We study the properties of a three-step approach to estimating the parameters of a latent structure model for categorical data and propose a simple correction for a common source of bias. Such models have a measurement part (essentially the latent class model) and a structural (causal) part (essentially a system of logit equations). In the three-step approach, a stand-alone measurement model is first defined and its parameters are estimated. Individual predicted scores on the latent variables are then computed from the parameter estimates of the measurement model and the individual observed scoring patterns on the indicators. Finally, these predicted scores are used in the causal part and treated as observed variables. We show that such a naive use of predicted latent scores cannot be recommended since it leads to a systematic underestimation of the strength of the association among the variables in the structural part of the models. However, a simple correction procedure can eliminate this systematic bias. This approach is illustrated on simulated and real data. A method that uses multiple imputation to account for the fact that the predicted latent variables are random variables can produce standard errors for the parameters in the structural part of the model.

Employee perceptions of HR practices are often assumed to play an important mediating role in the relationship between HR systems and HR outcomes. In a multisource, multilevel study of 2,063 employees and 449 managers in 119 branches of a single large firm, the authors tested how managers' perceptions of the HR practices implemented in the unit relate to employee perceptions of these HR practices. The authors' main aim is to explore managers' communication quality as a moderator of the relationship between manager-rated and employee-rated HR practices. They also tested whether perceived human resource management (HRM) perceptions in turn relate to perceived unit performance and satisfaction. Multilevel structural equation modeling analyses showed that HRM perceptions mediated the relationship between implemented HRM and both satisfaction and unit performance and that communication moderated the relationship between manager-rated and employee-rated HRM. These findings contribute to scholars' understanding of how HRM affects employee-related outcomes.

This is the accepted version of the paper.This version of the publication may differ from the final published version. Permanent repository link AbstractThe relationship between organizational performance and two dimensions of the widely known 'high performance work system' -enriched job design and high involvement management (HIM) -is assumed to be mediated by worker well-being.We outline the basis for three models: mutual-gains in which employee involvement increases well-being and this mediates its positive relationship with performance;conflicting outcomes which associates involvement with increased stress for workers that accounts for its positive performance effects; and counteracting effects which associates involvement with increased stress and dissatisfaction, and reduces its positive performance effects. These are tested using the UK's Workplace Finally, HIM is negatively related to job-related anxiety-comfort but this plays no mediating role in the link to performance. It is also unrelated to enriched job design. Keywords High involvement management Enriched job design Well-being Stress Job satisfaction Financial performance Labour Productivity Quality AbsenteeismMulti-level analysis 3Enriched job design, high involvement management and organizational performance:The mediating roles of job satisfaction and well-being Direct employee participation is one of the most widely advocated interventions for influencing organizational performance and worker well-being (Humphrey et al., 2007;Parker et al., 2001). It is central to modern organizational concepts such as Lawler's (1986) high involvement management (HIM), human resource management (HRM) (Guest, 1987), the mutual gains enterprise (Kochan and Osterman, 1994), and the high performance work system (Appelbaum et al., 2000;Benson and Lawler, 2003;Cappelli and Neumark, 2001).Two types of opportunity for direct participation are associated with these management models: a) the design of jobs that give their holders discretion, variety and high levels of responsibility; and b) organizational involvement methods that extend beyond the narrow confines of the job, such as teamworking, idea-capturing schemes and functional flexibility.Type a) is associated with the job redesign movement and the concept of job enrichment.Type b) is associated with the high involvement or commitment model that emerged out of this movement, particularly through its popularization by Lawler (1986) and Walton (1985).It is widely expected that these forms of employee involvement enhance the quality of individuals' working lives and their well-being and performance, and consequently the performance of organizations.Originating in the 1990s, following the emergence of high-involvement or highcommitment management, much of the research on workplace employment systems or HRM has concentrated on the performance effects of organization-level practices (e.g. Batt, 2002;Cappelli and Neumark, 2001;Huselid, 1995;MacDuffie, 1995;Wood and De Menezes, 2008). Involvement at the job level has, howe...

In multilevel modeling, one often distinguishes between macro-micro and micro-macro situations. In a macro-micro multilevel situation, a dependent variable measured at the lower level is predicted or explained by variables measured at that lower or a higher level. In a micro-macro multilevel situation, a dependent variable defined at the higher group level is predicted or explained on the basis of independent variables measured at the lower individual level. Up until now, multilevel methodology has mainly focused on macro-micro multilevel situations. In this article, a latent variable model is proposed for analyzing data from micro-macro situations. It is shown that regression analyses carried out at the aggregated level result in biased parameter estimates. A method that uses the best linear unbiased predictors of the group means is shown to yield unbiased estimates of the parameters.

Our results indicate that the multiple PS method is a feasible method to adjust for observed pretreatment differences in nonrandomized studies where the number of pretreatment differences is large and multiple treatments are compared.