EJ
About
- Username
- EJ
- Joined
- Visits
- 2,531
- Last Active
- Roles
- Member, Administrator, Moderator
Comments
-
Hmm this is really not my domain -- I just know of the papers that critiqued the idea. I'd do some Google'ing and feel free to report the results of your search here! Cheers, E.J.
-
Hi Tom, That is really strange and almost seems more like a bug. I'll attend the team to this but you could also post this issue on our GitHub page (for details see https://jasp-stats.org/2018/03/29/request-feature-report-bug-jasp/) Cheers, E.J.
-
I think you basically do a power analysis using the maximum likelihood estimator as the parameter value under H1, with the same planned sample size you actually had in the experiment. I don't think you can do this in JASP (because I would have put m…
-
Hi Raquel, This project is still marinating...I will put the pressure. If you post this as a feature request on our GitHub page, the people working on it will see it too. Cheers, E.J.
-
Hi Per, I will ask the team. Also, if you have any specific suggestions please let us know (maybe best to do this on our GitHub page, so the team won't forget!) Cheers, E.J.
-
I assume this is because the test is based on a pooled standard error, but I'll ask our expert to confirm this... Cheers, E.J.
-
Hi JT, When you say you want the "amount", I assume you'd like to standardized measure of effect size? -- because you could of course just report the cell means. I am generally a bit cautious about interactions that do not cross, because …
-
What was your t-value and what are the sample sizes? And what was your prior? E.J.
-
Reporting post hoc power is a terrible idea, as the number is just a transformation of the p-value. You can quote me on that. For a critique see Hoenig, J. M. and Heisey, D. M. (2001). The abuse of power: The pervasive fallacy of power calculations …
-
I think this issue has come up before, let me ask the team...
-
You probably intended to post this on a different forum. This is the place to discuss the interpretation of statistical results from JASP. Cheers, E.J.
-
Hello Tong, Almost correct -- the interpretation of the BF is that the data are about (1/0.272) time more likely under H0 than under the model with only cue_type. For examples see https://link.springer.com/article/10.3758/s13423-017-1323-7 https://w…
-
I do not immediately see how a quadratic contrast would be set up to find a U but not an inverted U. You could try generating data according to an inverted U (or take U-shaped data and transform it to give an inverted U shape), apply the software, a…
-
Early June. E.J.
-
Hi Per, This is why the Library allows you to click the csv icon (to the right of the JASP icon), which just gives the raw data without any analysis or annotation. Or is this not what you mean? If you have a feature request, it is best to use our Gi…
-
We don't produce the path diagrams. That certainly would be very nice to add. Cheers, E.J.
-
Correct E.J.
-
"Am I right to say that the cond model is the best model and adding presence into the model weakens it?" Yes. Second, "my concern here is whether I should be comparing cond + presence model with presence model (as per what I wrote in…
-
Yes, the same scale applies. Cheers, E.J.
-
Hi Tali, I am not aware of it, but I would advise you to check out our GitHub page (and, if you can't find it, post the issue there). Cheers, E.J.
-
Hmm I am not 100% sure what you would like. I assume that with "delta value" you mean the posterior point estimate of effect size? If so, see my earlier response. If not, could you clarify? Cheers, E.J.
-
The easiest way to check whether you did this correctly is to add "sense of presence" to the null model (under the tab "Model"). You might also want to tick "compare to best model" and "BF01" to the entries in…
-
Not directly. We are looking to implement blavaan (but it is not there yet); you could code the model in our JAGS module, but it is not yet there as a drag and drop option. It would be a good feature request to increase the pressure to include this …
-
Dear mlwilson, This is a topic for our GitHub page. If you post this issue there, you will have direct access to the JASP programming team. In order to diagnose the problem and fix it, they may need some more information. Cheers, E.J.
-
This would be a good feature request for our GitHub page. The more precise you can make your request the higher the probability that it will be implemented quickly (i.e., show a specific example, refer to a textbook, mention an R package). Cheers, E…
-
We are using the scientific E-notation for large numbers (https://en.wikipedia.org/wiki/Scientific_notation#E_notation). So 2.961 e+6 is shorthand for 2.961 * 10^6 = 2,961,000. If you use the log BF, you have to exponentiate, so compute exp(14.901).…
-
I'm not sure that is more enlightening, but you can make a feature request on our GitHub page so the team can debate this! Cheers, E.J.
-
The directional tests make sense only when you have a 2x2 table and the interest is in comparing two proportions. The data for these two proportions can be collected in several different ways (poisson, multinomial). Cheers, E.J.
-
I think the same advice holds here as well: you could send Erik-Jan an email. Of course it would be great if he could share the "solution" on this forum to avoid getting these emails on a regular basis :-)
-
Hello Nico, These options define the model to which all other models are compared. The fundamental information in the table does not change, however. Because the null model is often very poor, it is usually more interpretable to compare all models a…