Demystifying variance in performance: A longitudinal multilevel perspective

Abstract

Research summary:This study employs longitudinal multilevel modeling to re-examine the relative importance of business unit, corporation, industry, and year effects on business unit performance. Total variance in performance is partitioned into stable variance and dynamic variance. Sources of these two parts of variance are explored. Empirical results indicate that (1) stable effects of corporation-industry interaction are substantially important, but were unequally confounded with stable effects of business unit, corporation, and industry in results of previous studies; (2) stable effects of corporation, industry, and corporation-industry interaction, taken together, are of similar relative magnitude to stable effects of business unit; and (3) random and nonlinear year effects are very important in explaining dynamic variance. These findings extend our theoretical and empirical understanding of performance variability.Managerial summary: Whether stable or changing, business units themselves, corporate-parents, and industries influence business unit operations. This article investigates the relative effects of these factors on business unit performance. Although the traditional wisdom is that business unit is critical, this research finds that corporate-parent, industry, and interactions between these, taken together, are as influential as business unit. Specifically, interactions between corporate-parent and industry are important for over-time average business unit performance, indicating that a given corporate-parent unevenly influences its business units in different industries and that a particular industry unevenly influences business units within itself from different corporate-parents. This study also demonstrates that changes in business unit, corporate-parent, and industry are important drivers of over-time volatility of business unit performance and that effects of these changes differ.

On Ignoring the Random Effects Assumption in Multilevel Models: Review, Critique, and Recommendations

“…The difference between these approaches is that in the FE approach, cluster specific values are estimated for (e.g., by using dummy variables) whereas in the RE approach just the variance but not case values of are estimated. Outside any particular model, the phenomenon that some Level 2 units are higher on the dependent variable than others is referred to in econometrics as "unobserved heterogeneity" (Wooldridge, 2013, Chapter 13), and has been given various other labels in organizational research such as "stable variance" (Guo, 2017).…”

Using the SHAPLEY value approach to variance decomposition in strategy research: Diversification, internationalization, and corporate group effects on affiliate profitability

“…Our analysis of how diversification and internationalization affects corporate group influence demonstrates that beyond providing more accurate answers to old research questions, the Shapley Value method enables scholars to address novel research questions that could not be reliably answered using extant methods. In particular, an analysis of the contingencies that influence effect importance—which, as demonstrated, cannot be reliably accomplished using extant methods—has the potential to illuminate the ongoing debate about sources of variation in firm profitability (e.g., Crossland & Hambrick, 2007; Fitza, 2017; Fitza & Tihaniy, 2018; Guo, 2017; Ma et al, 2013; Quigley & Graffin, 2017). The greater reliability of the Shapley Value approach when applied to datasets with fewer observations should also prove an important advantage for researchers seeking to understand novel organizational phenomena such as the drivers of success in crowdfunding campaigns (e.g., Dushnitsky & Fitza, 2018), the factors influencing the growth trajectories of decentralized autonomous organizations (e.g., Hsieh, Vergne, Anderson, Lakhani, & Reitzig, 2018), and the role of artificial intelligence technologies in shaping the performance of firms and markets (Agrawal, Gans, & Goldfarb, 2019).…”

“…Alternative approaches have also been used, including simultaneous equation modeling (Brush, Bromiley, & Hendrickx, 1999), nonparametric estimation (Ruefli & Wiggins, 2003), and multilevel modeling (Guo, 2017; Hough, 2006; Misangyi et al, 2006). These are not exempt from criticism (Guo, 2017; Hough, 2006; McGahan & Porter, 2005). The multilevel approach appears to be the most promising, as it explicitly takes the cross‐nested structure of variation in firm performance into account.…”

Growth or profit? Strategic orientations and long‐term performance in China

“…We measure sales growth as the percentage of sales change from the previous year. ROA indicates the net profit as a percentage of total assets in each year, as applied in previous studies (Guo, 2017; McGahan & Porter, 1997).…”