Stat Consultant for Dissertation Recommendations

[myfuzzypen]

Hi all,

Due to a smaller sample size than I had previously planned for, I need stat help on my dissertation. There are so many options out there, I am not sure what websites are reputable, etc.

Does anyone have specific recommendations they would be willing to share for a good stat consultant?

Thanks!

---

Members don't see this ad.

 

[R. Matey]

Does your university have any free services? It be better to start there if you can.

Also, can you talk about this with your methodologist?

It would just be a bummer to pay for something you could get for free.


Sent from my iPhone using Tapatalk

 

[WisNeuro]

Best way to handle it, get some books and start reading, I recommend Andy Fields' stuff.

 

[Ollie123]

This will totally depend on the specifics. If the issue is a small N, I assume you or a committee member wants you to Bayesian approaches and report credibility intervals instead but it could be any number of other things (e.g. you planned to use latent variables, but don't have enough people to reliably apply the measurement model component, convergence issues on random effects at level 2 of an HLM, etc.).

Not all statisticians will be competent in what you may need to do.

Agree with the above that a combination of internal resources and self-study is likely a better starting point. If you are still at a "Here is the data, something went wrong and I don't know what to do" stage - a statistician is probably going to struggle to give a satisfying answer.

 

[psycho1391]

Best way to handle it, get some books and start reading, I recommend Andy Fields' stuff.

This will totally depend on the specifics. If the issue is a small N, I assume you or a committee member wants you to Bayesian approaches and report credibility intervals instead but it could be any number of other things (e.g. you planned to use latent variables, but don't have enough people to reliably apply the measurement model component, convergence issues on random effects at level 2 of an HLM, etc.).

Not all statisticians will be competent in what you may need to do.

Agree with the above that a combination of internal resources and self-study is likely a better starting point. If you are still at a "Here is the data, something went wrong and I don't know what to do" stage - a statistician is probably going to struggle to give a satisfying answer.

I second Andy Fields. He has a great and easy writing style.

 

[mypointlesspov]

Hi all,

Due to a smaller sample size than I had previously planned for, I need stat help on my dissertation. There are so many options out there, I am not sure what websites are reputable, etc.

Does anyone have specific recommendations they would be willing to share for a good stat consultant?

Thanks!

Does your university have a stats department? Mine does and they do an hour of free consultation which was enough for me to get on track. I'd say to start there or with other local universities.

 

[singasongofjoy]

Another vote for internal guidance, Fields, and also you might look into Laerd statistics website. It's like $5/month for access but it's not one of those stupid sites where it will charge a recurring fee until cancelled- at least it wasn't when I used it for my dissertation. I just paid for one month of it and it was super helpful thinking through checking assumptions and considering what options might be doable given my available data.

 

[mypointlesspov]

Another vote for internal guidance, Fields, and also you might look into Laerd statistics website. It's like $5/month for access but it's not one of those stupid sites where it will charge a recurring fee until cancelled- at least it wasn't when I used it for my dissertation. I just paid for one month of it and it was super helpful thinking through checking assumptions and considering what options might be doable given my available data.

I fully second this. Laerd was extremely helpful to me. Best $5 i spent by far.

 

[myfuzzypen]

Another vote for internal guidance, Fields, and also you might look into Laerd statistics website. It's like $5/month for access but it's not one of those stupid sites where it will charge a recurring fee until cancelled- at least it wasn't when I used it for my dissertation. I just paid for one month of it and it was super helpful thinking through checking assumptions and considering what options might be doable given my available data.

Thank you! Laerd is amazing!!!! I had no idea about this!