https://datascience.isr.umich.edu/events/coderspaces/

The idea of using synthetic data was mentioned. More concretely, how can synthetic data be used to facilitate reproducible research? Do you see synthetic data being created by a team dedicated to data management before being passed on to a team dedicated to data analysis, with the former having access to the synthetic and real data right off the bat, while the latter only having access to the real data after having pre-specified hypotheses?

Apologies if I missed an important piece of information. In case of the funnel graph, how do we know that lower frequency of published studies with positive results is NOT due to scarcity of studies that find positive associations but BECAUSE of the publication bias?

Re: funnel graphs - I also wondered how we would know if the finding is not due to the fact that a negative/positive association is more likely than the other

Do you think journals give proper credit for analyses that are doing things "the hard way"? And are there strategies for targeting journals that properly reward these reproducibility techniques for researchers that are doing them?

Thanks!

Thank you, got it.

Thank you!

OK great. Thanks!

AAPOR standard definitions were mentioned. Any thoughts on how applicable they are to EMA studies?

Software guide for the practical section. https://pdhp.isr.umich.edu/workshops/tools-for-reproducible-research/

Any idea about graduate programs with substantial focus on quantitative social sciences?

EMA = Ecological Momentary Assessments

Typically, several surveys are completed by a participant in a day

Thank you!

That will be great to have!!! I will do you an email, definitely. Thanks

Please note: Before you leave today please take a moment to fill out our feedback survey on this workshop. Very helpful for me/us to shape the upcombing workshops! https://pdhp.isr.umich.edu/workshops/workshop-feedback/

I will also send this link out via e-mail following the workshop. Thank you!

If we would like to again use packages in "D:/r/packages/" when we restart R studio, would we have to run .libPaths("D:/r/packages/") again? Does the call to .libPaths cause the change in where R looks for the libraries to persist even after closing R Studio?

Thanks!

Would it be possible to share these example script/Rmd files with attendees after the session?

My apologies, please. Which of Python, R and Stata is most advisable to use in research and practice for quantitative social scientists/social data scientists? Can we say a tool holds more promise for the future of work in those fields than others?

The prior RMD file would be helpful, but I can also rewatch the video. Thanks!

To @Elhakim's question, I think it depends, in part, if you rely on certain packages that are **only** available in R, Python, or Stata. This may be a constraint such that at least part of your workflow would **have** to be in R, Python, or Stata.

Great.

Thanks very much!

Thanks for the knowledge shared.

Thank you for a wonderful presentation!

Super useful

Thank you!!

Thanks for the presentation.

Thank you Alex!

Thanks, Alex! Great to see you!

Thank you!

https://pdhp.isr.umich.edu/workshops/workshop-feedback/

Thank you!