By the way I do have a bias in reviewing studies which is when scientists disagree on data interpretation I look for which interpretation...
-
-
@avantgame I need to double check that they didn't use Spearman or other ranked measures, but I don't think they did.#badproblem -
@Wikisteff this is really helpful and I have understood at least half of what you said :) if I had to answer a question about this study to
End of conversation
New conversation -
-
-
@avantgame I'm also a bit miffed at the weak sig. limits. If they say a p-value, we can correct for Bonferroni. A cut at p<0.05 is unuseful.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@avantgame If the test will only confidently find signals when comparing the limits, it won't get a lot better when you add the middle too.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@avantgame Incidentally you can tell that the hours played data are nonlinear, by noticing that the hours played SD > their means.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@avantgame Especially given the nonlinearity of "hours played per week". :)Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@avantgame They could only conditionally detect significance in the two-way test at d'~=0.5.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@avantgame 2. Their study was underpowered, obviously. The extreme-value test could only easily find significance with effect size >~1Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@avantgame This throws a lot of correlational analysis into qn, as linear analysis with an exponential variable'll be dominated by outliers.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@avantgame 1. Hours played is not a normally distributed variable; probably it's some sort of exponential variable, with a cap at 140/hr/wk.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@avantgame Okay, midway through Experiment 1, and some issues are already clear…Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.