The Application of Cognitive Processes to Organizational Surveys: How Informants Report About Interorganizational Relationships

1
The Application of Cognitive Processes to
Organizational Surveys How Informants Report
About Interorganizational Relationships

• Joan M. Phillips

Mendoza College of Business University of Notre
Dame
The author gratefully acknowledges the support
provided by the Institute for the Study of
Business Markets at Penn State.
2
Introduction

• Conceptual Issues
• Trend toward fewer but more purposive strategic
  partnerships
• Substantive Issues
• Growing interest among practitioners and scholars
  in understanding...
• How these partnerships are formed
• Reasons for failure
• Determinants of successful relationships

3
Introduction

• Methodological Issues
• Variables related to successful relationships are
  unobservable
• Obtaining valid measures of these latent
  constructs remains a challenge
• Empirical work used survey methods
• Application of cognitive psychology perspectives
  to improve survey data

4
Summary of Literature

• Satisfaction in Vertical Marketing Systems
• Growing interest in understanding
  interorganizational relations
• Reliability and Validity of Organizational
  Informant Reports
• Obtaining valid measures of organizational
  contructs remains a challenge
• Cognitive Aspects of Proxy Reporting
• Application of cognitive theories has improved
  the validity of self- and proxy-reports

5
Research Objectives

• Understand how informants report about their
  organizations and its relationships
• Cognitive processes used
• Data sources used
• Identify the factors associated with consensus
  among multiple informants
• Recommend strategies to improve measures (i.e.,
  questionnaire designs)

6
Sample

• Organizations
• 35 Industrial Manufacturers (5 Mil.-4 Bil.)
• Automotive/Industrial Parts Equipment 4
• Chemical Products 5
• Electrical/Electronic Components 4
• Heavy Equipment and Machinery 5
• Metal Products 6
• Machine Tools, Fasteners, Abrasives 6
• Sheet Metals 5
• Informants
• 109 Informants Interviewed - minimum of 3 per
  firm
• General Management 27
• Sales 31
• Marketing 26
• Customer Service/Administrative 25

7
Questionnaire

• 6 Objective Items
• of Sales Volume from Distribution
• of Sales Volume from Distributor 1
• of Sales Volume from Distributor 2
• Distributor 2s Sales Volume Rank
• Length of Relationship with Distributor 1
• Length of Relationship with Distributor 2

• 10 Subjective Items times 2 Distributors
• Distributors Dependence
• Manufacturers Dependence
• Distributors Influence
• Manufacturers Cooperation
• Distributors Cooperation
• Conflict
• Distributors Communication
• Trust
• Distributors Performance
• Manufacturers Satisfaction

8
Data Collection

• Face-to-face interviews
• Verbal protocol methodology
• Cognitive thinkalouds
• Audio recorded
• Transcribed
• Coded

9
Outcome Measures

• Data Sources
• Self/Job reference
• Documents
• Internal and external communication
• Participation in event
• Organizational cues

• Cognitive Processes
• Anchoring
• Decomposition
• Calculation
• Generalizing

• Consensus Agreement among informants.
• Variance for Subjective Questions
• Relative Error (variance/mean2) for Objective
  Questions

10
Consensus Hypotheses Supported

• Similar Data Sources
• Supported -- Use of similar data sources had a
  positive effect on informant consensus.
• Perceived Question Difficulty
• Supported for Objective Questions -- Perceived
  question difficulty has a negative effect on
  informant consensus for objective questions.
• Importance of Target Firm
• Supported for Objective Questions -- Importance
  of target firm had a positive effect on informant
  consensus

11
Consensus Hypotheses Not Supported

• Similar Cognitive Processes
• Not Supported -- Use of similar processes did not
  have an effect on informant consensus.
• Use of Anchoring Processes
• Not Supported -- Results for distributor 1 were
  mixed results for distributor 2 were in the
  opposite direction.

12
Data Source Hypotheses

• Objective Questions
• Supported -- Informants used self/job cues
  slightly more frequently than documents.
• Supported -- The odds of using documents were
  greater for objective questions.
• Subjective Questions
• Supported -- Informants relied primarily on
  distributor organizational cues.
• Not Supported -- The odds of using self/job cues,
  communication cues, and documents were greater
  for objective questions.

13
Anchoring Process Hypothesis

• Use of Anchoring Adjustment Processes
• Supported -- When documents were not used, the
  odds of using an anchoring procedure for
  subjective questions was greater than for
  objective questions.

14
Contributions

• Specify cognitive processes and information
  sources used by informants to report about
  interorganizational relationships
• Understand how this is impacted by question type
  and informant characteristics
• Identify the determinants of convergence among
  multiple informant reports
• Recommend strategies for improving organizational
  surveys

15
Findings and Recommendations

• FINDING
• Informant agreement seems to be related to
• use of similar information sources
• Informants use
• Organizational cues for subjective questions
• Self/job cues and documents for objective data
• RECOMMENDATION
• Ask specific rather than general questions
• e.g., Based on last months sales report...
• e.g., Based on the written communication your
  firm received from this distributor last week...

16
Findings and Recommendations

• FINDING
• Use of an anchoring process is greater for
  subjective questions
• RECOMMENDATION
• Encourage use of same reference point
• e.g., best distributor we have now, best ever,
  etc.
• Define scale values for informants
• Avoid Likert-type (e.g., SA/SD) scales

17
Findings and Recommendations

• FINDING
• Informants define organizational constructs
  differently (e.g., trust, performance,
  satisfaction, etc.)
• RECOMMENDATION
• Avoid single-item indicators of multi-dimensional
  constructs
• Use global scales with caution

18
Generalizations

• Can people really serve as informants?
• But my experience matters most!
• Do we know who our informants are?
• Papa Bear, Momma Bear, Baby Bear
• Are we asking the right questions?
• Which of your kids do you love the best?
• What is most important?
• Trust, Trust, Trust