Pushing the Methodological Frontiers of Survey Research

Session Type: PDW Workshop
Program Session: 246 | Submission: 14288 | Sponsor(s): (OM, RM)
Scheduled: Saturday, Aug 6 2016 8:00AM – 11:00AM at Sheraton Park Hotel in Garden C

Pushing the Methodological Frontiers of Survey Research: Endogeneity and Composite Modeling
Methodological Frontiers I

Organizer: Barbara B. FlynnIndiana U.
Organizer: Mikko KetokiviIE Business School – IE U.
Organizer: Daniel GuidePennsylvania State U.
Participant: Philip BromileyU. of California, Irvine
Participant: Mikko RönkköAalto U.

In 2015, the editors of Journal of Operations Management wrote an editorial that updated the methodological criteria for statistical research in the journal. The purpose of this workshop was to elaborate upon these criteria and explore various ways by which OM/SCM researchers can meet the updated criteria in their work.

Part I of the workshop focused on issues related to endogeneity and common factor vs. composite modeling. When we observe a statistical association between X and Y and we have only observational data, how do we plausibly argue that X is driving Y? The answer to this question hinges on the researcher’s ability to argue that the variance of X is exogenous to the model. In this first conversation, we examined the causes and consequences of the problem of endogeneity in OM research, as well as how to deal with the problem.

pdf_icon_4801_Endogeneity

The second conversation focused on the question “How should factors be modeled?” We contrasted common factor models to composite models and examined how researchers can make an informed choice between the two. In this conversation, we also touched on topics such as small sample sizes, explanation vs. prediction, and Partial Least Squares modeling.

pdf_icon_4802_composite_modeling

Each of the conversations featured 1-3 experts on the topic, facilitated by a well-known OM/SCM researcher.

Search Terms: Survey Research | Endogeneity | Composite Modeling

 

Session Type: PDW Workshop
Program Session: 359 | Submission: 14338 | Sponsor(s): (OM, RM)
Scheduled: Saturday, Aug 6 2016 1:00PM – 4:30PM at Sheraton Park Hotel in Garden C

Pushing the Methodological Frontiers of Survey Research: ‘Urban Legends’ and Common Method Bias
Methodological Frontiers II

Organizer: Barbara B. FlynnIndiana U.
Organizer: Daniel GuidePennsylvania State U.
Participant: Jeffrey R. EdwardsU. of North Carolina
Participant: Mikko KetokiviIE Business School – IE U.

In 2015, the editors of Journal of Operations Management wrote an editorial that updated the methodological criteria for statistical research in the journal. The purpose of this workshop was to elaborate upon these criteria and explore various ways by which OM/SCM researchers can meet the updated criteria in their work.

Part II of the workshop focused on issues related to ‘urban legends’ related to cutoff points and other rules of thumb, and common method bias. The research methods literature and empirical research is replete with various “rules of thumb” for topics such as Cronbach’s alpha, VIF, eigenvalues, etc., many of which do not survive methodological scrutiny. In this conversation, we discussed these “urban legends”, where they come from, and most importantly, how to avoid them.

pdf_icon_4803_urban_legends

In the second conversation, we focused on the sources of common method bias (CMB), which are complex and diverse, and there are no easy fixes. Yet, many authors continue to make strong claims about CMB not being a problem in their research. But how many of these claims fall prey to the “I can’t see it therefore it does not exist” fallacy? In this conversation, we examined the sources of CMB and its consequences. Most importantly, we discussed ways of avoiding it already in the research design phase.

pdf_icon_4804_Common_Method_Bias

Each of the conversations will feature 1-3 experts on the topic, facilitated by a well-known OM/SCM researcher.

Search Terms: Cutoff Points | Common Method Bias | Survey Research

pdf_icon_4805_Recommended_Reading