The effect of estimation method and sample size in multilevel structural equation modeling

Multilevel structural equation modeling (multilevel SEM) has become an established method to analyze multilevel multivariate data. The first useful estimation method was the pseudobalanced method. This method is approximate because it assumes that all groups have the same size, and ignores unbalance when it exists. In addition, full information maximum likelihood (ML) estimation is now available, which is often combined with robust chi‐squares and standard errors to accommodate unmodeled heterogeneity (MLR). In addition, diagonally weighted least squares (DWLS) methods have become available as estimation methods. This article compares the pseudobalanced estimation method, ML(R), and two DWLS methods by simulating a multilevel factor model with unbalanced data. The simulations included different sample sizes at the individual and group levels and different intraclass correlation (ICC). The within‐group part of the model posed no problems. In the between part of the model, the different ICC sizes had no effect. There is a clear interaction effect between number of groups and estimation method. ML reaches unbiasedness fastest, then the two DWLS methods, then MLR, and then the pseudobalanced method (which needs more than 200 groups). We conclude that both ML(R) and DWLS are genuine improvements on the pseudobalanced approximation. With small sample sizes, the robust methods are not recommended.     

Regional Inequalities in Consumption Patterns: A Multilevel Approach to the Case of Italy

The main aim of this paper is to evaluate the disparities in the Italian regions on the demand side. In more detail, an attempt will be made to find if the consumption behaviour of Italian households is different in the regions. With this in mind, Istat's 2000 Italian Family Budget data set was analysed. The data in question, which were collected through a two‐stage sample over Italy's 20 regions, contains information regarding the expenses of approximately 23,000 households. In this analysis, both households and regions are considered as units: households are nested in the regions so that the basic data structure is hierarchical. In order to take this hierarchical structure into account, a multilevel model was used, making it possible for parameters to vary randomly from region to region. The model in question also made it possible to consider heterogeneity across different groups (regions), such as stochastic variation. First, regional inequalities were tested using a simple model in which households constituted the first level of analysis and were grouped according to their region (the second level). As a second step, and in order to investigate the interaction between geographical context and income distribution, another model was used. This was cross‐classified by income and regions. The most relevant results showed that there is wide fragmentation of consumption behaviour and, at the same time, various differentiated types of behaviour in the regions under analysis. These territorial differentials become clear from income class and items of consumption.     

Contracts to Communities: A Processual Model of Organizational Virtue

abstract In the face of systemic challenges to corporate legitimacy, scholars and managers alike have been rethinking traditional answers to the question: What does it take to be a good company? We approach this question in two novel ways. We offer a normative answer, grounded in virtue ethics, by introducing a threefold typology of organizational forms. The moral goodness of each form depends on the congruence between its purpose and virtues. But we also offer a positive answer in the form of a processual model which traces corporate goodness to its empirical antecedents and consequences. The model defies a view of organizations as innately good or evil, but rather portrays virtue as the sediment of a value infusion process. We predict that if managers succeed in establishing in their organizations the kind of virtues necessary to support collective moral agency, they can expect to reap gains like enhanced effectiveness and legitimacy. However, when they neglect their moral responsibilities, the result will likely be organizational demise.     

The impact of generalized and institutional trust on donating to activist,leisure, and interest organizations: individual and contextual effects

In this paper, we answer the question as to what extent donating to activist, interest, and leisure organizations is affected by both individual and national levels of generalized and institutional trust. We use the European Social Survey 2002 to estimate multilevel random intercept models, based on more than 33 000 individuals living in 19 European countries and the USA. Our results show very consistent positive impacts of both individual generalized trust and institutional trust on donating to all types of organizations. The effects are strongest for donating to activist organizations and absent only for the relation between institutional trust and donating to interest organizations. At the national level, generalized trust positively affects donations to activist and leisure organizations, but not to interest organizations. Institutional trust at the national level is negatively related to donations to all types of organizations. This latter finding suggests that when institutions are perceived to function well, individuals estimate that their philanthropic donations are less needed. Copyright © 2011 John Wiley & Sons, Ltd.     

Longitudinal Meta-analysis

The goal of meta-analysis is to integrate the research results of a number of studies on a specific topic. Characteristic for meta-analysis is that in general only the summary statistics of the studies are used and not the original data. When the published research results to be integrated are longitudinal, multilevel analysis can be used for the meta-analysis. We will demonstrate this with an example of longitudinal data on the mental development of infants. We distinguish four levels in the data. The highest level (4) is the publication, in which the results of one or more studies are published. The third level consists of the separate studies. At this level we have knowledge about the degree of prematurity of the group of infants in the specific study. The second level are the repeated measures. We have data about the test age, the mental development, the corresponding standard deviations, and the sample sizes. The lowest level is needed for the specification of the meta-analysis model. Both the way in which the multilevel model has to be specified (the Mln-program is used) as the results will be presented and interpreted.     

Voting on Pensions: A Survey

Abstract. The paper presents a nonexhaustive survey of the literature designed to explain emergence, size and political sustainability of pay‐as‐you‐go pension systems. It proposes a simple framework of analysis (a small, open, two overlapping generation economy model), around which some variants are displayed. Dictatorship of the median voter is assumed. The text is organized to answer the following questions: (i) Do political equilibria with PAYG pension schemes exist? (ii) Why do they emerge? (iii) What are the conditions for the participation constraint of the pension game to be verified?, and finally, (iv) What is the size of the pension system chosen by the median voter and how is this size influenced by an exogenous (e.g. demographic) shock?     

Balancing priorities: Decision-making in sustainable supply chain management

The need for environmental protection and increasing demands for natural resources are forcing companies to reconsider their business models and restructure their supply chain operations. Scholars and proactive companies have begun to create more sustainable supply chains. What has not been fully addressed is how organizations deal with short-term pressures to remain economically viable while implementing these newly modeled supply chains. In this study, we use theory-building through case studies to answer the question: how do organizations balance short-term profitability and long-term environmental sustainability when making supply chain decisions under conditions of uncertainty? We present five sets of propositions that explain how exemplars in green supply chain management make decisions and balance short and long term objectives. We also identify four environmental postures that help explain the decisions organizations make when dealing with strategic trade-offs among the economic, environmental and social elements of the triple-bottom-line.     

Heteroskedasticity, autocorrelation, and spatial correlation robust inference in linear panel models with fixed-effects

This paper develops an asymptotic theory for test statistics in linear panel models that are robust to heteroskedasticity, autocorrelation and/or spatial correlation. Two classes of standard errors are analyzed. Both are based on nonparametric heteroskedasticity autocorrelation (HAC) covariance matrix estimators. The first class is based on averages of HAC estimators across individuals in the cross-section, i.e. “averages of HACs”. This class includes the well known cluster standard errors analyzed by Arellano (1987) as a special case. The second class is based on the HAC of cross-section averages and was proposed by Driscoll and Kraay (1998). The ”HAC of averages” standard errors are robust to heteroskedasticity, serial correlation and spatial correlation but weak dependence in the time dimension is required. The “averages of HACs” standard errors are robust to heteroskedasticity and serial correlation including the nonstationary case but they are not valid in the presence of spatial correlation. The main contribution of the paper is to develop a fixed-b asymptotic theory for statistics based on both classes of standard errors in models with individual and possibly time fixed-effects dummy variables. The asymptotics is carried out for large time sample sizes for both fixed and large cross-section sample sizes. Extensive simulations show that the fixed-b approximation is usually much better than the traditional normal or chi-square approximation especially for the Driscoll-Kraay standard errors. The use of fixed-b critical values will lead to more reliable inference in practice especially for tests of joint hypotheses.     

Leading for Long‐Term Performance: Matching the Behavior to the Situation

Is there one right way to lead, or does it all “depend?” The right answer will inform the right leadership development strategy. A recent study by McKinsey & Co. using survey data collected from 165 organizations and over 365,000 respondents globally suggests that the answer is both. Two types of behavior sets emerged from organizations that enjoyed higher member perceptions of leadership effectiveness and attained higher levels of organizational health (the latter predicts long‐term financial and operational performance). We combined these into a model that prescribes a set of “baseline” behaviors that should always be evident, complemented by another set of behaviors (“situational behaviors”) that should be differentially emphasized depending on the level of organizational health. Both the content of the behavior and its relative emphasis are critical to either set. The current paper combines these into a model of organization‐wide leadership and concludes with action implications for organizations’ leadership development strategy.     

The Problem of Time Dependent Explanatory Variables at the Context-Level in Discrete Time Multilevel Event History Analysis: A Comparison of Models Considering Mobility Between Local Labour Markets as an Example

The paper deals with the question of how to include time dependent explanatory variables at the context-level in multilevel event history models. In general, context-level explanatory variables in multilevel models are assumed to be time constant. Only time constant context-level explanatory variables perform the task of reducing context-level error variance. Thus, it will be suggested that the analysis should be extended to a three-level model. In this model, time periods of persons constitute level 1 units, time periods of contexts constitute level 2 units and the contexts themselves constitute level 3 units – in which in turn level 2 units are clustered. Considering mobility between local labour markets as an example, four different ways of modelling time varying context-level variables are compared. The result is that the proposed three-level model leads to the most conservative results.     

Simplified Jackknife Variance Estimates for Fuzzy Measures of Multidimensional Poverty

In this paper, we present a practical methodology for variance estimation for multi‐dimensional measures of poverty and deprivation of households and individuals, derived from sample surveys with complex designs and fairly large sample sizes. The measures considered are based on fuzzy representation of individuals' propensity to deprivation in monetary and diverse non‐monetary dimensions. We believe this to be the first original contribution for estimating standard errors for such fuzzy poverty measures. The second objective is to describe and numerically illustrate computational procedures and difficulties in producing reliable and robust estimates of sampling error for such complex statistics. We attempt to identify some of these problems and provide solutions in the context of actual situations. A detailed application based on European Union Statistics on Income and Living Conditions data for 19 NUTS2 regions in Spain is provided.     

Calibrating Public Sector Accountability: Translating experimental findings to public sector accountability

Accountability mechanisms are among the most important means with which governments guard and improve the performance of public sector organizations. However, research documents a plethora of accountability-failures. A key issue is: how can public sector accountability become more effective? This paper seeks to answer this question by connecting two largely separated strands of research: public administration research on real-world organizations and experimental research on the effects of different forms of accountability on decision-making. The paper develops the Calibrated Public Accountability-model (CPA-model) from experimental research findings which can be used to investigate how accountability can be calibrated to task requirements of organizations.     

What does a Monetary Policy Shock Do? An International Analysis with Multiple Filters*

What does a monetary policy shock do? We answer this question by estimating a new‐Keynesian monetary policy dynamic stochastic general equilibrium model for a number of economies with a variety of empirical proxies of the business cycle. The effects of two different policy shocks, an unexpected interest rate hike conditional on a constant inflation target and an unpredicted drift in the inflation target, are scrutinized. Filter‐specific Bayesian impulse responses are contrasted with those obtained by combining multiple business cycle indicators. Our results document the substantial uncertainty surrounding the estimated effects of these two policy shocks across a number of countries.     

Forging a ‘new’ organizational infrastructure for Los Angeles’ progressive community

What margins of maneuverability do urban‐based progressive movements have for affecting policy outcomes in entrepreneurial and neoliberal political systems? This article provides a partial answer to this question by examining how relations developed and stabilized between actors in the different sectors (community based organizations, labor, university) of Los Angeles’ progressive community. Such relations are a necessary but not sufficient condition for affecting policy outcomes. I argue that these relations have resulted from a 20‐year process of interactions between the more innovative agents of each of the sectors. Through their repeated experimentation in building frameworks to coordinate their partnerships, I argue that a variety of complex mechanisms have taken shape that nourish relations and coordinate complex forms of collective action. Functioning as ‘relational platforms’, these coordinating mechanisms have combined to form an emergent ‘organizational infrastructure’ that facilitates both ongoing relational processes and the mobilization of collective resources in politically effective ways. Thus, by examining the organizational infrastructure that makes such a broad based ‘movement’ possible and sustainable, the article offers the reader one insight into how urban progressives have been able to build the power necessary to affect policies in one of the world's most entrepreneurial and neoliberal cities.     

Multi‐level Environmentalism and the European Union: The Case of Trans‐European Transport Networks

In the European Union, trans‐European transport networks (TENs) are a vital element in the constitution of one European space in order to enable the free movement of people and goods throughout the Union. Their construction, however, often causes environmental degradation. Opposition to EU politics is mostly voiced at the level of individual nation‐states. As the case of TENs reveals, however, protest against European policy projects with environmental side effects can take the form of ‘multi‐level environmentalism’, linking lobbying and ‘conscientization’ in Brussels with direct action at the national and local levels. Civil society theory, social movement theory and governance theory help ensure a theoretically informed answer to the question of how the resistance to TENs is organized and framed. By questioning dominant problem definitions and solution strategies, environmental movements and movement organizations, both in Brussels and in the individual nation‐states, point to the possibility of looking at social and political reality from another, non‐hegemonic point of view. In this way, they contribute to challenging the often biased technocratic, growth‐oriented character of the European Union.     

Was eigentlich ist „gutes“ Personalmanagement? Einige Anregungen zur Professionalisierungsdebatte

It seems to be a simple question: What is meant by professional and successful human resource management? Finding a serious and theoretically grounded answer, which is empirically proved, turns out to be quite complicated. There is a long history of research in human resource management: Many debates at conferences facing the improvement of human resource management have been held. Lots of publications by practitioners, consultants and academics have been written. But there is still no concrete answer found which meets the demand for rigor und relevance.     

A cybernetic view of environmental management: the implications for business organizations

Current environmental management practice is judged to be mechanistic and based on previous experience with health and safety and with quality management systems. In support of the claim that systems approaches to management are required, the question is asked: what is the systemic nature of environmental management and will it lead an organization towards some concept of sustainable business practice? To try and answer this question the Viable System Model (VSM) developed by Stafford Beer was used to analyse current environmental management practice from a systems perspective in seven manufacturing organizations. The findings suggest that significant structural changes in organizations will be necessary as well as changes in management practices (core competencies) and attitudes if organizations are to become viable. Tomorrow's business strategy is likely to be driven much more by cooperation rather than by competition. Some possible implications for business are: (1) Corporate portfolios will be organized around the supply chain (or flows of energy and materials) instead of products and markets. (2) Strategy will not only continue to be made at the corporate and business unit levels but also at the industry level. (3) Such developments will require much higher degrees of integration and cooperation than is currently practised between companies. © 1997 John Wiley & Sons, Ltd and ERP Environment.     

Collaborative modes with Cultural and Creative Industries and innovation performance: The moderating role of heterogeneous sources of knowledge and absorptive capacity

Organizations can no longer be isolated entities in the current dynamic competitive environment but tend to establish alliances and networks with external parties. Recently, research has suggested that organizations develop collaborative modes with Cultural and Creative Industries (CCIs) to acquire and combine Heterogeneous Sources of Knowledge (HSK) within their organizational environment. Innovation thus becomes increasingly generated by a cross-process involving a diverse set of industries. Nevertheless, few studies have assessed how knowledge from distant industries, such as CCIs, can affect firms’ innovation performance. Accordingly, this empirical research on a sample of 187 firms evaluates whether formal and informal collaboration modes with CCIs affect innovation performance. Moreover, it assesses the moderating role of both HSK and absorptive capacity. In particular, it aims to answer the following questions. What are the effects of collaboration modes with CCIs on firms’ innovation performance? What are the differences (in terms of effects) between Formal Collaboration Modes (FCMs) and Informal Collaboration Modes (ICMs) with CCIs on firms’ innovation performance? Do heterogeneity of knowledge sources and absorptive capacity increase the possibility of benefiting from FCMs and ICMs with CCIs in terms of innovation performance? The results contribute to the theory indicating that HSK are important to benefit from knowledge from CCIs, while absorptive capacity does not fulfil the same role.     

课程改革的“突破口”:对教育评价的改革

本文通过对一个多世纪西方教育评价各流派的回顾,结合中国这一特殊国度的历史事实和当代教育评价的现状,认为课程改革的"突破口"应是对教育评价的改革。从而基于对我国学生学业评价中存在问题的剖析,试图在回答"什么是受过教育的人"这一教育哲学核心问题的同时,尝试建构一种新的教育评价观,即底线评价和自我评价的圆融互设。     

An inconvenient dataset: bias and inappropriate inference with the multilevel model

The multilevel model has become a staple of social research. I textually and formally explicate sample design features that, I contend, are required for unbiased estimation of macro-level multilevel model parameters and the use of tools for statistical inference, such as standard errors. After detailing the limited and conflicting guidance on sample design in the multilevel model didactic literature, illustrative nationally-representative datasets and published examples that violate the posited requirements are identified. Because the didactic literature is either silent on sample design requirements or in disagreement with the constraints posited here, two Monte Carlo simulations are conducted to clarify the issues. The results indicate that bias follows use of samples that fail to satisfy the requirements outlined; notably, the bias is poorly-behaved, such that estimates provide neither upper nor lower bounds for the population parameter. Further, hypothesis tests are unjustified. Thus, published multilevel model analyses using many workhorse datasets, including NELS, AdHealth, NLSY, GSS, PSID, and SIPP, often unwittingly convey substantive results and theoretical conclusions that lack foundation. Future research using the multilevel model should be limited to cases that satisfy the sample requirements described.