CommunityData:Message Walls: Difference between revisions

From CommunityData
Line 16: Line 16:
===Dataset construction===
===Dataset construction===
*Fix non-reverted edits bug (Nate)
*Fix non-reverted edits bug (Nate)
*Add session variables to wiki weeks (Nate)
*Add session variables to wiki weeks (Nate) - DONE
*Generate variables for newcomer edits to ns0, ns3 and ns1200 separately (Nate)
*Generate variables for newcomer edits to ns0, ns3 and ns1200 separately (Nate) - DONE
*Add variable for number of newcomers who make an edit in two consecutive weeks (Nate)
*Add variable for number of newcomers who make an edit in two consecutive weeks (Nate)
*Fix negative age in weeks bug (Nate)
*Fix negative age in weeks bug (Nate) - DONE
*Fix sessions.ns1201/sessions.ns1202 bug (Nate)
*Fix sessions.ns1201/sessions.ns1202 bug (Nate) - DONE


===Analysis===
===Analysis===
* Figure out what's causing the spike in total edits around the transition date - DONE
* Figure out what's causing the spike in total edits around the transition date - DONE
* Run a version of the analyses that takes into account whether newcomers ever edited  
* Run a version of the analyses that takes into account whether newcomers ever edited - DONE
* Use unique editors as an outcome variable - DONE
* Use unique editors as an outcome variable - DONE
*'''By March 22:'''  
*'''By March 22:'''  

Revision as of 17:21, 29 March 2018

Useful Resources

Task Management

Overview

(Updated March 15th)

Get missing wikis

  • ASAP Need to use wikilist3.csv to determine which wikis we don't have
  • ASAP Download the rest and put them through wikiq and build edit weeks

Dataset construction

  • Fix non-reverted edits bug (Nate)
  • Add session variables to wiki weeks (Nate) - DONE
  • Generate variables for newcomer edits to ns0, ns3 and ns1200 separately (Nate) - DONE
  • Add variable for number of newcomers who make an edit in two consecutive weeks (Nate)
  • Fix negative age in weeks bug (Nate) - DONE
  • Fix sessions.ns1201/sessions.ns1202 bug (Nate) - DONE

Analysis

  • Figure out what's causing the spike in total edits around the transition date - DONE
  • Run a version of the analyses that takes into account whether newcomers ever edited - DONE
  • Use unique editors as an outcome variable - DONE
  • By March 22:
    • Run negative binomial versions of models on Hyak
    • Determine models to report
  • By March 25:
    • Clean up code and change to knitrable format
  • On April 10th:
    • Run the analysis on the test dataset

Writing

  • By April 1st: have a full draft of the paper ready for collaborative editing
  • First week of April: Revise paper, present at Seattle meetup for comments/suggestions
  • April 16: CSCW abstract + metadata deadline
  • April 19: CSCW submission deadline

Archive