Import from SPSS - Continuous variables & levels

[Janusko]

Hi,
for the first time in my career I have hope that there is a stats programme that does not make my students give up before they start (unlike SPSS).

Just one question though: When importing data from spss, for continuous variables the levels do not copy over (e.g. likert scale, 1-5, 1= strongly agree etc.).

Is there a way around that?

Can I batch copy levels to multiple variables at once? It would be impractical to manually enter these for a large data set for each variable.

And if I leave the levels empty, how does that affect my analyses? Will I still get meaningful results tables?


I know this has been discussed somewhere else in the past, but I wonder if there is an update, and whether you have some practical advice on how to do that.

Thanks a lot in advance.

[jonathon]

hi,

> When importing data from spss, for continuous variables the levels do not copy over (e.g. likert scale, 1-5, 1= strongly agree etc.).

so the behaviour is that continuous variables *with* levels, will be imported as ordinal variables (because continuous variables shouldn't have levels, so they're probably supposed to be ordinal), and the levels should come through intact. if that's not the case, then provide us a .sav where this doesn't happen, and i'll take a look.

> Can I batch copy levels to multiple variables at once?

yes, with some caveats. you can select multiple columns at once (using shift or ctrl), and add labels to levels the way you normally would, and *all* the selected columns will be adjusted in this way. the caveat is that only levels that all selected columns share can be modified in this way. ... i might chat to damo about this, we might be able to improve it somewhat.

for example, if 3 columns contain the values 1, 2, and 3, but a fourth column *only* contains 1 and 2, then you won't be able to change the label for 3 if you have all four columns selected.

> And if I leave the levels empty, how does that affect my analyses?

so nominal and ordinal variables *always* have levels in jamovi, and continuous variables *never* do (which makes sense), so there shouldn't be a problem.

let me know if you have any more questions.

cheers

jonathon

[Janusko]

Great, thanks, that answers it.
I think if there was an easy way to batch copy, it would make integration from spss much easier.

[jonathon]

i assume selecting multiple columns at once, and then changing the level labels would achieve what you want with 'batch copy'?

with thanks

[Janusko]

Yes that would do it. Cheers

[jonathon]

so that *mostly* works at the moment ... it's just where a level is missing from one of the columns which is a problem.

cheers

jonathon