Follower behavior renders leader behavior endogenous: The simultaneity problem, estimation challenges, and solutions

https://doi.org/10.1016/j.leaqua.2020.101441Get rights and content

Abstract

Traditionally, leadership research has focused on unidirectional questions in which leader attributes are considered to determine follower outcomes. However, many phenomena between leaders (x) and followers (y) involve a simultaneous influence process in which x affects y, and y also affects x (i.e., simultaneity). Unfortunately, this simultaneity bias creates endogeneity and is often not properly addressed in the extant leadership literature. In three studies, we demonstrate the challenges of simultaneity bias and present two methodological solutions that can help to correct problems of simultaneity bias. We focus on simultaneity that occurs between follower resistance and leader control. We mathematically demonstrate the simultaneity bias using a simulated dataset and show how this bias can be statistically solved using an instrumental variable estimation approach. Furthermore, we present how the simultaneity bias can be resolved using an experimental design. We discuss how our approach advances theory and methods for leadership research.

Section snippets

Theoretical background

Common definitions of leadership assume that leadership involves an influencing process between a leader and his/her followers that is determined by leader traits and behaviors, by follower perceptions and behaviors towards the leader, and by the context in which leaders interact with their followers (Bass and Bass, 2008). Essentially, this definition implies that leader-related variables are a cause of follower-related variables but also that leader-related variables can be a consequence for

Overview of studies

To demonstrate the challenges associated with simultaneity bias and offer solutions in terms of estimation approaches and research designs, we present three research studies (one empirical observational study, one study using simulated data, and one experimental study). Starting with Study 1, an empirical observation study, our goal is to demonstrate the theoretical idea behind simultaneity, that is, followers and leaders mutually influencing each other. Because the leadership field is

Study 1

We use the case of follower resistance as an example of followers' influence on their leader, thereby highlighting the fact that leader behavior can be endogenous to follower behavior. Specifically, we theorize about the influence of follower resistance on leader control. Because followers who express resistance may be perceived as criticizing and interrupting the workflow, leaders commonly associate resistance as being counterproductive (Waddell and Sohal, 1998). Due to this negative

Study 2

We conducted a demonstration using simulated data to illustrate the consequences of simultaneity bias on the results of an OLS regression (e.g., as part of a simple regression or an ANOVA, which is commonly used in leadership research to estimate the relationship between leaders and their followers). To this end, we contrasted the frequently used method of OLS to the less known instrumental variable estimation method of 2SLS and demonstrate how 2SLS helps to alleviate simultaneity bias.

Design

We used an experimental ABAB design (Hersen and Barlow, 1976), which is typically employed to evaluate the relationship between an intervention and the frequency or magnitude of a dependent variable. It contrasts the effects of an intervention (A) to the effects of a baseline (B) by alternating intervention and baseline conditions. The ABAB design offers several benefits, such as increasing internal validity, limiting the influence of confounding variables, and allowing the investigation of

General discussion

The present research illustrates the severe consequences associated with simultaneity bias in leadership research and provides different approaches to deal with this bias, with the hope of inspiring more rigorous research practices that correct for endogeneity resulting from simultaneity bias. In Study 1, we used lag sequential analysis allowing us to identify that the likelihood of leader control within the leader-follower interaction process is increased by preceding follower resistance.

Conclusion

This set of studies illustrated how leadership research suffers from the endogeneity problem, especially with respect to simultaneity bias, and provides potential solutions for this issue. The issue of considering all information relevant to correctly identifying a leadership model has plagued leadership research. Whereas the limitations of observational research regarding causal analysis are well established, many leadership and organizational behavior studies have not paid sufficient

Declaration of competing interest

None.

Acknowledgements

We are grateful to our editor John Antonakis for his valuable guidance throughout the review process and we thank our three anonymous reviewers for their constructive feedback. Furthermore, we appreciate the helpful suggestions by Holger Steinmetz regarding statistical analysis and the help by Isabel van Oorschot in collecting data.

References (84)

  • P.M. Podsakoff et al.

    Relationships between leader reward and punishment behavior and subordinate attitudes, perceptions, and behaviors: A meta-analytic review of existing and new research

    Organizational Behavior and Human Decision Processes

    (2006)
  • P.M. Podsakoff et al.

    Experimental designs in management and leadership research: Strengths, limitations, and recommendations for improving publishability

    The Leadership Quarterly

    (2019)
  • T. Sy

    What do you think of followers? Examining the content, structure, and consequences of implicit followership theories

    Organizational Behavior and Human Decision Processes

    (2010)
  • M. Uhl-Bien et al.

    Followership theory: A review and research agenda

    The Leadership Quarterly

    (2014)
  • J.D. Angrist et al.

    Instrumental variables and the search for identification: From supply and demand to natural experiments

    Journal of Economic Perspectives

    (2001)
  • J.D. Angrist et al.

    Mostly harmless econometrics: An empiricists' companion

    (2008)
  • J. Antonakis et al.

    Charisma: An ill-defined and ill-measured gift

    Annual Review of Organizational Psychology and Organizational Behavior

    (2016)
  • R. Bakeman et al.

    Analysis of behavioral streams

  • R. Bakeman et al.

    Analyzing interaction: Sequential analysis with SDIS and GSEQ

    (1995)
  • R. Bakeman et al.

    Sequential analysis and observational methods for the behavioral sciences

    (2011)
  • B.M. Bass et al.

    The Bass handbook of leadership: Theory, research, and managerial implications

    (2008)
  • R.F. Baumeister et al.

    Psychology as the science of self-reports and finger movements: Whatever happened to actual behavior?

    Perspectives on Psychological Science: A Journal of the Association for Psychological Science

    (2007)
  • G. Blair et al.

    estimatr: Fast estimators for design-based inference. R package version 0.18.0

  • J. Bound et al.

    Problems with instrumental variables estimation when the correlation between the instruments and the endogenous explanatory variable is weak

    Journal of the American Statistical Association

    (1995)
  • J. Bratton et al.

    Organizational leadership

    (2004)
  • E.R. Burris

    The risks and rewards of speaking up: Managerial responses to employee voice

    Academy of Management Journal

    (2012)
  • B.J. Byiers et al.

    Single-subject experimental design for evidence-based practice

    American Journal of Speech-Language Pathology

    (2012)
  • A.C. Cameron et al.

    Microeconometrics using Stata

    (2009)
  • S. Cameron

    The economics of crime deterrence: A survey of theory and evidence

    Kyklos

    (1988)
  • A.H. Castorr et al.

    The process of rater training for observational instruments: Implications for interrater reliability

    Research in Nursing & Health

    (1990)
  • D.V. Cicchetti

    Guidelines, criteria, and rules of thumb for evaluating normed and standardized assessment instruments in psychology

    Psychological Assessment

    (1994)
  • A.S. Cook et al.

    Observing leadership as behavior in teams and herds–An ethological approach to shared leadership research

    The Leadership Quarterly

    (2019)
  • E.L. Deci et al.

    Self-determination in a work organization

    Journal of Applied Psychology

    (1989)
  • E.L. Deci et al.

    Intrinsic motivation and self-determination in human behavior

    (1985)
  • C.M. Falbe et al.

    Consequences for managers of using single influence tactics and combinations of tactics

    Academy of Management Journal

    (1992)
  • J.D. Ford et al.

    Resistance to change: The rest of the story

    Academy of Management Review

    (2008)
  • S.A. Furst et al.

    Employee resistance to organizational change: Managerial influence tactics and leader-member exchange

    Journal of Applied Psychology

    (2008)
  • R. Gifford

    The dragons of inaction: Psychological barriers that limit climate change mitigation and adaptation

    The American Psychologist

    (2011)
  • L.H. Glynn et al.

    Manual for the Client Language Easy Rating (CLEAR) coding system: Formerly “Motivational Interviewing Skill Code (MISC) 1.1”

  • J.A. Hausman

    Specification tests in econometrics

    Econometrica

    (1978)
  • M. Hersen et al.

    Single case experiment designs

    (1976)
  • J.M. Howell et al.

    The role of followers in the charismatic leadership process: Relationships and their consequences

    Academy of Management Review

    (2005)
  • Cited by (28)

    • Where is “behavior” in organizational behavior? A call for a revolution in leadership research and beyond

      2023, Leadership Quarterly
      Citation Excerpt :

      Measurement error, as described above, and common method bias, of which self-reports are a common component, often lead to endogeneity problems (Clougherty et al., 2016). Other problems such as omitted variables and simultaneity can contribute to endogeneity bias, such as when follower behaviors cause leader behaviors in the study of leadership (Güntner et al., 2020). Endogeneity issues are problematic because they bias effects; in such instances, the predictor variable is correlated with the error term in the model, which influences the true effect of the predictor on the outcome (Antonakis et al., 2010; Hill et al., 2021).

    View all citing articles on Scopus
    View full text