search

uchicago-computation-workshop / ed_awh

Repository for Ed Awh's presentation at the CSS Workshop (1/10/2018)
0 stars 0 forks source link

data and experimental design #42

Open RuxinChen opened 5 years ago

RuxinChen commented 5 years ago

Thanks for presenting! I am wondering how do you determine the sample size that suffices for convincing/valid results. I also notice that both experiments exclude some subjects and in experiment 2, the exclusion criterion was determined during the course of data collection. Was your original experimental design/procedure affected by this decision?

edawh commented 5 years ago

For studies in which the findings are very similar across different subjects, fewer subjects are required to determine the pattern of findings. The link between alpha topography and attention is super-robust across subjects.