Evaluating estimation methods for ordinal data in structural equation modeling

This study examined the performance of two alternative estimation approaches in structural equation modeling for ordinal data under different levels of model misspecification, score skewness, sample size, and model size. Both approaches involve analyzing a polychoric correlation matrix as well as adjusting standard error estimates and model chi-squared, but one estimates model parameters with maximum likelihood and the other with robust weighted least-squared. Relative bias in parameter estimates and standard error estimates, Type I error rate, and empirical power of the model test, where appropriate, were evaluated through Monte Carlo simulations. These alternative approaches generally provided unbiased parameter estimates when the model was correctly specified. They also provided unbiased standard error estimates and adequate Type I error control in general unless sample size was small and the measured variables were moderately skewed. Differences between the methods in convergence problems and the evaluation criteria, especially under small sample and skewed variable conditions, were discussed.     

Partial least squares structural equation modeling in HRM research

Abstract

Partial least squares structural equation modeling (PLS-SEM) has become a key multivariate analysis technique that human resource management (HRM) researchers frequently use. While most disciplines undertake regular critical reflections on the use of important methods to ensure rigorous research and publication practices, the use of PLS-SEM in HRM has not been analyzed so far. To address this gap in HRM literature, this paper presents a critical review of PLS-SEM use in 77 HRM studies published over a 30-year period in leading journals. By contrasting the review results with state-of-the-art guidelines for use of the method, we identify several areas that offer room of improvement when applying PLS-SEM in HRM studies. Our findings offer important guidance for future use of the PLS-SEM method in HRM and related fields.     

The impact of non-normality,sample size and estimation technique on goodness-of-fit measures in structural equation modeling: evidence from ten empirical models of travel behavior

Ten empirical models of travel behavior are used to measure the variability of structural equation model goodness-of-fit as a function of sample size, multivariate kurtosis, and estimation technique. The estimation techniques are maximum likelihood, asymptotic distribution free, bootstrapping, and the Mplus approach. The results highlight the divergence of these techniques when sample sizes are small and/or multivariate kurtosis high. Recommendations include using multiple estimation techniques and, when sample sizes are large, sampling the data and reestimating the models to test both the robustness of the specifications and to quantify, to some extent, the large sample bias inherent in the χ ² test statistic.     

Estimating structural equation models with non-normal variables by using transformations

We discuss structural equation models for non-normal variables. In this situation the maximum likelihood and the generalized least-squares estimates of the model parameters can give incorrect estimates of the standard errors and the associated goodness-of-fit chi-squared statistics. If the sample size is not large, for instance smaller than about 1000, asymptotic distribution-free estimation methods are also not applicable. This paper assumes that the observed variables are transformed to normally distributed variables. The non-normally distributed variables are transformed with a Box–Cox function. Estimation of the model parameters and the transformation parameters is done by the maximum likelihood method. Furthermore, the test statistics (i.e. standard deviations) of these parameters are derived. This makes it possible to show the importance of the transformations. Finally, an empirical example is presented.     

Statistical power of structural equation models in SCM research

Prior research has emphasized the relevance of adequate statistical power for covariance-based structural equation modeling (CSEM). Nevertheless, reviews in domains other than supply chain management (SCM) found that the magnitude of power tends to be inadequate. This finding is worrisome because statistical power directly affects the meaningfulness of the conclusions based on CSEM. The issue is particularly relevant for the field of SCM in light of the increasing use of CSEM. An investigation of the statistical power of CSEM published in seven major SCM journals since 1999 confirms this criticism. Specifically, an analysis of 988 applications of CSEM indicates that 32% of all applications have too little power, increasing the probability of Type II errors, and that another 43% of all applications exhibit excessive power, increasing the probability of Type I errors. This paper emphasizes the importance of adequate statistical power for CSEM in SCM.     

Dynamic inoperability input-output modeling for economic losses estimation in industries during flooding

Among all natural disasters, flood stands as a recurrently happening disaster. It holds the aptitude to disrupt the organizations and to cause absenteeism of the workforce in industries. As the workforce is directly involve in the functioning of industries, work force absenteeism can cause reduced production and inoperability which outcomes in financial losses of industrial sectors. This research objects to estimate inoperability of industries due to distraction of workers by incorporating Dynamic Inoperability Input-Output Model (DIIM). Economic losses are determined from inoperability. Industrial area which is selected for the research includes local industries in Peshawar, Khyber-Pakhtunkhwa, Pakistan. Various industries are chosen and are ordered according to inoperability and economic losses. Industries having highest financial damages are: (i) Agriculture; (ii) Sugar mills; and (iii) Marble industry. These three industries hold liable for 40.6% of the overall financial losses of fifteen industries. Industries suffering from highest inoperability include (i) Sugar mills; (ii) Agriculture and (iii) Marble industry. A risk analysis frame work has also been developed to help industrial sectors to recover after a disaster. Besides, data of three different floods has also been taken for the above mentioned critical sectors to plot probability distributions for predicting economic losses of most frequent floods. Furthermore, this research methodology has been applied to flooding but it can be applied to any other disaster, everywhere.     

Nonparametric identification and estimation of sample selection models under symmetry

Under a conditional mean restriction Das et al. (2003) considered nonparametric estimation of sample selection models. However, their method can only identify the outcome regression function up to a constant. In this paper we strengthen the conditional mean restriction to a symmetry restriction under which selection biases due to selection on unobservables can be eliminated through proper matching of propensity scores; consequently we are able to identify and obtain consistent estimators for the average treatment effects and the structural regression functions. The results from a simulation study suggest that our estimators perform satisfactorily.     

Robustness of normal theory statistics in structural equation models*

A condition is given by which optimal normal theory methods, such as the maximum likelihood methods, are robust against violation of the normality assumption in a general linear structural equation model. Specifically, the estimators and the goodness of fit test are robust. The estimator is efficient within some defined class, and its standard errors can be obtained by a correction formula applied to the inverse of the information matrix. Some special models, like the factor analysis model and path models, are discussed in more detail. A method for evaluating the robustness condition is given.     

A multilevel examination of high‐performance work systems and unit‐level organisational ambidexterity

This paper examines the process linking high‐performance work systems (HPWS) and organisational ambidexterity both at the unit and firm level of analyses by integrating strategic HRM, human capital and social capital perspectives. Multisource and multilevel data from 2,887 employees and 536 managers of 58 banks was collected. Results revealed that firm‐level HPWS were positively related to unit‐level employee human capital. Unit‐level employee human capital partially mediated the relationship between firm‐level HPWS and unit organisational ambidexterity. Furthermore, firm‐level social climate moderated the effect of firm‐level HPWS on unit organisational ambidexterity through unit‐level employee human capital. This paper contributes to HPWS and ambidexterity research by revealing the impacts of firm‐level HPWS and mediating mechanisms, as well as identifying boundary conditions for pursuing unit‐level organisational ambidexterity.     

The Effectiveness of Recruitment Advertisements and Recruitment Websites: Indirect and Interactive Effects on Applicant Attraction

This study compares the effect of printed recruitment advertisements and recruitment websites on applicant attraction and shows how these recruitment activities interact with one another. Our results indicate that websites have a significantly stronger impact on applicant attraction than printed advertisements. We show that websites and printed recruitment advertisements have an indirect effect on applicant attraction that is mediated by employer knowledge. Furthermore, printed advertisements supplement the positive effects of websites and, thus, the simultaneous use of multiple recruitment activities directly influences applicant attraction. © 2014 Wiley Periodicals, Inc.     

含水层底板标高的两种估值方法比较

对于许多区域水资源问题，用数值方法进行潜水水流模拟时，需要给出每个节点上含水层底板标高值。首先简单介绍了趋势面方法，然后着重阐述了克里格方法的基本原理及它们在合水层底板标高估值中的应用，通过比较两种方法的计算结果可以得出克里格方法是进行含水层底板标高估值的空间最优估计方法。     

改进的郭涛算法解一类有限元离散常微分方程

文章提出了改进的郭涛算法,并运用该算法解区域分裂、有限元离散之后的非线性常微分方程两点边值问题。数值结果表明该算法能很好地收敛到准确解,精度较高。     

Continuous time modeling of panel data: SEM versus filter techniques

Although convincing arguments have been put forward for continuous-time modeling, its use in panel research is rare. In one approach, classical N  = 1 state-space modeling procedures are adapted for panel analysis to estimate the exact discrete model (EDM) by means of filter techniques. Based on earlier less satisfactory indirect methods, a more recent approach uses structural equation modeling (SEM) to get the maximum likelihood estimate of the EDM by the direct method. After an introduction into continuous-time state-space modeling for panel data and the EDM, a thorough comparison is made between the two distinct approaches with quite different histories by means of Monte Carlo simulation studies. The model used in the simulation studies is the damped linear oscillator with and without random subject effects.     

The Measurement and Structural Modeling of the Reasonableness of Workplace Accommodations

The Americans with Disabilities Act of 1990 (ADA) guarantees protection from discrimination for persons with a disability. The ADA requires employers to make reasonable accommodations for persons with physical and mental disabilities unless doing so would produce undue hardship on the organization. Fifteen years after the passage of the ADA, the question as to what is a reasonable accommodation remains controversial, especially for some types of disabilities. In this study, a reasonable accommodation scale is proposed and tested using structural equation modeling in hopes that such a scale will aid the courts and organizations in determining what is reasonable. Individuals’ level of awareness of disability issues is also examined in the model to examine the relationship between awareness and willingness to accommodate. Results indicate acceptable fit of the model to the data and provide support for the proposed scale. Furthermore, the relationship between awareness of disability issues and willingness to accommodate was significant. Implications for research and practice are discussed.     

Human resource management and developing proactive environmental strategies: The influence of environmental training and organizational learning

The need to implement advanced approaches to protect the environment is forcing companies to refocus their internal procedures and actions. To match employees' capabilities and the organization itself to these new requirements, the human resource management department can offer some key tools. This article analyzes whether environmental training (ET) and organizational learning (OL) positively influence the development of proactive environmental strategies (PESs) and compares the two processes, which differ in the time, costs, and difficulty required to implement them. Companies in the tourism sector are currently facing a highly dynamic environment where innovativeness is a decisive factor for achieving competitiveness. As such, we analyze whether the presence or absence of innovativeness influences the development of these two mechanisms. Using a sample of 252 tourism companies, we tested these relationships using structural equation modeling. The findings showed that (1) innovativeness is an antecedent of implementing ET and OL in the companies sampled, (2) both mechanisms promote implementing PESs, and (3) ET is equally as effective as OL for this purpose. Managers should take these findings into account when deciding which mechanism to apply when striving to achieve environmental proactivity.     

On finite sample properties of alternative estimators of coefficients in a structural equation with many instruments

We compare four different estimation methods for the coefficients of a linear structural equation with instrumental variables. As the classical methods we consider the limited information maximum likelihood (LIML) estimator and the two-stage least squares (TSLS) estimator, and as the semi-parametric estimation methods we consider the maximum empirical likelihood (MEL) estimator and the generalized method of moments (GMM) (or the estimating equation) estimator. Tables and figures of the distribution functions of four estimators are given for enough values of the parameters to cover most linear models of interest and we include some heteroscedastic cases and nonlinear cases. We have found that the LIML estimator has good performance in terms of the bounded loss functions and probabilities when the number of instruments is large, that is, the micro-econometric models with “many instruments” in the terminology of recent econometric literature.     

Expanding Insights on the Diversity Climate–Performance Link: The Role of Workgroup Discrimination and Group Size

The present study extends knowledge of the performance consequences of workgroup diversity climate. Building upon Kopelman, Brief, and Guzzo's ( 1990 ) climate model of productivity, we introduce workgroup discrimination as a behavioral mediator that explains the positive effects of diversity climate on workgroup performance. In addition, we investigate group size as a moderator upon which this mediated relationship depends. We test these moderated‐mediated propositions using a split‐sample design and data from 248 military workgroups comprising 8,707 respondents. Findings from structural equation modeling reveal that diversity climate is consistently positively related to workgroup performance and that this relationship is mediated by discrimination. Results yield a pattern of moderated mediation, in that the indirect relationship between workgroup diversity climate (through perceptions of workgroup discrimination) and group performance was more pronounced in larger than in smaller workgroups. These results illustrate that discrimination and group size represent key factors in determining how a diversity climate is associated with group performance and, thus, have significant implications for research and practice. © 2014 Wiley Periodicals, Inc.     

Investigating the Number of Non-linear and Multi-modal Relationships Between Observed Variables Measuring Growth-oriented Atmosphere

This study investigates the number of non-linear and multi-modal relationships between observed variables measuring the Growth-oriented Atmosphere. The sample (N = 726) represents employees of three vocational high schools in Finland. The first stage of analysis showed that only 22% of all dependencies between variables were purely linear. In the second stage two sub samples of the data were identified as linear and non-linear. Both bivariate correlations and confirmatory factor analysis (CFA) parameter estimates were found to be higher in the linear sub sample. Results showed that some of the highest bivariate correlations in both sub samples were explained via third variable in the non-linear Bayesian dependence modeling (BDM). Finally, the results of CFA and BDM led in different substantive interpretations in two out of four research questions concerning organizational growth.     

Small sample properties of maximum likelihood versus generalized method of moments based tests for spatially autocorrelated errors

Many applied researchers have to deal with spatially autocorrelated residuals (SAR). Available tests that identify spatial spillovers as captured by a significant SAR parameter, are either based on maximum likelihood (MLE) or generalized method of moments (GMM) estimates. This paper illustrates the properties of various tests for the null hypothesis of a zero SAR parameter in a comprehensive Monte Carlo study. The main finding is that Wald tests generally perform well regarding both size and power even in small samples. The GMM-based Wald test is correctly sized even for non-normally distributed disturbances and small samples, and it exhibits a similar power as its MLE-based counterpart. Hence, for the applied researcher the GMM Wald test can be recommended, because it is easy to implement.     

Elucidating the Relationship Between Satisfaction and Citizen Involvement in Public Administration

Abstract

This study analyzes the relationships between public administration satisfaction (PAS), political efficacy (PE), and active citizen involvement (ACI). The study group consists of 345 respondents residing in Taiwan. Respondents are interviewed by phone to obtain their perceptions regarding local municipality operations (i.e. professionalism and sympathy (PS), ambition to improve (AI), neutrality and morality (NM)), PAS, ACI, and PE). These relationships are verified by conducting path analysis. Results of this study demonstrate that AI and NM positively affect PAS, that PAS positively affects PE and ACI, and that PE plays a mediating role in the relationship between PAS and ACI.