for your final question:
How should these ...
- 109k
8
votes
Accepted
Different results between ANOVA and Linear Mixed Effects
The anova function returns the sequential tests on the model (the aov function would be used if you wanted to compare to ...
- 50.6k
1
vote
2-way Anova on Unequal Group Proportions
Which method or technique should I use to account for the binary response variable?
For a binary response you can use logistic regression. In SAS that would be ...
- 56.5k
1
vote
Test for synergy using Bliss independance model for two drugs
This could be quite a long answer as it's something I've also been working on recently. The Bliss Independence model (Bliss, 1939; Greco et al., 1995, Geary, 2013) is an Effect-Based method (in ...
- 56.5k
5
votes
Performing ANOVA with missing data
This is not the usual missing data problem. Usually "missing data" means you didn't measure it, or someone didn't answer a question, or the scale broke, or something like that.
Here, it ...
- 109k
2
votes
Confusion about replication in ANOVA
"Replication", at least in my experience, usually means something else. It means repeating the whole experiment or study. (However, use of terms varies by field, so maybe "replicate&...
- 109k
7
votes
Non-significant results when running Kruskal-Wallis, significant results when running Dwass-Steel-Critchlow-Flinger pairwise comparisons
What is ANOVA?
It is important to note some of the reasons that ANOVA exist, as it can provide some context about how you should consider your omnibus ANOVA and the pairwise tests that accompany them (...
- 8,210
11
votes
Non-significant results when running Kruskal-Wallis, significant results when running Dwass-Steel-Critchlow-Flinger pairwise comparisons
My question is - is DSCF a post-hoc test which is supposed to only be applied when statistically significant differences are found on Kruskal-Wallis?
Yes, the Kruskal-Wallis test is a non-parametric ...
- 56.5k
4
votes
Non-significant results when running Kruskal-Wallis, significant results when running Dwass-Steel-Critchlow-Flinger pairwise comparisons
To me, this question is the same as whether to do any kind of pairwise comparisons after ANOVA. The fact that you used a nonparametric test doesn't change the key issues.
This issue has been discussed ...
- 109k
2
votes
Synergistic effects: what statistics should I use?
Statistics cannot tell you if an interaction is synergistic!
You need to characterise the dose-response relationship of both agents and see if both together yields an effect that is greater than ...
- 13.6k
0
votes
How can I explain the intuition behind ANOVA?
I've noticed that there are few sources that actually explain why variation between is close to variation within under null hypothesis. For simple ANOVA, the intuitive explanation is:
Variation within ...
- 171
1
vote
T test or a different analysis? advice
A t-test compares means. But you have two independent variables (depth and treatment) so a t-test can't be right. You probably want some form of regression, but it might have to be a multilevel model, ...
- 109k
0
votes
non-central F-statistic confidence intervals seem inconsistent with ANOVA p-values
You need to solve numerically for the non-centrality parameter (NCP) such that your estimate agrees with the corresponding quantile of the F distribution. You can't just plug the estimate as the true ...
- 11.6k
1
vote
linear regression vs anova or ancova in clinical trial
Aren't these tests specific cases of linear regression?
Yes, ANOVA and ANCOVA are both special cases of linear regression.
How should I compute when doing analysis of variance or covariance.
ANOVA ...
- 56.5k
1
vote
non-central F-statistic confidence intervals seem inconsistent with ANOVA p-values
You are on the right track in that the central $F$ distribution assumes $H_0$ whereas the non-central one assumes $H_A$, but what you are doing with the latter does not make a lot of sense.
To ...
- 1,836
1
vote
Accepted
How to consider independent samples in a mixed-design experiment?
In this type of design, involving repeated measures, because of the interplay between between-subjects factors (the two populations) and within-subjects factors (the differing conditions for each ...
- 56.5k
3
votes
Regression coefficient before vs. after demeaning X
Here's something more ...intuitional, if that's a word.
The $\beta_i$ coefficients are the predicted increase in $Y$ for each unit increase in $X_i$ and the increase is linear.
Removing the mean (or ...
- 109k
5
votes
Regression coefficient before vs. after demeaning X
Intuition is as follows:
Without demeaning, $\hat{\beta}$ is the MLE for
$$Y_i = \beta_0 + x_{i 1}\beta_1 + \dots + x_{i p}\beta_p + \varepsilon_i,$$
assuming $\varepsilon_1, \dots, \varepsilon_n$ are ...
- 151
1
vote
Accepted
When detecting outliers for an ANVOA test for multiple groups, should I do this for the whole population sample or by group?
Dave answered your first question in a comment: Look group by group.
You also asked if it will "affect the results of this ANOVA?" This is a little trickier question than it appears. Every ...
- 109k
3
votes
Accepted
How to calculate slope and linear regression in R
I'm not entirely sure what your "carcass" variable is here, but I think I can still answer your question. The reason you are likely confused is because the factors in a regression (the ...
- 8,210
0
votes
Nested ANOVAs in R vs Averaging Data Points
Hmm, i may have the same concern. By testing the main factor in your case as F(Condition) = MS(Condition)/MS(Residuals), one doesnt know whether the result supports extra variation in Condition or ...
1
vote
Comparing statistical difference between two trendlines/slopes
You can test for this using a linear model (or various flavors of ANOVA that use the same underlying model). What I'm going to say assumes the slope is in fact linear - what you hope to show - and ...
- 1,836
1
vote
Use of simple statistics for matched study designs
Does this mean that given a matched design, one cannot use simple statistics such as a T-test/Anova to compare the different outcome groups, since such methods do not allow for adjusting for the ...
- 60.4k
1
vote
Non-constant treatment periods for A/B test samples
I would aggregate your data so it looks like this:
...
- 34.7k
3
votes
Non-constant treatment periods for A/B test samples
There is no "right" approach here so I will give you a couple of options.
Remove multiply exposed users This is the typical approach in AB testing. However, this is usually done when there ...
- 33.3k
2
votes
Sample size calculation for RM ANOVA; the difference between WebPower package in R and G*Power 3.1.9.7
I don't get the same output as you do from G*Power; the lowest critical value I can achieve for your case 1 is $F>3.0819664$ using the provided parameters. Still, I think a key point is that you ...
- 1,836
0
votes
Unsure of statistics needed while working with replicates
Not very neat to do by hand, but multi-way ANOVA is completely possible here and runs easily in R or other languages. Consider the following example data set:
...
- 3,400
0
votes
Which type of ANOVA should I select for a Power Analysis of a Factorial Design
I reviewed the simulations I was running and understood that I was running the ANOVA type III wrongly (using car::Anova function).
In a balanced design as the one I ...
- 105
1
vote
ANOVA type III strange results
I was re-reading several posts online and the documentation for the car::Anova function and I was setting the contrasts outside of the lm, thus causing the error. ...
- 105
2
votes
Accepted
Measuring the effect of treatment on variable over time
A very easy way to accomplish this would be through a mixed model, with the treatment as a fixed effect and random slope, and the period of measurement as the random intercept. As an example below ...
- 8,210
Top 50 recent answers are included