This is a mini-project to study the recent paper by Jager and Leek on the science-wise discovery rate published in Biostatistics and associated commentary. The aim is to replicate the work and in the process improve our R skills and trial various collaborative and open science tools. Basically this is an alternative version of a traditional journal club.
The paper and other references are available from the Simply Statistics blog
Is most science false? The titans weigh in
The outcomes of doing this work will hopefully be a nice replication of the results of Jager et. al., an assessment of the work, some improvements and criticisms of our own, and a write-up in some form (at the very least in figshare so we and others can highlight a replication). We should also have a nice system for organizing and publishing future projects.
- Richard T. Gray Rgray@kirby.unsw.edu.au
- Andrew P. Craig Acraig@kirby.unsw.edu.au
- Github - primary tool we want to learn and use to store and track our progress.
- Can use as a todo list and comment tracker possibly replacing email
- Final repository could be stored on Figshare to get a DOI.
- Evernote
- Create a shared note stack for storage of useful bits and pieces related to the project
- Postach.io blog for open notebook about research progress??
- Google Drive - possibly use for synced storage but primarily for creating nicely formated collaborative documents.
- Dropbox - synced storage. Might not be necessary but could be useful for things we don't want to put in the repository.
- MarkdownPad2 - software to write everything in Markdown format.
- trial writing collaborative publications (given we don't use latex might not be that useful google docs may work better, we will see). Any final publication is likely to be in word given the work evironment we are in.
- Use to write up all notes and other peices of communication as text files in Markdown format
- Trello - project management
- Github has an issue tracker which can be used as a todo list but Trello is so pretty to use
- Qiqqa - for setting up a common library of references related to the project
In addition to trialing tools we are exploring various ways to organize the project, with an eye to improving the organization and management of more serious research projects in the future. This is what we have come up with so far.
All the project files are stored in 5 main directories.