Hierarchical Linear Modeling of Multilevel Data

Abstract

Most data involving organizations are hierarchical in nature and often contain variables measured at multiple levels of analysis. Hierarchical linear modeling (HLM) is a relatively new and innovative statistical method that organizational scientists have used to alleviate some common problems associated with multilevel data, thus advancing our understanding of organizations. This article presents a broad overview of HLM’s logic through an empirical analysis and outlines how its use can strengthen sport management research. For illustration purposes, we use both HLM and the traditional linear regression model to analyze how organizational and individual factors in Major League Baseball impact individual players’ salaries. A key implication is that, depending on the method, parameter estimates differ because of the multilevel data structure and, thus, findings differ. We explain these differences and conclude by presenting theoretical discussions from strategic management and consumer behavior to provide a potential research agenda for sport management scholars.

Post-materialism, Religiosity, Political Orientation, Locus of Control and Concern for Global Warming: A Multilevel Analysis Across 40 Nations

“…Contrary to traditional ANOVA approaches, MLM estimation does not require balanced data and utilizes all available information in an unbalanced dataset. Todd et al (2005) argued that this particular characteristic makes it possible to detect relationships that might have previously gone undetected, which, in turn, allows for performing comprehensive analysis of complex policy issues by clarifying the effects among variables measured at different levels.…”

How the Economic and Financial Situation of the Community Affects Sport Clubs’ Resources: Evidence from Multi-Level Models

“…The different levels are mirrored in the equations which, consequently, have two-sub-indexes (i for organizational level and j for community level). In line with Todd et al [43], the initial model in this study is in the form of a general linear model: Yij = β0j + β1j XOij + rij (1) where Yij is the outcome of interest for club i in community j; β0j the intercept for each community; β1j the expected change in the outcome of interest (Yij) with a one-unit increase in XOij; and rij the residual. Within multi-level analysis, every organizational-level estimate is calculated in separate community-level equations, which are of the following form: β0j = γ00 + γ01XCj + u0j…”

Does context matter? Analysing structural and individual factors of member commitment in sport clubs

“…As a result, we still know little about the infl uences of factors from different levels as well as their interplay on member commitment. In both sport sociology and sport management research, only a few studies currently relate contextual factors to individual behaviour in a suitable way (in sport participation research: Hallmann, Wicker, Breuer & Schüttoff, 2011;Hallmann, Wicker, Breuer & Schönherr, 2012;van Tuyckom & Scheerder, 2010;; player salaries in baseball teams : Todd, Crook & Barilla, 2005; member action in sport clubs: Schlesinger & Nagel, 2013a;2013b The article is structured as follows: The theoretical framework starts with some general refl ections on the relevance of social context conditions associated with individual action. These general assumptions are used to develop a specifi c multilevel framework for explaining the commitment of members in sport clubs.…”

“…Therefore, context analysis can contribute to a deeper understanding of individual behaviour (e.g. Todd et al, 2005); and if there are pertinent theoretical considerations and the data structure allows it, the context should be considered in the analysis.…”