EJ
About
- Username
- EJ
- Joined
- Visits
- 2,532
- Last Active
- Roles
- Member, Administrator, Moderator
Comments
-
Dear biopsy, The analysis assumes that the order with which the data arrive is given by the order of the rows. So the analysis is executed on the first n rows, where n is increased until the last row has been reached. The order with which the data …
-
Dear FLPS, The interpretation is straightforward: nothing much is going on. The error % provides an indication of the numerical error of approximation relative to the size of what is estimated. The BF_M term compares prior model odds (0.063/(1-0.063…
-
Dear HTIDCam, It might help us if you add the .jasp file and/or a screenshot of the error message. I will bring this to the attention of one of our team members. Usually Johnny would handle this but he's on vacation, so you might have to wait a few…
-
Hi isakohrlund, From the perspective of Miller & Chapman, the problem cannot be overcome -- it is an inherent confound. I believe Pearl had a different opinion, but there's no easy fix. So at the very least it's prudent to issue a caveat emptor…
-
Hi Isak, JASP assumes that every subject/unit is a row. So it assumes a "wide" format, not a "long" format. I've suggested on GitHub that we increase the priority level of including support for the "long" format. The wa…
-
We're working on it (or I should say, a student is working on it :-)) E.J.
-
Hi mrdata2017, I am not 100% sure I understand the question, but if you double-click on the csv data, your favorite csv editor will open -- just create a fourth column that equals the sum of the first three, press "save", and the JASP data…
-
Yes, I think you're spot on. Cheers, E.J.
-
Susanne, on GitHub you said that you resolved the issue. What was the cause of the discrepancy? Cheers, E.J.
-
Hi Bill, Thanks. I will look into it, but I can already tell you that this will most likely be an issue with an R package. E.J.
-
Courtesy of Erik-Jan van Kesteren, the underlying R code: v <- plotDat[plotDat$group == level,]$yquantiles <- quantile(v, probs=c(0.25,0.75))IQR <- quantiles[2]-quantiles[1]plotDat[plotDat$group == level,]$outlier <- v < (quantiles…
-
I am not sure what outputs to compare, exactly, but I googled this problem and it has been noted a few times before. From what I see, most of the time the issue is rounding (!). For instance see https://stats.stackexchange.com/questions/113314/spss-…
-
JASP uses the standard R code for boxplots. Typing ?boxplot in R gives the specifications. Outliers are values that extend beyond the whiskers. The whiskers depend on the argument "range". From the R documentation on "range": &qu…
-
Hi Toc, I think Sebastiaan was not beging serious when he proposed "Baws" as a name. You might just want to describe what models you were averaging over to compute the inclusion probabilities. Oh boy, we do need that manual and some decen…
-
At least (c) seems the other way around E.J.
-
Dear cpaeye, Thanks! H0 is the model without any predictors; H1 is the model with predictors (which predictors are in H1 is indicated per row, so each row is a different H1). Some examples and explanations here: http://pcl.missouri.edu/sites/defaul…
-
Thanks. We are working on it! E.J.
-
Hi Eniseg2, The next version of JASP will have the option to put the best model on top as a comparison model, which will make all of this easier. With respect to your text, my comments inline: "Table 11 shows that the model which was most str…
-
Hi Tcarsten, I have not tried MorePower 6.0, but I will say that for RM designs, there is a more recent paper by Masson that seems relevant: https://web.uvic.ca/psyc/masson/NM16.pdf In general, I think the problem with the model is twofold. First,…
-
Hi Inge, From memory, yes. But if you want to know for sure you should drop Richard Morey an Email. Another way to check is to consult the BayesFactor documentation, or run the same analysis in JASP and BayesFactor. It is always best to check for y…
-
Hi Duplex, I'll inform the team member who created the factor analysis. There were some earlier suggestions to document this better and expand on the functionality. I'll discuss this. Cheers, E.J.
-
Hi Spi, This is my take on the issue: interactions are inherently complex --they can take on all kinds of shapes-- and this means that they spread their predictions out relatively widely; this entails a relatively strong penalty for complexity. Mor…
-
Hi Bill, I can have someone look into this, but it would certainly help us if you could provide a specific example so we have some concrete numbers to work with. Cheers, E.J.
-
Hi Charles, Good question. I'll attend Richard to it. Cheers, E.J.
-
Hi Inge, The ANOVA model is not so simple. I advice you to start with http://www.ejwagenmakers.com/inpress/RouderEtAlinpressANOVAPM.pdf We are working on a JASP Manual that will explain these in a language that is perhaps easier to understand, but t…
-
Hi Eniseg, Best to take this issue to the GitHub pages, where you can interact directly with the programming team. We hope to release a new version soon, and if this is a bug we'd love to fix it asap. Cheers, E.J.
-
Hi Guillaume, In Jeffreys's opinion, BFs in between 1/3 and 3 are "not worth more than a bare mention", and I agree. So that BF01 = 1.951 really doesn't tell us much. The effect of group is also not really compelling, so I would not lose s…
-
Hi Guillaume, When you put "Time" and "Group" as nuisance, you are basically comparing the full model to the model with two main effects; that is, you are asking whether the inclusion of the interaction adds anything compared to…
-
Good points. The 95% CI is an inferential measure and we didn't feel it belongs in descriptives. But I'll discuss this with the team. Making the "split by" option more general is highly desirable, we agree. E.J.
-
These are all good points, I will pass them on to the people responsible. I think we want to stick with "log" instead of introducing "ln" (see also how R does this). Thanks for helping out with these great suggestions! Note that …