2
$\begingroup$

I am interested in bootstrap, especially how it's entering the psychologist's world.

Boostrapping is used to 'omit' classical assumptions to get statistics. Bootstrapping starts with creating new samples from an original sample (that is considered as a population). Those new samples are generated via resampling with replacement (so that in a new sample observations can be repeated). Doing this (for example) 1000 times, you can get 1000 new samples and then calculate statistics of interest on each of a new sample. Thus you get 1000 of means (as an example). In the end you can calculate (bootstrapped) confidence interval.

I am looking for papers on bootstrap in social sciences (like this one Wright, D.B, London, K., Field, A.P. (2011) Using Bootstrap Estimation and the Plug-in Principle for Clinical Psychology Data, Journal of Experimental Psychology, DOI: 10.5127/jep.013611), where the authors explain bootstrap technique to psychologists in their subfield or those papers (in social science) where bootstrap is used extensively.

Has anyone has seen, read or known such papers?

$\endgroup$
  • $\begingroup$ Would you mind explaining in greater depth what bootstrap is and how it's typically used? The reference to the paper is useful, but a quick paragraph explaining the principle would be greatly appreciated. $\endgroup$ – Seanny123 Dec 29 '15 at 17:34
  • $\begingroup$ ... in greater depth... I'd like to address you to wikipedia, but it's used to 'omit' classical assumptions to get statistics. Bootstrap starts with creating new samples from an original sample (that is considered as a population). Those new samples are generated via resampling with replacement (so that in a new sample observations can be repeated). Doing this (f.e.) 1000 times, you can get 1000 new samples and then calculate statistics of interest on each of a new sample. Thus you get 1000 of means (as an example). In the end you can calculate (bootstrapped) confidence interval. $\endgroup$ – Lil'Lobster Dec 29 '15 at 17:48
2
$\begingroup$

When I embarked on using bootstrapping in my research I found this chapter/lecture notes by Cosma Shalizi very useful:

I see now that he has also written a short introduction for American Scientist

We published this paper, which uses bootstrapping in the third analysis (and so may directly answer your question).

Aside from mere egotism on my part, the analysis code for the paper is available, so you can see mechanically how I did it for my analysis (in Python)

$\endgroup$
  • $\begingroup$ I'm unsure why you deleted this, but I took the liberty to undelete it. Ping me if you disagree please. $\endgroup$ – AliceD Mar 5 '18 at 8:53
  • $\begingroup$ @AliceD Who deleted? Me or Tom? I may have done this only by accident (so thx). $\endgroup$ – Lil'Lobster Mar 5 '18 at 15:46
  • $\begingroup$ @Lili, it was Tom's doing and I don't understand why they deleted their post. $\endgroup$ – AliceD Mar 5 '18 at 18:50
  • $\begingroup$ I don't know why I deleted this either, most likely it was an accident, so happy for you to undelete $\endgroup$ – Idiot Tom Mar 6 '18 at 11:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.