ReplicationWiki
 | |
| Status | public |
| Table of Contents | |
|---|---|
ReplicationWiki
EDITA database of empirical studies (so far mainly in economics), the availability of replication material for them and of replication studies. It can help teaching replication to students. Seminars at several faculties internationally were already taught for which the information of this database was used. In the starting phase the focus was on some leading journals in economics. Replication results can be published as replication working papers of the University of Göttingen's Center for Statistics.
Involved with this wiki? Write a feature or story about your semantic experiences
Teaching and providing access to information will raise awareness for the need for replications, provide a basis for research about the reasons why replications so often fail and how this can be changed, and educate future generations of scientists about how to make research replicable. The starting phase of this project was funded by the Institute for New Economic Thinking (INET).
Usage[edit]
Pageviews have increased very substantially to more than 7.2 million as of February 17, 2022, especially after the very widely used pages of the Research Papers in Economics (RePEc) service IDEAS link back to each article page of the wiki that links to an entry at their pages (cf. this example and check in their entry under "lists"). Several academic blogs and the media reported. The American Economic Association lists the wiki as useful resource for economists, the Economics Network recommends it for the teaching of econometrics, and several academic journals point their readers to it.[1] [2] [3] [4] It was presented at the Spring Meeting of Young Economists in Aarhus, 2013, the Economics Network conference in Exeter, 2013, the American Economic Association Annual Meeting in Philadelphia, 2014, the Annual Meeting of the French Economic Association in Lyon, 2014, the joined conference of the Econometric Society and the European Economic Association in Toulouse, 2014, the 2014 Research Transparency Forum of the Berkeley Initiative for Transparency in the Social Sciences, the 2015 National Conference on Teaching and Research in Economic Education (CTREE) in Minneapolis, the Annual Meeting of the Austrian Economic Association (NOeG 2015), in Klagenfurt, the IACSEE Conference “Political and economic systems under challenge – assessing the role and potential of citizenship education” in 2015 Göttingen, in the session on Replication in Economics during the American Economic Association Annual Meeting, 2016, in San Francisco, the RepScience2016 - 1st International Workshop on Reproducible Open Science, Hannover, 2016, the Statistical Week, Augsburg, 2016, the Beyond the Numbers Workshop, St Louis, 2016, the 2017 American Economic Association Annual Meeting in Chicago, the American Political Sciences Association Annual Meeting, Boston, 2018, the digitalsocieties2019, 2nd Annual Conference of the Academy of Sociology (AS), Konstanz, (recording "Open Science Initiative" starting at about minute 20), Semantic Media Wiki Conference 2020, 2021 FOSDEM Open Research Tools and Technologies devroom, Tenth Annual Conference on Teaching & Research in Economic Education (CTREE) 2021, Barcamp open data 2021, and Open Science Days of the Max Planck Digital Library 2021.
Even though quite a number of researchers have registered, few participate actively. To change this we try to get more students involved and to further expand the database to let it cover a significant part of the empirical literature in economics and other social sciences.
We also need to fix some technical issues, e.g., with the forms we use so far only one replication can be entered for every original study but the most interesting studies get replicated more often. We can display this with a boilerplate but the code is too difficult to understand for most users. We plan to solve this with multiple instance templates and subobjects. There is also a bug with the placement of the slideshow on the main page. And searchability has to be improved significantly because the standard ask form is not user friendly. The semantic drilldown extension allows more intuitive searching but a bug so far makes combined searches difficult,[5] which are pivotal in order to find studies as researchers need it. We are currently trying to get easily clickable properties on each page with category pages that show where else to find the same values like for methods and software used. This already works for authors, keywords, and JEL codes. We are hoping for a cooperation with a software company that offers an improved search. We use the Community portal to keep track of plans for improvement.
A relaunch became necessary at the end of 2019 because our software was very outdated and no longer worked with the updated version of PHP on the server.
Extensions[edit]
- Page Forms allow users who are not familiar with wiki syntax to easily enter new data: [6], [7], [8]
- Confirm account ensures only users with an academic background who register with their real name can edit pages, which to us seemed important given prevalent conflicts about replication of scientific studies and the experience with the page science-fraud.org that was shut down due to legal threats
- JavascriptSlideshow is used on the main page for a quick introduction to the wiki
- VoteNY
is used to allow a rating of studies that should be replicated - DynamicPageList helps to show the most recently added replications on the main page
- Replace Text was used because a lot of the data that was originally imported from an Excel sheet was not standardized (cf. [9])
- Semantic Drilldown helps to search in categories and properties more intuitively than with the ask form (though there is an important bug to solve)
- HitCounters allows to show how many times each page has been accessed (this had been a standard feature in earlier versions of media wiki but was then removed)
- Delete History proved useful as the wiki was first set up with a team before it was launched to the general public that we thought should not be confused by changes that we had needed to try out the wiki. It was later removed as it does not work with the updated software and now there is only DeletePagesForGood
