Error messages and other issues
Hi Jasp team,
I am a great fan of the product. Keep it up. I have a few questions and issues with it though.
First, a bug report:
This analysis terminated unexpectedly.
Error in if (is.finite(value)) {: argument is of length zero
Stack trace
analysis(dataset = dataset, options = options, perform = perform, callback = the.callback, state = oldState)
.rmAnovaWithinSubjectsTable(dataset, options, perform, model, stateSphericity, status)
.clean(result[index, 'F'])
This is shown in the results window when I am trying to run a non-bayesian 4x2x2x5 repeated measures ANOVA from .csv file data (which I can upload if necessary to reproduce the error). This error only occurs after I delete one of the factors and add a new one after I have already drag&dropped variables in the RM cells window.
Second, I noticed that when adding my fourth factor (with 5 levels), the repeated measures cells window gets messed up. More specifically, the fourth level is shown as the third when hovering over the individual cells. This can only be fixed by deleting a factor and adding it again (which then causes the error described above).
Third, I have some issues with drag&dropping variables into the repeated measures cells window. When marking all and drag&dropping them, it doesn't do so in the order that is shown left to the cells window (i.e. in the part where you select variables for drag&drop). While the order for some variables is rather specifically messed up - e.g. the 6th in the left window is shown as the first in the RM cells window - for others, it seems that they are ordered by alphabet. The first is really frustrating, while the second is just very unintuitive. I think it should just maintain the order it had in the left window.
All these issues occur for v8.5 as well as for the most recent version of JASP.
I also wonder whether it will be possible in a future version to change the size of the individual sub-windows/sections (i.e. RM factors or RM cells windows). They are awfully tiny and clumsy to work with when (especially because there is no horizontal slider) in most cases.
Finally, is there already an ETA/news on when we will be able to edit the data in JASP rather than before loading it into JASP?
Thanks
Comments
Dear Michif,
Thanks.
1. As far as data editing is concerned, see https://jasp-stats.org/2018/05/15/data-editing-in-jasp/.
2. Resizing those windows is a good feature request.
3. The bug in the RM ANOVA is interesting; I'll direct the team to it. Note that we prefer to handle bug reports and feature request through our GitHub page; this way you can communicate directly with the programming team. See https://jasp-stats.org/2018/03/29/request-feature-report-bug-jasp/.
Cheers,
E.J.
HI Michif,
I just tried to reproduce your bug in the build version of JASP, and could not reproduce it. It is probably solved in the next release of JASP, since we improved some state issues, but I will check back with you after it is released to see if this is indeed the case.
Kind regards,
Johnny