I've run a multi-tasking behavioural paradigm with volunteers grouped according to 2 characteristics 1) Age (young, mid) 2) Risk (high, low). The multi-tasking paradigm asked volunteers to complete an attentional paradigm divided into 3 blocks of varying difficulty (lets call them a, b, c), alongside a concurrent working memory paradigm. I've analysed the data for the attentional task, but am bit stuck with how to analyse the WM data.
The WM paradigm asked participants to count how many times they saw an obscure target over the course of the attention block, and so for each block I am left with a binary score of were they correct or incorrect. I've converted this into a percentage e.g. 30% of all mid-age high risk volunteers got it correct, compared with 70% of all young low risk volunteers.
The data looks really interesting when eyeballing it, but I wanted to know if there was a way to include attention block (a, b, c), age (mid-young), and risk (low-high), and potentially model the interaction in looking at the percent who got it correct? So is there a difference between the groups in the proportion who got it correct, and does this differ across attention blocks.
So far I have tried a log linear regression but wasn't sure if this was the best way to tackle this problem.
Thanks for all your help and please let me know if anything is unclear.
Advertise on Talk Stats