And shoutouts to @shannonsauerzavala.bsky.social and the TIPS Lab crew for being patient with this paper-in-a-year-and-a-half! 🥳

It was just accepted at Psych Assessment today so the published version should be available soon.

So if you want to capture between-person differences, use any instructions but if you might want to capture within-person differences, matching the instructions to your timeframe is probably the way to go.

Feedback, comments, and questions very much welcome!

We think focusing people on a consistent timeframe improves their within-person reliabilities but keeping other aspects of the measures consistent ensures we're still assessing the same personality construct. 👯

Within-person relations with anxiety, depression, quality of life, & functioning:

✅ Generally stronger for weekly instructions (as expected)

osf.io/gznav/files/...

Within-person means & standard deviations:

Weekly instructions condition:
⬇️ Somewhat lower means (as expected)
⬆️ Higher standard deviations (as expected)

Are they measuring the same underlying personality construct?

✅ Generally, yes!

Although model fit significantly improved for some measures when we let the 2 instruction conditions differ, these improvements were fairly modest & mostly with the NEO (general personality).

Test-retest reliability:

✅ Generally .70+ in both conditions, although typically higher in the general instructions condition

Within-person internal consistencies:

✅ Adequate-to-good in the weekly instructions condition

❌ Poor-to-questionable in the general instructions condition

Between-person internal consistencies:

✅ Excellent and generally the same regardless of randomization condition

We randomly assigned 257 University of Kentucky undergrads to complete 6 weekly personality assessment batteries that either:
(a) asked about their personality in general or
(b) over the prior week

If we're assessing personality when we think it might change (like in treatment), asking people repeatedly how they are in general might get weird 😵‍💫

If I've only had a week of therapy, my personality can't have changed much *in general* but might be shifting a lot.

Most personality measures ask people how they are *in general* or don't specify a timeframe for people to consider, which makes sense when we want to capture how people are in general but do so at one timepoint.

For example, the BFI-2: www.colby.edu/wp-content/u...

We know personality changes over our lifespan (h/t @dingdingpeng.the100.ci & @ajwright.bsky.social), but what about the granular changes contributing to it?

Or when we want to test for personality changes in treatment? psycnet.apa.org/fulltext/201...

Can we just use the same measures for each?

Please check it out, share the post with anyone you think might be a good fit, and absolutely reach out via DM or email with any particular questions!

I'll leave it open as long as folks need, though the first deadline we have listed is Oct 2.