Doktoratsstudium

Research Seminar: Measurement in Business Research


SALZBERGER (Statistics and Mathematics / Marketing Management)
Winter Term 2013 | PI SE 1029


Course language | Learning Outcomes | Topics overview | Grading | Schedule

Dates: October 10, 24, 25 | November 7, 8 | December 19 | January 16, 17


General Information

Latest update of this page / News

  • Latest update November 13, 2013
  • Registration via WWW/LPIS.
  • If you are not yet registered for the course, but in principle eligible, please attend the introductory class on October 10.

Course lnaguage

[en]
The language of this seminar will be English unless all participants are fluent in German and agree otherwise. (In the past, the course was always held in English.)

Learning outcomes

[en]
Quantitative research typically involves some sort of measurement. Measuring the length of a table is easy, measuring temperature is more complicated but thanks to age-long research, today, simple instruments are available and every child can do it.
In business research (marketing, management), which is in the context of this course subsumed under social sciences, things are much more intricate. What about, say, trust in a business relationship - can it be measured as simply as its duration, which would just require a calendar? Customer satisfaction, perceived product quality, attitudes, etc. - can these constructs be properly quantified? In many cases, the literature offers instruments, scales consisting of a few items, that claim being able to measures these constructs reliably and validly. But can we rely on the measures we get? Can we treat those measures as numbers expressing the amount of some property and use them as linear, interval-scaled measures in statistical analyses? Maybe. Maybe not. In fact, in many cases there is not just one instrument but many slightly different ones which all purportatly measures the same thing. Which shoulöd we choose? And what should we do if there is no instrument that we can take off the shelf? What does it take to develop one's own scale? And can every construct we might want to define really be measured?

This seminar deals with problems of measurement in business research, specifically problems of the measurement of so-called latent variables. Some issues are more fundamental, philosophical if you like, others are more hands-on. Both levels are of utmost importance. If you develop a scale, or use an existing one, you should know what you do and why you do it.
The seminar should not be an ordinary lecture but rather a sort of round table. The participants will critically discuss what measurement means in the social sciences, learn what the current standards are (mostly classical or traditional test theory), where the strenghts and shortcomings of these standards lie, and what the alternatives are (modern test theory, item response theory, etc). After taking part in this seminar, you will be in a better position to critically evaluate different approaches to measurement and to evaluate existing scales.
Research always requires compromises. This is especially true for doctoral research. Time and funding are limited. The objective of the seminar is to enable the participants to make informed choices of how to tackle the problem of measurement in the best possible way given constraints and limitations.

The specific goals of the seminar will be discussed in the introductory class in order to accommodate the needs of the participants.

Topics overview

[en]
The seminar (and hopefully lively discussions) will evolve around the following key topics:
  • Foundation of quantitative empirical research with an emphasis on measurement
  • Concepts of measurement in business research (marketing, management, etc.), comprising traditional as well as modern/innovate approaches.
  • Scale analysis
  • Development of measurement instruments

Grading

[en]
The grading will be based on the participant's contributions (discussions, presentations) during the seminar.

Schedule

Topics: to be discussed in the introductory class; the following is preliminary


Thursday,
October 10,
2013
16-17

Introduction - Organization
Particiation required


Thursday,
October 24,
2013
15-19

Block I
The Problem of Measurement in the Social Sciences


Friday,
October 25
2013
15-19

Block I (cont.)
Current Practice, Validity
(Traditional paradigm of measurement)


Thursday,
November 7,
2013
15-19

Block 2
Alternative approaches


Friday,
November 8,
2013
17-19

Block 2 (cont.)
Alternative approaches (cont.)


Thursday,
December 19,
2013
15-19

Block 3
Scale development and analysis


Thursday,
January 16,
2014
15:30-19
TC.5.04

Block 3 (cont.)
Scale development and analysis (cont.)


Friday,
January 17,
2014
15-19
TC.5.16

Block 3 (cont.)
Scale development and analysis (cont.)