#### Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Supported by

# andersony3k

andersony3k
Joined
Visits
196
Last Active
Roles
Member

• This is a fine approach. (I've been forced--by a journal editor--to use it in the past). However, I think the result can also be misleading. The interval is drawn around a particular mean, suggesting that it pertains to confidence about the location…
Comment by andersony3k May 17
• I think that if Excel handles it just fine, Excel is doing it wrong. For an ANOVA, you need to estimate variance across conditions as well as error variance (variance between conditions). An ANOVA can be calculated only if at least one of the six ce…
Comment by andersony3k April 27
• It seems to me that you want to test two different alternative hypotheses (relative to two different nulls): The hypothesis that the control condition's mean is different from 0.10, and the hypothesis that the treatment condition's mean is differen…
Comment by andersony3k January 20
• Hi. Perhaps you should take a look at this . . . https://cran.r-project.org/web/packages/emmeans/vignettes/basics.html
• No. The results may not be the same, especially if you have unequal group sizes. For example, the mean response value for Condition X1 could be different when all response values in X1 count equally, versus when X1 consists of a large subgroup (X1Z1…
• I meant to say, you need at least two observations per cell to do an ANOVA.
• You need at least one observation per cell to do an ANOVA.
• MaximumLuminis, It seems that you are doing a 2 X 2 repeated-measures ANOVA, but with the addition of two covariates--the two data columns containing the "pre" values. I think what you are doing would be correct if you were not seriously v…
• A correlation between B (the "before" value) and D (a before-after difference score that you compute).
• Hi. This actually does not work in JASP 0.15.0 (Windows 10). It generates an error, as shown below. https://forum.cogsci.nl/uploads/723/8JBAWDSZJJX9.png
• It's not so surprising to me. To what degree do your data meet all of the underlying assumptions? For example, nonlinear relationships between the predictors and the criterion variable could really screw things up. Obviously, the ideal way to identi…
• Here's a relevant and recent in-press paper: https://sites.google.com/view/complexcognitionlab-bgsu/research/publications-references/lab-publications/biasing-the-input
• My thought is, that would only make sense in a narrow range of circumstances. One circumstance is when there are just two levels of a factor being compared, in which case a directional t test could be used instead of an ANOVA. Another is when the de…
Comment by andersony3k June 2021
• In any t test, the theoretical, sampling distribution of t scores is distributed according to a t distribution, not a normal distribution. It is the population(s) of raw scores, from which the data scores are sampled, that are assumed to be normally…
Comment by andersony3k June 2021
• I believe that what you want to do is run an Analysis of Covariance, with two groups and one covariate.
Comment by andersony3k March 2021
• The solution (without having to write code) was given to me some time ago in another thread: https://forum.cogsci.nl/discussion/4041/feature-request-allow-each-row-within-a-loop-to-run-different-sequence#latest Also, I've created a simple, working e…
• https://forum.cogsci.nl/uploads/903/PQSDRP24D10G.jasp In the attached file, you can see that when an ANOVA is conducted on the raw Y data) along with Levene's test for variance inequality, Levene's test shows that the variance for Group A is signif…
Comment by andersony3k July 2019
• FYI: My finalized slide. https://forum.cogsci.nl/uploads/169/0K4IXDAJ7M88.png
Comment by andersony3k June 2019
• OK. My slide is getting really close. I hope it is permitted to speak of a "prior probability" distribution (even though the probabilities don't sum to 1.0), such that: "Prior to knowing the data, this is a distribution of the subjec…
Comment by andersony3k June 2019
• OK. This is good. I'm now feeling more positive about increasing the emphasis I place on Bayesian statistical analysis in my classes and my research. Regarding my annotations: (1) Perhaps I can say "This distribution is NOT NEEDED for hypothes…
Comment by andersony3k June 2019
• Yes you are welcomed to use it. But I guess it would be more accurate to use version below, which says that the null hypothesis is "not shown." http://forum.cogsci.nl/uploads/277/TFLJC3L76H8O.png
Comment by andersony3k June 2019
• OK. Here's a slide that I could use for teaching: I've annotated the JASP output to indicate how its various components are to be interpreted. I would be grateful for any feedback concerning errors in the annotation. http://forum.cogsci.nl/uploads/…
Comment by andersony3k June 2019
• However, that "posterior" distribution does not equal: ( 50:50 times ( P(D|"A zero-variance distribution located at ES = 0.0") / P(D|"A Cauchy distribution located at ES = 0.0 with scale = .707") ) ) That's because t…
Comment by andersony3k June 2019
• Dear MSB, Thank you. I think you are right and that I was quite wrong. But what this as done is help me better conceptualize and describe what may now be stronger reasons to backslide. I think that my misconception stemmed from what might be seen as…
Comment by andersony3k June 2019
• But the variances would not approximate normal distributions, which would be problematic for a t test. Wouldn't the most straight-forward, well-established way to do this be to simply run a repeated-measures ANOVA with Condition (with levels X, Y, a…
Comment by andersony3k May 2019
• jploenneke Yes. Another way to put it is: Levene's test is just a oneway ANOVA on dispersion scores (absolute deviations from the group mean) that indirectly tests the equality of variance in the un-transformed scores. So a Bayesian one-factor ANOVA…
Comment by andersony3k May 2019
• OK. Thanks! I now see how the RUN IF functionality, combined with sequence-within-a-sequence, can indeed randomly order the blocks/inner-sequences. So it's not necessary to implement the feature I requested. Perhaps adding a section on it, in the do…
Comment by andersony3k April 2018
• And the tree representation would look like this: (Image)
Comment by andersony3k April 2018
• Thanks. That solution had not occurred to me with regard to randomizing the order of different sequences (as opposed to randomly selecting which sequence(s) to run or not run). But suppose I have inner_sequence_1, inner_sequence_2, inner_sequence_…
Comment by andersony3k April 2018
• I believe Vesna's question was about randomizing the order in which the blocks are presented (where each block includes its own loop of trials), not randomization within a loop of trials.
Comment by andersony3k April 2018