So... help me understand stats

[CanIBeAnonymous]

Incoming M1. I want to hit the ground running with clinical research, and I'm quite proficient in 'R' which I feel will come in handy. So I've got a lot of experience exploring and analyzing data as directed, and making some pretty graphs, but as directed. I want to get a good understanding of stats so I can be self-reliant, and really understand what it is I'm doing. I imagine this would make it easy to propose my own questions after an exploratory analysis.

So - any resource yall can recommend for me to use over the next 3-4 weeks before M1 starts to get a solid ground in statistics, preferably specifically geared toward the type of clinical research med students would normally do. If it matters, I plan to continue using R. It's just such a powerful tool, and I don't want to waste time learning something new during med school.

Also it's cool if it goes "overboard." I'm very interested in data and plan to continue learning more about this stuff throughout med school and beyond.

---

Members don't see this ad.

 

[ArteryStudyPainting]

Not research per se.

But.

The highest of yields, is a two part stat video on YouTube. Run by a doctor. You need to watch that, and life will be so much easier.

 

[CanIBeAnonymous]

Not research per se.

But.

The highest of yields, is a two part stat video on YouTube. Run by a doctor. You need to watch that, and life will be so much easier.

Thanks! Do you have a link to the video directly? My google-fu seems to be weak this morning.

 

[ArteryStudyPainting]

By Randy Neil.

 

[CanIBeAnonymous]

By Randy Neil.

Oh I'm mad dumb, that popped up too when I was googling. Thanks!

 

[OnePunchBiopsy]

A quick and dirty way to learn stats is learning what “tests” best determine if differences in your data are significant. Learn when to apply them, and when you can’t apply them.

Stanford has a quick guide that outlines the majority of tests I use for my research, which consists mostly of retrospective chart

https://web.stanford.edu/~kcobb/hrp259/correlateddata.pdf

 

[CanIBeAnonymous]

A quick and dirty way to learn stats is learning what “tests” best determine if differences in your data are significant. Learn when to apply them, and when you can’t apply them.

Stanford has a quick guide that outlines the majority of tests I use for my research, which consists mostly of retrospective chart

https://web.stanford.edu/~kcobb/hrp259/correlateddata.pdf

This is exactly what I'm looking for, thanks!