Research Methodology in Management: Current Practices, Trends, and Implications for Future Research
RESEARCH METHODOLOGY IN MANAGEMENT: CURRENT PRACTICES, TRENDS, AND IMPLICATIONS FOR FUTURE RESEARCH
Authors
Terri A. Scandura, University of Miami
Ethlyn A. Williams, University of Colorado at Colorado Springs
Introduction
This study compares research strategies in management between two periods: 1985-87 and 1995-97.
Conducted a content analysis of articles from key journals:
Academy of Management Journal
Administrative Science Quarterly
Journal of Management
Investigated the use of research methods addressing issues of:
Triangulation
Validity (internal, external, construct, statistical conclusion validity)
Results indicate:
Notable trends in research strategies
Decreases in validity measures across studies
Emphasizes the importance of research method assessment for management studies to advance.
Theoretical Overview
Triangulation:
Defined as utilizing multiple methods to corroborate research findings.
Relates to the idea originated by McGrath (1982) regarding inherent flaws in all research methods, necessitating corroborative evidence.
Can apply to:
Strategies (e.g., qualitative vs. quantitative methods)
Data collection settings
Data sources (single vs. multiple)
Example: Jick (1979) details a study assessing anxiety and job insecurity during a merger using multiple data sources (self-reports, interviews, observations).
Advocates (Webb et al., 1966; Smith, 1975) suggest various methods to study phenomena to achieve a well-rounded understanding.
Research Strategies
Increased triangulation may enhance conclusion validity.
McGrath’s (1982) framework categorizes research strategies into eight types:
Formal theory
Sample surveys
Laboratory experiments
Judgment tasks
Computer simulations
Experimental simulations
Field studies
Field experiments
Adjustments for this analysis expanded to nine categories by combining formal theory/literature reviews and differentiating field studies based on data collection type.
Discuss how each method correlates with internal and external validity.
Validation Types
Internal Validity
Concerned with causal relationships. Requires:
Covariation between variables
Evidence that the cause precedes the effect
Alternative explanations dismissed
External Validity
The ability to generalize findings across different populations, settings, and contexts.
Construct Validity
Validity of measures relative to the theoretical constructs they are intended to reflect.
Required assessments for consistency and reliability of measures (Sackett & Larson, 1990).
Statistical Conclusion Validity
Ability to draw reliable conclusions based on statistical evidence of relationships among variables.
Involves consideration of sample size and the appropriateness of statistical tests used for data analysis.
Methodology
Content analyses conducted on research methods sections of articles across both time periods:
Articles selected if they utilized empirical methods or formal theory/literature reviews.
Total articles analyzed:
1985-87: n = 280 empirical studies coded.
1995-97: n = 334 empirical studies coded.
Results analyzed by 13 coding dimensions across various aspects of each study, including:
Research strategy employed
Substantive content area
Level of dependent variable analysis (individual, group, organizational)
Results Overview
Triangulation Trends
Comparison of the two decades showed:
Decline in sample surveys and laboratory experiments
Increased use of field studies, particularly those using secondary data
Significant shifts in substantive content, with a decrease in studies focusing on strategic planning, careers, and turnover, yet rising studies on groups/teams and international management.
Internal and External Validity
Increase of cross-sectional studies at the expense of longitudinal ones
Rise in usage of private sector samples while public sector studies declined.
Construct Validity
Less emphasis was placed on construct validation measures between the periods; reports on discriminant, convergent, and predictive validity decreased while reliability reports were low but stable.
Statistical Conclusion Validity
Sample size medians increased, which may suggest improvements in statistical power.
A notable trend includes a decline in usage of analytical methods such as ANOVA while increasing tendencies in regression techniques and time-series analysis.
Discussion
General trends indicate a shift toward methods compromising validity and triangulation.
Findings stress an existing gap between rigor and relevance in management research, highlighting issues of validity.
Recommendations from prior literature (e.g., Mitchell, 1985) on practices to enhance validity appear neglected over the periods studied.
Future Directions
The need for improved triangulation in management studies is emphasized to build robust research findings.
Discussion of innovative research methodologies, such as online samples and simulation tools to enhance methodological rigor.