for those of you who asked about licensing - the book is now licensed under CC-BY-NC. I don't love the NC but it's necessary for a commercial publisher to consider printing a hardcopy version.
Show this thread
My new open undergraduate statistics textbook is now available at http://thinkstats.org/ - more context at http://www.russpoldrack.org/2018/11/statistical-thinking-for-21st-century.html …
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I love this! While brushing over it I noticed the regression to the mean subsection feels missplaced. You write General Linear Model but restrict yourself to the linear model - also did you mean generalized linear model? Fig. 15.4 is likely missing aes(group(subject))
-
for this course I'm sticking to simple general linear model. perhaps one day there will be an advanced version that also covers generalized linear models...
-
That makes a lot of sense about the generalized model. But I'm probably missing something, I thought the general linear model is the multivariate version (multiple dependent variables) of the linear model?
-
in my head (and this is the way the concept is used in the neuroimaging literature) a general linear model is Y = X*B + E where Y is a Nx1 vector and X is a NxJ design matrix with J regressors. so univariate.
-
Ok now I understand. According to Wikipedia glm is the multivariate case, and we usually have multiple voxels/electrodes so multivariate, which are then tested univariatly. I think massive univariate linear model would be the way better name to not confuse with true multivariate
-
At some point I tried finding ""true" general linear model analyses and everyone confused everything (i.e. with generalized) and multivariate is confused with multiple - what a definition mess :(
End of conversation
New conversation -
-
-
Hey, thanks for making this book! I have a suggestion for the suggested reading in the DataVisualization chapter which is this free online book: https://serialmentor.com/dataviz/ . Also, a minor note: Chapter 18 is not showing the references but they are shown in the next chapter.
-
thanks! yes, that's some funkyness that I needed to make the latex version compile properly. will try to fix it soon.
End of conversation
New conversation -
-
-
Russ, this looks amazing. Can't wait to use it as a resource for my data analysis class this spring.
-
great! please send along any suggestions/corrections you come up with
End of conversation
New conversation -
-
-
Thanks so much for this! Love to see the disdain for NHST and shout-out to
@lakens course on improving statistical inferences. Was surprised to see Meehl not mentioned. IMO philosophy of science needs to accompany early statistics training. -
thanks! yeah, would be great to work Meehl into the mix.
End of conversation
New conversation -
-
-
This looks great, I love all the R examples. On Figure 2.1, the "valid but not reliable" example seems misleading, since lack of reliability precludes validity (always, right?). That said, the targets example is very clear and I have used it before myself.
-
well, one can in principle have a measure that is valid but not reliable in the test/retest sense, because there is too little between-subject variability - see https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5990556/ … - but that's probably too deep in the weeds for this book!
End of conversation
New conversation -
-
-
This looks great. Would you consider giving it a clear creative commons licence so it can be used elsewhere? Apologies if I missed it. https://creativecommons.org/licenses/
-
I will be assigning a license soon, once I figure out how it interacts with potential publication of a hard copy version
End of conversation
New conversation -
-
-
What service/framework did you use for the online styling and pdf generation? Looks great and I'd love to use the same simple format for a few ideas I've had over the years. Thanks for the neat and OPEN resource!
-
I wrote it in RMarkdown using Bookdown https://bookdown.org/yihui/bookdown/html.html#gitbook-style … and
@rstudio
End of conversation
New conversation -
-
-
I'm still working on reading it and writing comments....not sure how quickly I'll get to it :)
-
thanks! the version that I just posted has lots of changes since the earlier draft I sent you.
End of conversation
New conversation -
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.