search

LSC-MEM-course / random-effect-simulation-assignment-1-rebeccabieber

random-effect-simulation-assignment-1-rebeccabieber created by GitHub Classroom
0 stars 0 forks source link

Not a very important issue, probably #1

Open rebeccabieber opened 4 years ago

rebeccabieber commented 4 years ago

Just working my way through this assignment (finally), and not sure how to answer the extension exercise question 1.e - how do I interpret the size of the random effect variance in my data? Maybe you all already talked about this in the class that I missed.

shoestringpsycholing commented 4 years ago

Write up some code to get you that far (namely, running a simple lmer on your own data), and then look at the summary output. Interpreting it depends on how big it is...

rebeccabieber commented 4 years ago

Random effects: Groups Name Variance Std.Dev. Subject (Intercept) 0.01671 0.1293
Residual 0.13529 0.3678
Number of obs: 8400, groups: Subject, 56

Not sure what to make of it, though. I think it's relatively small, meaning that subjects aren't very different from one another?

scottrjackson commented 4 years ago

What do the coefficients look like (fixed effects)?

rebeccabieber commented 4 years ago

Fixed effects: Estimate Std. Error df t value Pr(>|t|)
(Intercept) 5.948e-01 1.913e-02 7.454e+01 31.086 < 2e-16 Block 2.509e-03 4.637e-04 8.343e+03 5.411 6.44e-08

rebeccabieber commented 4 years ago

Fixed effects: Estimate Std. Error df t value Pr(>|t|)
(Intercept) 5.948e-01 1.913e-02 7.454e+01 31.086 < 2e-16 Block 2.509e-03 4.637e-04 8.343e+03 5.411 6.44e-08

rebeccabieber commented 4 years ago

Sorry - ugly formatting. But also pretty small. So is it more about the relative amount re: fixed effects?

scottrjackson commented 4 years ago

So your effect of block is only 0.0025, and subjects vary from each other in their intercepts with a standard deviation of 0.129, so...