EJ
About
- Username
- EJ
- Joined
- Visits
- 2,598
- Last Active
- Roles
- Member, Administrator, Moderator
Comments
-
Sorry about the delay! The "e" is scientific notation (https://en.wikipedia.org/wiki/Scientific_notation, see E-notation) for large numbers. So 2.12e+11 means 2.12 and then move the period 11 positions to the right, so it is 212000000000. …
-
Hi eniseg2, I am not sure about the discrepancy -- when the prior inclusion plot suggests a small increase in probability, the inclusion BF suggests weak evidence in favor of adding that component. There is nothing special about a BF value of 3. If …
-
Hi JC, Sorry about the tardy response, this one slipped through. First of all, there is no way to compare BFs directly. Instead, you need a BF for the difference. So what you'd need to do is combine the two conditions into one analysis file, and tes…
-
If JASP recognizes the variable as nominal, it suggests some entries (at least one) has a character string. This could be an undefined "not a number" indicator, or perhaps the Excel version you used denotes fractions with a comma: that is,…
-
I'll forward this to Richard
-
Hi roundcircle, Well, you have B01 (from R or JASP), and you can easily compute the numerator, so that then determines the denominator. However, (I have not checked this) I'd be surprised if that equation shows the two marginal likelihoods -- I susp…
-
Hi MSB, "r" is the sample value. The underlying model parameter is "rho". Cheers, E.J.
-
Hi AnaOnar, Nothing is wrong. This procedure uses numerical integration and therefore the results will be slightly different each time. For enhanced reproducibility, the next version will include a random seed that you can set in order to reproduce …
-
Hi WWU, Ah, I see. Strange. Well we do run SEM to do the CFA but I'll pass this on to Erik-Jan (with a k) who might have bright ideas. Cheers, EJ
-
Dear WWU, This warning suggests that the model is overparameterized. I would try a simpler model first, and see whether that works... Cheers, E.J.
-
Hmm yes, that is quite categorical support for H0. This is usually difficult to obtain. Generation has 7 levels? The more levels there are, the easier it will be to get evidence for the null I think. Still, this is pretty extreme, especially because…
-
Can you give a concrete example?
-
I'll pass this on to our lavaan expert!
-
Hi Sophie, When you do an independent samples t-test, the sequential BF will remain at 1 until you feed it data from the other group. Without having seen any data from the other group, there can be no evidence of an effect of the manipulation. So if…
-
Hi Chribo, These tests are based on different data (DVs) so they can't easily be combined, afaik. I'd make absolutely sure that the direction of the effect has been coded correctly -- to get BF+0 = 0.0001 the effect must go clearly in the direction …
-
Hello irepet, I think it would help if you showed the results for a specific example, this would make the discussion more concrete. Cheers, E.J.
-
Hi Gabriel, The not-so-subtle statistical practice in experimental psychology would be to average the 8 time points for each participant and conduct the regression/correlation on the averaged numbers. A more subtle approach would be to apply some so…
-
van Doorn, J., van den Bergh, D., Boehm, U., Dablander, F., Derks, K., Draws, T., Evans, N. J., Gronau, Q. F., Hinne, M., Kucharsky, S., Ly, A., Marsman, M., Matzke, D., Komarlu Narendra Gupta, A. R., Sarafoglou, A., Stefan, A., Voelkel, J. G., &…
-
I'll ask Don van den Bergh, who programmed this
-
Well, JASP basically uses Lavaan code, so when you know the correct model syntax for Lavaan, you can simply copy-paste it and it should work. E.J.
-
That's a good suggestion, I'll pass this along. Cheers, E.J.
-
Well, that depends. One could argue that for a subjective Bayesian it does not matter what other hypotheses you did or did not test, the evidence is simply the evidence. However, the fact that you were testing several hypothesis might indicate that …
-
Hi WWU, Can you provide more information please? (description of what exactly you want to do, concrete example, url with background information, possibly R code). Cheers, EJ
-
What are the t-test supposed to correct for? EJ
-
Hi Gabriel, Although it is possible for test a four-way interaction, with realistic sample sizes it would have to be a whopper to detect it. What we hope to do in the near future is allow the user to specify the particular kind of interaction (i.e.,…
-
Dear WWU, We are currently in the midst of revamping the Reliability functionality, so if you post this issue on our GitHub page we can either tell you we already have it, or it could be added in the future. (for details see https://jasp-stats.org/…
-
yes, although the labels are only heuristic guidelines. I would not bet my house on such evidence. One way to interpret the number is that if H0 and H1 were equally likely a priori (0.50 for H0) a BF of 3.73 gives a posterior model probability for H…
-
Hi Elena, Gabriel's link is the one I'd also point to for design planning; Given that the data are already in, all that matters is the evidence. So I agree with you that, from an inference perspective (not from a planning perspective) power analyses…
-
Hi Philip, That's not easy and its pretty impressive that you got it to work. I'd recommend to take a look at the Scott & Berger papers that explain the procedure. Cheers, E.J.
-
Yes, so the data are 1/0.268 = 3.73 times more likely under the two main effects model than under the model that also includes the interaction. JASP uses the principle of marginality, which means that if the interaction is included in the model, so …