Developing an Effective Empirical Dissertation in Management and Strategy - PowerPoint PPT Presentation

1 / 7
About This Presentation
Title:

Developing an Effective Empirical Dissertation in Management and Strategy

Description:

Developing an Effective Empirical Dissertation in Management and Strategy ... Scott Stern, Northwestern and NBER ... While coursework focuses (nearly) ... – PowerPoint PPT presentation

Number of Views:62
Avg rating:3.0/5.0
Slides: 8
Provided by: scott221
Category:

less

Transcript and Presenter's Notes

Title: Developing an Effective Empirical Dissertation in Management and Strategy


1
Developing an Effective Empirical Dissertation in
Management and Strategy
  • BPS Managing Your Dissertation Workshop
  • August, 2006
  • AOM Meetings
  • Scott Stern, Northwestern and NBER

2
There are three crucial inputs into empirical
management research...
Phenomena
Measurement
Theory
....And even excellent research projects choose
among these inputs for different types of
projects.
3
While coursework focuses (nearly) exclusively on
the use of specific tools and models to test
given hypotheses within a given data
environment....
4
An empirical dissertation requires an integrated
approach, from a clear understanding of the
phenomena, to the development of behavioral and
statistical foundations, to the recognition of
data limitations, to understanding the elements
of a persuasive empirical argument
5
  • At least in part, these other objectives can only
    happy by getting your hands dirty with real
    datasets and thinking through data issues,
    replication, and extension

6
Lets think about a simple but crucial example
  • Suppose we would like to evaluate the performance
    impact of an organizational practice.
  • Typical empirical approach
  • Variations of this regression are pervasive in
    strategy, operations, decision sciences,
    marketing, finance, and accounting, etc....
  • Under H1, best-practice policy recommendation!

7
The Behavioral Challenge
  • Suppose
  • Doesnt this imply that firms should adopt an
    infinite level of X?
  • Even if X is discrete, why arent all firms at
    the highest level?
  • One immediate idea lets add in a squared term
    to allow for a nonlinear relationship
  • Suppose that
  • Doesnt this imply that all firms should adopt an
    optimal level, X? Why is there variation in
    adoption if there are well-defined benefits to
    adoption?
  • In principle, a clear understanding of the
    phenomena should allow us to identify factors
    that result in a given value of X, independent of
    the performance benefits arising from adoption?
  • Garbage can theories of organization
  • Interest Group Politics within the firm
  • The Timing and Details of the Adoption Process
  • In other words, rather than a false divide
    between phenomena and applied econometrics,
    careful understanding of phenomena can allow us
    to craft a far more persuasive empirical
    argument, and the econometrics can reinforce the
    insights and issues raised by the
    phenomena-driven theory!
Write a Comment
User Comments (0)
About PowerShow.com