Digital Methods for Exploratory Research

Companion website for the ABRI doctoral seminar

NOTE: This is a live syllabus and subject to change. Make sure you always check the live version and not the static copy or the print.

Last updated: 2019-06-24 13:45:14 +0000

Teaching Faculty

Mahmood Shafeie Zargar (KIN / VU - Coordinator) Website | Email

Joey van Angeren (KIN / VU) Website | Email

Philipp Hukal (Copenhagen Business School - Guest) Website | Email

Brian Pentland (Michigan State - Guest) Website | Email

Course Description

The recent advances in the availability of digital trace data as well as the corresponding data analysis methods have opened up new venues for social science research. These trends allow new types of analysis and make new ways of theorising possible. This course intends to initiate the participants with a catalogue of contemporary tools and techniques necessary to navigate and analyse the emerging data landscape, with an emphasis the exploratory analysis of socially meaningful similarities. In particular, it aims to provide research students from diverse backgrounds with the basic skills and methodological insights to take advantage of digital trace data and computational tools. This course is agnostic to research paradigms and leaves the door open to participants from qualitative and quantitative backgrounds, as we believe the methods covered here are most useful as pre-processing aids for other analysis techniques. The course is aimed towards research students from fields related to business administration and organisation studies who are either interested in discovering different methods or aiming to enrich their methods toolbox by engaging hands-on with digital trace data. Familiarity with R and a knack for hack are mandatory (see below).

Learning Activities & Objectives

During the course the students will:

Target Audience

PhD and research master students from the VU, The Netherlands and abroad, engaged in research projects broadly related to business administration or organisation studies. We expect a maximum of 25 participants.

Course Outline[1]

Session Date Topic Location Description
1 / J & M Thursday Oct 31 / 13:00–18:00 Theory Building in the Age of Digital Data / Intro to R HG-5A91[2][3] We start with an introduction to computational qualitative research design and theorising relations, processes and meaning using computational methods.
2 / M Thursday Nov 7 / 13:00–18:00 Data Collection & Preparation / Intro to Course Datasets HG-5A91[2][3] This session is organised as a hands-on tutorial on data formats, data sources, as well as data collection and preparation techniques.
3 / M Thursday Nov 14 / 13:00–18:00 Social Network Analysis: Analysing Relations HG-5A91[2][3] Social network analysis methods allow mapping and measuring of relationships and flows between individuals or collectives. They are the cornerstone of the relational approach to sociology.
4 / P & B Thursday Nov 21 / 13:00–18:15 Sequence Analysis: Analysing Processes HG-5A91[2][3] Human behaviour often follow certain patterns of more or less formalised routines and processes. In this session we will learn to analyse and theorise such behavioural patterns in this session.
5 / J Thursday Nov 28 / 13:00–18:00 Semantic Analysis: Analysing Meaning HG-5A91[2][3] Topic modelling and latent semantic analysis can reveal underlying patterns of meaning in large, distributed communities and organisations. We will learn the techniques and discuss their implications for theory building in this session.
6 / J & M Thursday Dec 5 / 13:00–18:00 Detecting Semantic & Syntactic Similarity in Text HG-5A91[2][3] To be completed: We will use semantic as well as syntactic similarity methods to detect similar text content across text corpora.
7 Thursday Dec 12 / 13:00–18:00 Personal Appointments - During the personal appointments the students will get the chance to seek the help of the instructors on their data wrangling and analysis process, in order to progress towards the final paper.
8 / J & M & P Thursday Dec 19 / 13:00–18:00 Project Presentation HG-5A91[2][3] This is a marathon presentation session where the students will share their final paper, and will receive feedback from their peers as well as the instructors.

[1] The outline is tentative, although we will do our best to make sure the dates won’t change.

[2] Main (Hoofdgebouw) building

[3] The classroom is within the staff-only perimeter of the building. You can use the inter-phone on the ground floor, or the buzzers on the 5th floor to call the course coordinator, once you are here. Alternatively, drop him an email in advance and set an appointment.

View Map

Class Schedule (To be revised)

All sessions, except Session 4, are organized as follows:

For Session 4 (Sequence Analysis):

Deliverables

The students are supposed to read the mandatory course readings and prepare for the class discussions.

For the theory sessions the students will be designated as paper discussants on a voluntary basis. The paper discussants have five to ten minutes to present the gist of the paper, before launching and leading the class discussion. About 30 minutes is allocated to each paper. Depending on the number of students each student will be assigned once or twice during the term to be the class lead.

A hands-on tutorial will follow each theory session. The students must come prepared to the tutorials as well. They have to familiarise themselves with the software package used in the session beforehand. The tutorial exercises contain questions that the students must respond to and submit to the session instructor(s) 3 days in advance. For a Thursday class at 13:00 the exercises will be due on Monday at 13:00. The students are allowed to partner with another student for the submissions, as long as the two partners work together. The hands-on exercises will be communicated to the students at least a week in advance of the submission deadline.

The students will have to pick one of the presented methods to implement in their term paper. A short paper proposal (2-3 pages) and a usable dataset are due before the personal appointment (Dec. 12).

A presentation session is scheduled on the last day of the course, so that the students receive constructive feedback from their colleagues and the instructors.

Course Material (To be updated)

Grading

Pre-requisites

The assumption is that the students are initiated to the statistics for research, including the use of R programming language for analysis. It is crucial that students are familiar with the basic use of computers (e.g. know how to navigate their way around the Internet) and are proficient with at least one software suit for data handling (e.g. using Excel, SPSS, SQL or other software). The course will require a limited amount of coding as a requisite condition for the deliverables, what we will facilitate by offering examples and code snippets. The students are encouraged to spend time before and during the course to improve their R skills, so that they make the most out of the course. The students who lack a basic understanding of the common descriptive and inferential statistics methods may want to take a preliminary statistics course before taking this course. The students who have little experience with R are required to do the following tutorials as an absolute minimum: Beginner’s Guide to R (Computer World) and Introduction to R (DataCamp). All the students are encouraged to go through Hadley Wickham’s R for Data Science and practice the topics they have not covered before. We will also hold a short R workshop at the beginning of the term, where we will focus on R’s specificities and the troublesome aspects of working with it.

Credits

The seminar will count as 6 ECTS under European credit transfer and accumulation system. If required, a certificate can be provided upon the completion of the course.

Registration

If you have not registered yet, but you would like to attend the seminar, you can use this form to enroll. For additional information on the administrative formalities, contact graduateschool.abri@vu.nl.

Appendix: Resources