brushing up on statistics- recommendations?

[biscuitsbiscuits]

I'm starting a doctoral program in September and I'd like to use this time to brush up on statistics so I can be better prepared. It's been over a decade since my last stats class and I want to make sure I don't drown in the first semester! I'm not too too worried because math has always come easily to me and I've been working in clinical research and using statistical principles since then. But I haven't personally done any "hard stats" in a long time, so I'm going to be rusty.

Any advice for books or online resources? I've started a stats overview on Khan Academy, but I'm worried it's too general/basic and not focused enough on social sciences.

In my first semester I'll be taking Statistical Design in Research; in semester two, I'll be taking Multivariate Analysis for Psychological Research.

Thanks!

---

Members don't see this ad.

 

[MCParent]

Discovering Statistics Using SPSS or Using R (depending on what your lab uses).

 

[singasongofjoy]

Andy Field's book- discovering statistics using SPSS- as MCParent mentioned. A cheaper, older edition would probably still be quite useful (though there is a good chance that your class will use this book- it's a very widely used one). I found it provided good explanations of why/when to use various stats tests. Also once you get into using SPSS and having to decide which analyses to run on actual data, Laerd Statistics is where it's at. You can pay for a month at a time (I think it's like $3 or $6 a month? and it is not one of those annoying automatic renewal things either, you can literally just pay for 1 month at a time and then forget about it when you're through using it) to access the full content - that website was a godsend for me at many points in grad school.

 

[NeuroLady]

Andy Field's book- discovering statistics using SPSS- as MCParent mentioned. A cheaper, older edition would probably still be quite useful (though there is a good chance that your class will use this book- it's a very widely used one). I found it provided good explanations of why/when to use various stats tests. Also once you get into using SPSS and having to decide which analyses to run on actual data, Laerd Statistics is where it's at. You can pay for a month at a time (I think it's like $3 or $6 a month? and it is not one of those annoying automatic renewal things either, you can literally just pay for 1 month at a time and then forget about it when you're through using it) to access the full content - that website was a godsend for me at many points in grad school.

Seconding Laerd Statistics. Amazing resource. Very much worth the few dollars.

 

[deleted343839]

I'm starting a doctoral program in September and I'd like to use this time to brush up on statistics so I can be better prepared. It's been over a decade since my last stats class and I want to make sure I don't drown in the first semester! I'm not too too worried because math has always come easily to me and I've been working in clinical research and using statistical principles since then. But I haven't personally done any "hard stats" in a long time, so I'm going to be rusty.

People taking their first-year graduate stats courses are all over the map in terms of conceptual understanding, so I wouldn't worry too much about brushing up. But if that's how you want to spend your free time, definitely find out what statistical package your PI/lab uses and become more knowledgeable about how to use it. If you can learn command syntax while working your way through basic stats, you can avoid becoming dependent on the graphical interface and use text commands for all or most of your analyses.

 

[singasongofjoy]

People taking their first-year graduate stats courses are all over the map in terms of conceptual understanding, so I wouldn't worry too much about brushing up. But if that's how you want to spend your free time, definitely find out what statistical package your PI/lab uses and become more knowledgeable about how to use it. If you can learn command syntax while working your way through basic stats, you can avoid becoming dependent on the graphical interface and use text commands for all or most of your analyses.

Yesssss.... if you learn to use syntax early on, you'll thank yourself later.

 

[ellenew]

Seconding Laerd Statistics. Amazing resource. Very much worth the few dollars.

What is it exactly?

 

[PsyDr]

Reread Friedmans grand conjecture.

 

[biscuitsbiscuits]

Great recommendations and advice, everyone, thank you! I bought a used copy of the Andy Field book and I'll look into Laerd.

 

[NeuroLady]

SPSS Statistics Tutorials and Statistical Guides | Laerd Statistics

 

[glowstick]

Obviously, some labs will strongly recommend a stats package – but it's also worth considering what the courses are taught in. i.e., if you're using SPSS (which I would recommend against, but I get that some people prefer it) but your stats professors are all teaching with R or Stata (plausible), you might want to consider which direction you want to go in.

If you do go down the R route, there are tons of free online resources, and you can spend some time playing around with swirl, e.g., and learning it yourself. The nice thing about R is that it's free, powerful, open-source, and widely used. The less nice thing about R is that it takes some getting used to.

 

[pediatric_psydoc]

If you will be using SPSS, I recommend Julie Pallant's SPSS Survival Manual. That helped me during my statistics classes and I will also be using that when I do the stats for my dissertation.


Sent from my iPhone using SDN mobile