Timeline for Choosing between nested and crossed design
Current License: CC BY-SA 4.0
11 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 23, 2021 at 18:20 | comment | added | usεr11852 |
Absolutely, yes. I would suggest looking at something like simr in more detail for that matter. (I will try to write all this in an answer over the weekend.)
|
|
| Apr 23, 2021 at 17:15 | comment | added | Green44 | Thanks again. This all makes sense. Getting off topic a little, but would you recommend using simulations to calculate sample size? We have no pilot data, and there aren't really similar studies to go on. Some people in my sub-field seem to just guess the sample size in this situation. I'd like to avoid guessing, if possible, though. | |
| Apr 23, 2021 at 2:49 | comment | added | usεr11852 |
No problem at all. I would say yes, if we would think that there might be a framing effect but no, if we think that it won't (be nested). Ultimately the distinction being having "nested" or "crossed" effect would mean distinguishing between using the error structure (1|item_pairings) + (1|item:item_pairings) and (1|item_pairings) + (1|item) respectively. Do note that: (1|item:item_pairings) is the same as (1|item) if item is coded uniquely across item_pairings. Conceptually if we assuming no framing effects, then it's not uniquely coded.
|
|
| Apr 23, 2021 at 2:21 | comment | added | Green44 | Great. Thanks. Just one follow-up question. If each item appears in multiple pairs (as per the 2nd design), that still counts as a nested design? I was thinking that to count as a nested design, each item would have to appear in only 1 pair. Apologies if my basic understanding of the nested/crossed distinction is falling short here! | |
| Apr 23, 2021 at 1:50 | comment | added | usεr11852 |
Cool. That's great then. I was thinking you wanted the first design because of some small sample restrictions. In that sense then, the second design would be an obvious choice for me. And right, now that I think of it I understand why participant/item was not picked initially. Yeah, not using probably won't be harmful, if anything it can cause identification issues so dropping it (as you initially did) is probably better.
|
|
| Apr 23, 2021 at 1:40 | comment | added | Green44 | Thanks, that's very helpful. I take your point about the 15-pair option being more robust. I haven't determine the sample size yet, but we can definitely run as many as 1500 participants. | |
| Apr 23, 2021 at 1:17 | comment | added | usεr11852 | Finally the first design case seems to me suboptimal as certain participants won't see a particular item (or item-pairing) by design so how will we distinguish from say item 1 being great (getting high scores all-round) and participant 1 being affluent with her scoring (giving high scores all-round) if item 1 is say in the 1-2 item-pairing and participant 1 is within that pairing. The 15 pair option seems much more robust. | |
| Apr 23, 2021 at 1:15 | comment | added | usεr11852 |
Based on what is described I would assume that the nesting participant/item is relevant too. And also, so many random slopes... are you sure this won't lead to model identifiability issues? At first instance mod <-lmer(DV~ 1 + P1 + P2 + ... + P8 + (1|Participant/item) + (1 |item_pairings/item), data =mydata) should be adequate for the second design option.
|
|
| Apr 23, 2021 at 0:48 | comment | added | usεr11852 | How many participants are expected? I am thinking that inadvertently we might end up with a crossed design, despite recognising we have a potentially nested design, just because certain items are not replicated enough. | |
| Apr 22, 2021 at 16:59 | history | edited | Green44 |
edited tags
|
|
| Apr 21, 2021 at 23:33 | history | asked | Green44 | CC BY-SA 4.0 |