Pavlovia datafiles in DB misaligned and conditions not saved

[wendyross]

I wonder if anyone can help me save my data! I have three problems with my pavlovia output - some might be salvageable - others I suspect not.

1. The conditions were not saved despite having explicit code to do so - I can recreate these off the participant numbers but I would like to be able to understand why they weren’t.
2. The columns seem to be misaligned in a non systematic manner so the data cleaning file I had built on a soft launch of 10 doesn’t seem to work - it is a complex design so I don’t now even know where to start.
3. (This is probably not salvageable) I randomised my conditions (6) by dividing the participant number by 6 but this has led to big differences in my groups when I have gone back and tried to recreate the conditions. I guess that participant number is random rather than sequential as I assumed?

Any thoughts would be greatly received - I had my settings as database but if you even know where the raw csv files are stored, I may be able to save this!

[wakecarter]

Hi Wendy,

1. What code did you use to save the condition? Where did you do this in comparison with your code to allocate the condition.

2. This sounds like a Excel file opening issue. Have you inspected the CSV file in a text editor?

Download excel - damaged?

Hi all, I am having problems when downloading results from pavlovia, as I got an excelsheet but is like damaged, I cannot see the columns properly, not separated. A colleague did that in her computer and it works fine for her, so data is stored, but we don’t understand why that’s not the case when I do that. I download it after selecting “CSV” but a zip file is downloaded with the excelsheet in it, and I got that problem. Could you help me? Thanks so much

3. I’m surprised you haven’t already noticed that participant numbers are random by default. You could use one of my tools.

Automatically incrementing participant number?

I’d like my experiment (run on pavlovia, interfacing with sona) to automatically assign a participant number, based on how many participants have completed the experiment. In other words, I want the first participant to be assigned participant number 1, and the next participant to be assigned participant number 2 (unless participant 1 failed to complete), and so on. I understand that it’s possible to incorporate participant IDs into links from sona, but unless I’ve missed something, those have t…

New web app. VESPR Study Portal

I’ve just written an app to allow researchers to host their PI sheet and debrief on my site (written in markdown), which has the advantage of a download link for the PI sheet before and after the study. Participants are also able to withdraw their consent at the debrief stage which sends the researcher an email but the participant stays anonymous. Hosted studies will be listed at the link above (in random order) or an individual study can be addressed using a variable (e.g. https://moryscarter…

[wendyross]

Yes, I always used to use VESPR and it was brilliant and then Survey got better and I got complacent which is a lesson to me and anyone else reading this.

I’m not sure if it’s a CSV file issue. It is just some of the rows are overlapping into the wrong columns if that makes sense. For example, row 148 and 165 are all over the place as well as others. But some are in the right place.

df.csv (3.6 MB)