Minding the Time: A Critical Look at Longitudinal Design and Data Analysis in Quantitative Public Management Research

• Published date: 01 June 2017
• DOI: 10.1177/0734371X17697117
• Date: 01 June 2017
• Author: Justin M. Stritch
• Subject Matter: Articles

https://doi.org/10.1177/0734371X17697117

Review of Public Personnel Administration

2017, Vol. 37(2) 219 –244

© The Author(s) 2017

Reprints and permissions:

sagepub.com/journalsPermissions.nav

DOI: 10.1177/0734371X17697117

journals.sagepub.com/home/rop

Article

Minding the Time: A Critical

Look at Longitudinal

Design and Data Analysis

in Quantitative Public

Management Research

Justin M. Stritch1

Abstract

Public management scholars are looking to longitudinal research designs and data

to help overcome the many limitations associated with cross-sectional research.

However, far less attention has been given to time itself as a research lens for

scholars to consider. This article seeks to integrate time as a construct of theoretical

importance into a discussion of longitudinal design, data, and public management

research. First, I discuss the relative advantages of longitudinal design and data, but

also the challenges, limitations, and issues researchers need to consider. Second, I

consider the importance of time as a theoretical construct of interest in the pursuit

of longitudinal public management research. Third, I offer a brief look at the use

of longitudinal design and panel data analyses in the current public management

literature. The overview demonstrates a notable absence of public management

research considering the attitudes, motives, perceptions, and experiences of individual

public employees and managers. Finally, I consider why there are so few longitudinal

studies of public employees and point out the issues public management researchers

interested in individual employee-level phenomena need to consider when advancing

their own longitudinal research designs.

Keywords

public management theory, longitudinal data, survey research, cross-sectional data

1Arizona State University, Phoenix, AZ, USA

Corresponding Author:

Justin M. Stritch, School of Public Affairs, Center for Organization Research and Design, Arizona State

University, 411 N. Central Avenue, 480-C, Phoenix, AZ 85004, USA.

Email: jstritch@asu.edu

697117ROPXXX10.1177/0734371X17697117Review of Public Personnel AdministrationStritch

research-article2017

220 Review of Public Personnel Administration 37(2)

Introduction

Cross-sectional survey research has facilitated the development and initial testing of

theories related to public employees’ and managers’ behaviors, motivations, attitudes,

and perceptions of life in public organizations. However, there are growing concerns

that public management research is overly dependent on cross-sectional designs and

that scholars have given insufficient attention to the limits of cross-sectional survey

data in testing causal relationships. Today, research articles commonly cite the threats

common methods and common source (or rater) biases pose to estimates and statistical

inference as major weaknesses of cross-sectional research (Favero & Bullock, 2015;

Jakobsen & Jensen, 2014). To a large extent, however, these specific issues might be

addressed through the design of research (see Hassan & Hatmaker, 2015; Jacobsen &

Andersen, 2015).

A more fundamental problem, however, is that cross-sectional data are static. In

contrast, theories of management are, in essence, theories of change and are dynamic

(Mitchell & James, 2001; Oberfield, 2014a; Ployhart & Vandenberg, 2010). With

respect to causality, public management researchers working with cross-sectional data

are unable to observe changes among variables across time and cannot directly test for

causal relationships (Popper, 1959). This is extremely problematic for public manage-

ment researchers who are interested in the attitudes, perceptions, motivations, and

experiences of individual public employees, but who do not have repeated observa-

tions for variables across multiple periods of time.

Public management researchers acknowledge the field’s methodological underde-

velopment and are actively advancing methodological approaches that address the

limitations of cross-sectional research (see Anderson & Edwards, 2015; Blom-Hansen,

Morton, & Serritzlew, 2015; Grimmelikhuijsen, Tummers, & Pandey, 2017; Zhu,

2013). As public management scholars strive for methodological advancement, how-

ever, researchers must continue to keep an eye on the advancement and development

of theory. There is an inherent interplay among method and theory. Different methods

may offer different opportunities to test different and, potentially, more complex theo-

retical relationships. Researchers must exercise caution to ensure that the advance-

ment of methods does not become the end game at the expense of substantive research

interests and the development of theory.

Although a longitudinal research and design is not a single method per se, it is an

approach to research that is receiving attention given the limitations of cross-sectional

research. Current public management research has emphasized the statistical advan-

tages and trade-offs in longitudinal analyses and panel data methods (Marvel & Pitts,

2014; Oberfield, 2014a, 2014b; Zhu, 2013). However, much less consideration in pub-

lic management research is given to time as a theoretical construct of interest in longi-

tudinal studies (for an exception, see O’Toole & Meier, 1999). In this article, I argue

that thinking of time as a theoretical construct will open new windows for theory

development in public management research just as it has in the general management

and organization science literatures (Ancona, Goodman, Lawrence, & Tushman, 2001;

Mitchell & James, 2001; Ployhart & Vandenberg, 2010).