Applied Quantitative Methods for Management Research - 2019

Course Syllabus

 

Course coordinator: Michael S. Dahl

Lecturers: Tünde Cserpes, John Thøgersen and Michael S. Dahl

Work load: 5 ECTS

Administrative assistance: Lisbeth Widahl


Time and place

8 lecture days (9.15-15.00):
Week 41: 8 and 10 October 2019
Week 43: 22 October 2019
Week 44: 29 October 2019
Week 45: 5 November 2019
Week 46: 12 November 2019
Week 47: 21 November 2019
Week 48: 28 November 2019

Aarhus BSS, Aarhus University; Fuglesangs Allé 4, DK-8210 Aarhus V. Room: 2628-M211/2628-M303.

Objectives

What does it mean when seminar participants or journal referees claim that your paper has an endogeneity problem or no identification strategy? How can you deal with these challenges? With a point of departure in the ideal randomized experiments, this PhD course at the Department of Management introduces a tool box with various techniques to deal with these questions in quantitative analysis of non-experimental data. We do this from two different perspectives, i.e. structural equation modelling and quasi-experimental design.

This course will introduce participants to state-of-the-art quantitative empirical methods in management with a focus on application and understanding the underlying intuition. This will include topics ranging from regression models to quasi-experimental designs and structural equation models. We will address the discussion of identification, correlation and causality with a point of departure in recent empirical studies from the broad field of management and marketing. This course will introduce two software packages, Stata and AMOS, which are frequently used among empirical scholars.

The goal is to provide participants with a foundation for collecting and analyzing datasets in management research with a focus on establishing causality. Even for students who are not themselves working with quantitative empirics we aim at building an understanding of the central topics, enabling them to read, understand, review and comment on quantitative empirical contributions for their thesis, at conferences and as reviewers for international journals. The ultimate goal is to spark an interest for quantitative analysis and de-mystify the concepts and skills involved.


Topics

Identification, Causality, Types of research questions, Quality of data, Matching, Differences-in-differences, Instrumental variable regression, Regression discontinuity design, Structural equation modelling, Stata, AMOS.

Main literature (you need to buy these)

-   Joshua D. Angrist and Jörn-Steffen Pischke (2014), Mastering ‘Metrics. Princeton University Press.

-   Niels Blunch (2013), Introduction to Structural Equation Modeling using IBM SPSS Statistics and Amos, Sage Publishing.

See Lecture plan below for additional readings for each session (find these on Blackboard).


Application

Application/registration - no later than 15 September 2019 - to: Lisbeth Widahl, Aarhus BSS, Aarhus University, Department of Management (by email). The application form can be downloaded from the website (Application Form). Please note that your registration is binding.


Fee

Participants from other universities will be charged a fee that covers meals during the course (for more information, please contact Lisbeth Widahl). Participants are required to find their own accommodation.


Software

The techniques introduced in the first part of this course will be demonstrated using Stata. All participants are expected to obtain and install the Stata system before the beginning of the course and to be familiar with the basic operations. Aarhus BSS students should contact the IT department to obtain a Stata license (IC-version is sufficient for this course). A 1-year subscription is a good option, if you do not plan to use it after the course.

Structural equation modelling is demonstrated with SPSS and the SPSS AMOS software package, both of which can be obtained in the Aarhus BSS webshop or at the IT department. This part of the course will be based on SPSS and SPSS AMOS, so both are needed.

(See more here: http://studerende.au.dk/it-support/software/ (under Stata, SPSS and SPSS AMOS)).


Prerequisites

A basic understanding of probability and statistics is required. Most master level programmes in management (including cand.merc.) have courses providing the prerequisites for participating in this course. Revisiting the text books and material from these basic classes would be recommended. Several universities are providing excellent and easily accessible online courses (so-called MOOCs, short for massive open online courses) in basic statistics and regression analysis, which also provide an opportunity to revisit these topics. While this course focuses on intuition and application rather than proof and formal equations, some technical content should be expected.


Exam form

Full and active participation is expected during lectures. Each participant will submit a written assignment. The assignments will be evaluated in terms of pass or fail. Deadline for submission is 11 December. Assignments will be available on Blackboard after the final lecture, but introduced during the first lecture.


Lecture plan

MSD: Michael S. Dahl, TC: Tünde Cserpes, JT: John Thøgersen

Day 1:

Oct 8

9.15 - 15.00

Room: 2628-M211

Introduction to quantitative research and Stata

 

Readings:

-   TBA

TC
MSD

Day 2:

Oct 10

9.15 - 15.00


Room: 2628-M211

Experiments


Readings:

-   TBA

TC
MSD

Day 3:

Oct 22

9.15 - 15.00

 

Room: 2628-M211

Regression and matching

 

Readings:

-   TBA

TC
MSD

Day 4:

Oct 29

9.15 - 15.00

 

Room: 2628-M211

Instrumental variable regression

Readings:

-   TBA

TC
MSD

Day 5:

Nov 5

9.15 - 15.00

 

Room: 2628-M211

Regression discontinuity design

Readings:

-   TBA

TC
MSD

Day 6:

Nov 12

9.15 - 15.00

 

Room: 2628-M303

Differences in differences

Readings:

-   TBA

TC
MSD

Day 7:

Nov 21

9.15 - 15.00

Room: 2628-M211

Structural equation modeling, the basics

Readings:

-   TBA

JT

Day 8:

Nov 28

9.15 - 15.00

 

Room: 2628-M211

Structural equation modeling, multi-group and cross-lagged panel analysis

 

Readings:

-   TBA

 

JT