Replication Workshop

Prerequisites

This module is part of the Social Science Research Methods Centre training programme which is a shared platform for providing research students with a broad range of quantitative and qualitative research methods skills that are relevant across the social sciences.

This module will introduce students to the process of reproducing published work. Replicating other scholars’ work is an essential tool to get familiar with methods, learn to select suitable models, and get a chance to publish early during their PhD. This replication module will therefore provide students with a deeper understanding of statistical modelling and professionalism in their field. With the right amount of value added, a replication study is publishable after the module.

• Graduate students from participating departments taking the Social Science Research Methods Centre training programme as part of their research degree
• Students who are already competent at using quantitative methods for their own research (=at a minimum, good knowledge of multivariate regression and data handling in R).
• Because of the high workload involved, this course is most suitable for PhD students

• This is an advanced course intended for students are already competent at using quantitative methods for their own research (=at a minimum, good knowledge of multivariate regression and data handling in R).
• Students will have to commit 2-6 hours self-directed work per week.
• Please send an email to the course lecturer (nj248@cam.ac.uk) that your supervisor agrees to your participation, and that you have basic statistics and R knowledge prior to the course starting.
• Preparation for first session

Please bring a copy of the article(s) you would like to replicate so we can discuss if it is suitable. A rough guideline: 1) Pick a paper where the data set is available at the Journal’s webpage, the author’s webpage or a data verse. Do not pick a paper where you cannot locate the data and code! 2) To increase the probability that your paper will eventually be published, pick an article from a top journal. 3) The article should be published 2008 or later. 4) Your paper should use methods you can manage or learn over the Christmas break. 5) Bring the paper(s) including the data on your laptop to the first session.

• Because of the above, this course is primarily intended for PhD students.

Number of sessions: 6

#	Date	Time	Venue		Trainer
1	Thu 5 Nov 2015   17:00 - 19:00	17:00 - 19:00	Department of Sociology, Seminar Room	map	N. Janz
2	Thu 19 Nov 2015   17:00 - 19:00	17:00 - 19:00	Department of Sociology, Seminar Room	map	N. Janz
3	Thu 26 Nov 2015   17:00 - 20:00	17:00 - 20:00	Department of Sociology, Seminar Room	map	N. Janz
4	Thu 3 Dec 2015   17:00 - 20:00	17:00 - 20:00	Department of Sociology, Seminar Room	map	N. Janz
5	Thu 21 Jan 2016   17:00 - 20:00	17:00 - 20:00	Department of Sociology, Seminar Room	map	N. Janz
6	Thu 28 Jan 2016   17:00 - 20:00	17:00 - 20:00	Department of Sociology, Seminar Room	map	N. Janz

Introductory meeting: Approaches to the Replication of Research

Practical Part: Get help with picking a paper

Creating a project plan for your replication study

Practical Part: Access data & codes in R; start writing your project plan

Exploratory Analysis

Practical Part: Loading the data; data management; identify each variable; doing transformations; summary statistics; a first look at statistical models

Replication Analysis

Practical part in R: Reproduce all results in text, tables, figures exactly as in the original paper (=duplication); Compare your results with original study

Cross-Check & How to add value to your replication

Practical part in R: Students will discuss each others’ results and code; we will discuss individualized options on how to add value to the project (=full replication study)

Wrap-up & Final assignment completion: Uploading your project to our class Dataverse

Practical part: Outlook: How to bring your results into a publishable format; how to submit to a conference and journal

Students will work on a published paper (where data are online) that was approved as suitable by the lecturer. In the first sessions, students will be guided through replicating the paper. In the following sessions, they will add value to the replication and bring it into a publishable format.

Each session consists of a short lecture part, and an extensive practical part. There will be a weekly assignment in which you work on your replication, and bring the results (and problems) to the next session.

Please note the prerequisites!

6 sessions over Michaelmas and Lent terms.

Sessions during Michaelmas and Lent term 2015/16