Explore the vast range of titles available.

Classical path analysis is a statistical technique used to test causal hypotheses involving multiple variables without latent variables, assuming linearity, multivariate normality, and a sufficient sample size. The d-separation (d-sep) test is a generalization of path analysis that relaxes these assumptions. Although model selection using Akaike's information criterion (AIC) is well established for classical path analysis, this model selection technique has not yet been developed for d-sep tests. In this paper, I explain how to use the AIC statistic for d-sep tests, give a worked example, and include instructions (supplemental material) to implement the analysis in the R computing language.

Communities of plants, biological soil crusts (BSCs), and arbuscular mycorrhizal (AM) fungi are known to influence soil stability individually, but their relative contributions, interactions, and combined effects are not well understood, particularly in arid and semiarid ecosystems. In a landscape-scale field study we quantified plant, BSC, and AM fungal communities at 216 locations along a gradient of soil stability levels in southern Utah, USA. We used multivariate modeling to examine the relative influences of plants, BSCs, and AM fungi on surface and subsurface stability in a semiarid shrubland landscape. Models were found to be congruent with the data and explained 35% of the variation in surface stability and 54% of the variation in subsurface stability. The results support several tentative conclusions. While BSCs, plants, and AM fungi all contribute to surface stability, only plants and AM fungi contribute to subsurface stability. In both surface and subsurface models, the strongest contributions to soil stability are made by biological components of the system. Biological soil crust cover was found to have the strongest direct effect on surface soil stability (0.60; controlling for other factors). Surprisingly, AM fungi appeared to influence surface soil stability (0.37), even though they are not generally considered to exist in the top few millimeters of the soil. In the subsurface model, plant cover appeared to have the strongest direct influence on soil stability (0.42); in both models, results indicate that plant cover influences soil stability both directly (controlling for other factors) and indirectly through influences on other organisms. Soil organic matter was not found to have a direct contribution to surface or subsurface stability in this system. The relative influence of AM fungi on soil stability in these semiarid shrublands was similar to that reported for a mesic tallgrass prairie. Estimates of effects that BSCs, plants, and AM fungi have on soil stability in these models are used to suggest the relative amounts of resources that erosion control practitioners should devote to promoting these communities. This study highlights the need for system approaches in combating erosion, soil degradation, and arid-land desertification.

Skill-sharing services have the potential to foster regional development and mutual aid within a community through residents’ social participation. Despite the growing social demand for skill-sharing services, few cases have utilized individuals’ knowledge, skills, and other intellectual assets. To widely diffuse such services, it is necessary to clarify user factors (motivations and barriers to use services) and reflect on the service design process. However, there is limited knowledge regarding user analysis and skill-sharing services. Thus, this study explores user factors that affect the intention to use skill-sharing services and derives guidelines for skill-sharing service design and development. A hypothetical user factor model was constructed through a literature review of user research in sharing services and empirical analysis of actual skill-sharing services. The hypothetical model was applied to a survey on the use of skill-sharing services by residents in Hino city, the western part of Tokyo (n = 358). The results revealed that social motivation and self-actualizational motivation significantly affected the service use intention of skill providers. Economic motivations and enjoyment of service activities derive the service use intention of skill receivers. Moreover, familiarity was identified as a significant factor for both skill providers and receivers. These findings generated practical propositions for service designers to foster the further diffusion of skill-sharing services.

Partial least squares structural equation modeling (PLS-SEM) has become more popular across many disciplines including health care. However, articles in health care often fail to discuss the choice of PLS-SEM and robustness testing is not undertaken. This article presents the steps to be followed in a thorough PLS-SEM analysis, and includes a conceptual comparison of PLS-SEM with the more traditional covariance-based structural equation modeling (CB-SEM) to enable health care researchers and policy makers make appropriate choices. PLS-SEM allows for critical exploratory research to lay the groundwork for follow-up studies using methods with stricter assumptions. The PLS-SEM analysis is illustrated in the context of residential aged care networks combining low-level and high-level care. Based on the illustrative setting, low-level care does not make a significant contribution to the overall quality of care in residential aged care networks. The article provides key references from outside the health care literature that are often overlooked by health care articles. Choosing between PLS-SEM and CB-SEM should be based on data characteristics, sample size, the types and numbers of latent constructs modelled, and the nature of the underlying theory (exploratory versus advanced). PLS-SEM can become an indispensable tool for managers, policy makers and regulators in the health care sector.

This study focuses on the predictive power of linguistic (i.e., general English proficiency; identified simply as “proficiency” in this paper) and non-linguistic (i.e., language learning anxiety and self-regulation) factors on the academic success of English medium instruction (EMI) students studying in engineering and social sciences programs in a Turkish university setting. Data were collected from 705 conveniently sampled EMI students of four academic subjects (international relations; N = 158; business administration; N = 184; mechatronics engineering; N = 181; mechanical engineering: N = 182) representing two disciplines (i.e., social sciences and engineering) from a public university. Pearson correlation and SEM analyses were run to determine the relationships among language learning anxiety, self-regulation, proficiency and EMI success. Findings revealed that anxiety and self-regulation skills do affect EMI students’ proficiency irrespective of academic disciplines. Both self-regulation and proficiency impacted EMI students’ academic success in engineering, while only proficiency predicted academic success in the social sciences. These results are discussed and pedagogical implications are given related to the impact of linguistic and non-linguistic factors in EMI contexts.

The relevance and prominence of the partial least squares structural equation modeling (PLS-SEM) method has recently increased in higher education research, especially in explanatory and predictive studies. We therefore first aim to assess previous PLS-SEM applications by providing a systematic review; second, we aim to highlight and summarize important guidelines for conducting a rigorous PLS-SEM analysis of the current state of results reporting in higher education journals. Specifically, this study focuses on empirical PLS-SEM applications in 14 major higher education journals indexed in the Thomson Reuters Web of Science and in the Elsevier-Scopus databases between 1999 and 2018. We initially identified 49 relevant papers published in 10 higher education journals. Based on these papers’ generally followed guidelines, we thereafter identified various issues related to data screening, model characteristics, measurement model evaluation, structural model evaluation, and the application of state-of-the-art PLS-SEM advanced methods requiring particular attention. Furthermore, we recommend recent guidelines to improve PLS-SEM applications and practices, besides providing specific suggestions regarding utilizing the method’s strength in terms of relevant higher education research questions. Our findings remind researchers, reviewers, and journal editors to remain vigilant, should help them avoid inaccuracies in future publications, and ensure rigor.

The relationship between social network sites and social capital has received much research attention. However, two research gaps can be identified in the existing literature. First, only few studies have examined online social capital as a resource in online social networks. In this regard, it is not clear how to validly measure online social capital. Second, while the factors influencing social capital, among them properties of an individual’s social network, have been investigated in offline settings, such factors have not yet been investigated in terms of online social capital. Addressing these gaps, we asked 1000 Facebook users to provide information on their Facebook usage and online friendship network. Employing structural equation modelling for analysing the survey data, we show that Williams’ Internet Social Capital Scales, which are commonly used to assess social capital in offline settings, can be used to validly measure online social capital. Moreover, we find that some of the variables influencing offline social capital, among them similarity in terms of sociodemographic attributes, seem less important in an online setting.

PurposeOne popular method to assess discriminant validity in structural equation modeling is the heterotrait-monotrait ratio of correlations (HTMT). However, the HTMT assumes tau-equivalent measurement models, which are unlikely to hold for most empirical studies. To relax this assumption, the authors modify the original HTMT and introduce a new consistent measure for congeneric measurement models: the HTMT2.Design/methodology/approachThe HTMT2 is designed in analogy to the HTMT but relies on the geometric mean instead of the arithmetic mean. A Monte Carlo simulation compares the performance of the HTMT and the HTMT2. In the simulation, several design factors are varied such as loading patterns, sample sizes and inter-construct correlations in order to compare the estimation bias of the two criteria.FindingsThe HTMT2 provides less biased estimations of the correlations among the latent variables compared to the HTMT, in particular if indicators loading patterns are heterogeneous. Consequently, the HTMT2 should be preferred over the HTMT to assess discriminant validity in case of congeneric measurement models.Research limitations/implicationsHowever, the HTMT2 can only be determined if all correlations between involved observable variables are positive.Originality/valueThis paper introduces the HTMT2 as an improved version of the traditional HTMT. Compared to other approaches assessing discriminant validity, the HTMT2 provides two advantages: (1) the ease of its computation, since HTMT2 is only based on the indicator correlations, and (2) the relaxed assumption of tau-equivalence. The authors highly recommend the HTMT2 criterion over the traditional HTMT for assessing discriminant validity in empirical studies.

Background: Our purpose is to clarify whether psychological distress (PD) affects the rate of continuous sedentary behavior (CSB). Materials and Methods: In this secondary analysis, a sample population of 80 healthy older adults aged 65 years or older participated in a health club of college A from 2016 to 2017. We conducted Structural Equation Modeling (SEM) using the cross-lagged and synchronous effects models. We adopted the following as proxy variables: CSB (based on the ratio of 1.5 METs sessions or more continuing for over 30 min) CSB and PD (based on the Kessler psychological distress scale: K6). Results: “2016 K6” had a significant influence on “2017 CSB” (standardization factor (β) = 0.136, p = 0.020) using the cross-lagged effects model, and “2017 K6” significantly influenced “2017 CSB” (β = 0.166, p = 0.039) using the synchronous effects model. Fit indices were Adjusted Goodness-of-Fit Index (AGFI) = 0.990, Confirmatory Fit Index (CFI) = 1.000, and Root Mean Square Error of Approximation (RMSEA) = 0.000. Conclusion: The results suggest that PD may affect the ratio of CSB one year later.

Discriminant validity assessment has become a generally accepted prerequisite for analyzing relationships between latent variables. For variance-based structural equation modeling, such as partial least squares, the Fornell-Larcker criterion and the examination of cross-loadings are the dominant approaches for evaluating discriminant validity. By means of a simulation study, we show that these approaches do not reliably detect the lack of discriminant validity in common research situations. We therefore propose an alternative approach, based on the multitrait-multimethod matrix, to assess discriminant validity: the heterotrait-monotrait ratio of correlations. We demonstrate its superior performance by means of a Monte Carlo simulation study, in which we compare the new approach to the Fornell-Larcker criterion and the assessment of (partial) cross-loadings. Finally, we provide guidelines on how to handle discriminant validity issues in variance-based structural equation modeling.