Identifying and treating unobserved heterogeneity with FIMIX-PLS: part I – method

2016 ◽  
Vol 28 (1) ◽  
pp. 63-76 ◽  
Author(s):  
Joe F. Hair, Jr. ◽  
Marko Sarstedt ◽  
Lucy M Matthews ◽  
Christian M Ringle
Keyword(s):  
Structural Equation ◽  
Latent Class ◽  
State Of The Art ◽  
Unobserved Heterogeneity ◽  
Information Criteria ◽  
Equation Modeling ◽  
Content Type ◽  
Measurement Models ◽  
Research Problems ◽  
European Business

Purpose – The purpose of this paper is to provide an overview of unobserved heterogeneity in the context of partial least squares structural equation modeling (PLS-SEM), its prevalence and challenges for social science researchers. Part II – in the next issue (European Business Review, Vol. 28 No. 2) – presents a case study, which illustrates how to identify and treat unobserved heterogeneity in PLS-SEM using the finite mixture PLS (FIMIX-PLS) module in the SmartPLS 3 software. Design/methodology/approach – The paper merges literatures from various disciplines, such as management information systems, marketing and statistics, to present a state-of-the-art review of FIMIX-PLS. Based on this review, the paper offers guidelines on how to apply the technique to specific research problems. Findings – FIMIX-PLS offers a means to identify and treat unobserved heterogeneity in PLS-SEM and is particularly useful for determining the number of segments to extract from the data. In the latter respect, prior applications of FIMIX-PLS restricted their focus to a very limited set of criteria, but future studies should broaden the scope by considering information criteria, theory and logic. Research limitations/implications – Since the introduction of FIMIX-PLS, a range of alternative latent class techniques have emerged to address some of the limitations of the approach relating, for example, to the technique’s inability to handle heterogeneity in the measurement models and its distributional assumptions. The second part of this article (Part II) discusses alternative latent class techniques in greater detail and calls for the joint use of FIMIX-PLS and PLS prediction-oriented segmentation. Originality/value – This paper is the first to offer researchers who have not been exposed to the method an introduction to FIMIX-PLS. Based on a state-of-the-art review of the technique in Part I, Part II follows up by offering a step-by-step tutorial on how to use FIMIX-PLS in SmartPLS 3.

Identifying and treating unobserved heterogeneity with FIMIX-PLS

2016 ◽  
Vol 28 (2) ◽  
pp. 208-224 ◽  
Author(s):  
Lucy M. Matthews ◽  
Marko Sarstedt ◽  
Joseph F. Hair ◽  
Christian M. Ringle
Keyword(s):  
Structural Equation ◽  
Latent Class ◽  
Structural Model ◽  
Unobserved Heterogeneity ◽  
Equation Modeling ◽  
Content Type ◽  
Measurement Models ◽  
European Business ◽  
Number Of Segments

Purpose Part I of this article (European Business Review, Volume 28, Issue 1) offered an overview of unobserved heterogeneity in the context of partial least squares structural equation modeling (PLS-SEM), its prevalence and challenges for social sciences researchers. This paper aims to provide an example that explains how to identify and treat unobserved heterogeneity in PLS-SEM by using the finite mixture PLS (FIMIX-PLS) module in the SmartPLS 3 software (Part II). Design/methodology/approach This case study illustrates the application of FIMIX-PLS using a popular corporate reputation model. Findings The case study demonstrates the capability of FIMIX-PLS to identify whether unobserved heterogeneity significantly affects structural model relationships. Furthermore, it shows that FIMIX-PLS is particularly useful for determining the number of segments to extract from the data. Research limitations/implications Since the introduction of FIMIX-PLS, a range of alternative latent class techniques has appeared. These techniques address some of the limitations of the approach relating to, for example, its failure to handle heterogeneity in measurement models, or its distributional assumptions. This research discusses alternative latent class techniques and calls for the joint use of FIMIX-PLS and PLS prediction-oriented segmentation. Originality/value This article is the first to offer researchers, who have not been exposed to the method, an introduction to FIMIX-PLS. Based on a state-of-the-art review of the technique, the paper offers a step-by-step tutorial on how to use FIMIX-PLS by using the SmartPLS 3 software.

Rethinking balanced scorecard (BSC) measures: formative versus reflective measurement models

2017 ◽  
Vol 66 (1) ◽  
pp. 92-110 ◽  
Author(s):  
Sungbum Park ◽  
Heeseok Lee ◽  
Seong Wook Chae
Keyword(s):  
Balanced Scorecard ◽  
Latent Variables ◽  
Structural Equation ◽  
Model Comparison ◽  
Empirical Studies ◽  
Equation Modeling ◽  
Content Type ◽  
Formative Measurement ◽  
Measurement Models ◽  
Formative Indicators

Purpose Most empirical balanced scorecard (BSC) studies have shown a tendency to wrongly employ reflective indicators instead of the more theoretically suitable formative indicators. However, formative indicators are difficult to apply due to the lack of statistical software support and a standardized model testing method. The paper aims to discuss these issues. Design/methodology/approach This study empirically compares the reflective and formative measurement method with standardized model comparison criteria. After collecting 217 valid questionnaires from companies in South Korea, the authors applied a structural equation modeling technique to analyze the data. Findings The result shows that the formative measure provides greater validity for the corporate performance measurement using BSC. Further, this study shows the indicators’ relative influence on each BSC perspectives using the formative measure. Practical implications This study proved the usefulness of the formative measure analysis method and suggested its practical use, focusing on the indicators most useful in developing corporate strategies. In addition, the authors showed that formative indicators could be used in the corporate environment by overcoming the limitations of conventional studies that were confined to causal relationships with latent variables. Originality/value This study may be the pioneering work that compares formative and reflective indicators simultaneously, addressing the usefulness of formative measurement and its application validity in the existing empirical studies using reflective measurements.

Expected and experienced quality as predictors of intention to purchase four new processed beef products

2014 ◽  
Vol 116 (3) ◽  
pp. 451-471 ◽  
Author(s):  
Faiza Saeed ◽  
Klaus G. Grunert
Keyword(s):  
Structural Equation Modeling ◽  
Purchase Intention ◽  
Structural Equation ◽  
Causal Structure ◽  
Purchase Intentions ◽  
Equation Modeling ◽  
Content Type ◽  
Measurement Models ◽  
Latent Constructs ◽  
Beef Products

Purpose – This paper aims to explore consumers' perception of quality of new processed beef products and the role of expected and experienced quality in the formation of consumer's purchase intentions. Based on the Total Food Quality Model, a conceptual framework is developed that relates cue evaluation, expected quality, experienced quality, purchase motive fulfilment and purchase intention. Design/methodology/approach – Structural equation modeling is used to test the framework with data from a sample of 201 respondents, involving three steps. First, principal component analyses were applied to explore underlying factor structures within each construct. Based on the exploratory factor analyses, measurement models were estimated, with the measured variables as indicators of latent constructs for all the four products. Finally, structural models were estimated for the relationships among the latent constructs. Findings – Results show that cue evaluations, expected/experienced quality and purchase motive fulfilment are all predictors of purchase intention, but that their weight and causal structure differ between purchase intentions before and after trial. Practical implications – Implications for the introduction of new beef products are discussed. Originality/value – This paper is an attempt to quantitatively estimate the relationships between quality cues, expected and experienced quality, and purchase motives as determinants of purchase intention for new products using structural equation modeling.

Analyzing heterogeneity on the value, satisfaction, word-of-mouth relationship in retailing

2017 ◽  
Vol 55 (7) ◽  
pp. 1558-1577 ◽  
Author(s):  
María Fuentes-Blasco ◽  
Beatriz Moliner-Velázquez ◽  
Irene Gil-Saura
Keyword(s):  
Customer Satisfaction ◽  
Customer Value ◽  
Structural Equation ◽  
Word Of Mouth ◽  
Unobserved Heterogeneity ◽  
Behavioral Outcomes ◽  
Marketing Strategies ◽  
Equation Modeling ◽  
Content Type ◽  
Dynamic Markets

Purpose The literature recognizes the need to study differences in consumer behavior in highly competitive and dynamic markets. In this paper, the authors look at how the heterogeneous evaluation of retailing influences customer satisfaction and loyalty. The purpose of this paper is to analyze unobserved heterogeneity on customer value dimensions perceptions in retail establishments, and their potential effects on positive forms of behavioral outcomes considering customer satisfaction as a mediating variable. Design/methodology/approach On a sample of 820 retail customers, the authors apply a finite mixture structural equation modeling that analyzes unobserved heterogeneity simultaneously. In this model, the authors study the influence of heterogeneous perceptions of excellence, efficiency, entertainment and aesthetics on customer satisfaction and of satisfaction on word-of-mouth (WOM) referral and WOM activity. Findings The results show two latent segments where the intensity of causal relations varies, which means that the effect of value dimensions and satisfaction are over or underestimated when heterogeneity is ignored. Originality/value The main value of the paper has been to analyze the potential heterogeneity of value dimensions (intravariable approach), and their links with satisfaction and some dimensions of loyalty (intervariable approach). Customer heterogeneity must be studied to understand the satisfaction process and WOM responses in order to design more efficient and effective relationship marketing strategies.

When to use and how to report the results of PLS-SEM

2019 ◽  
Vol 31 (1) ◽  
pp. 2-24 ◽  
Author(s):  
Joseph F. Hair ◽  
Jeffrey J. Risher ◽  
Marko Sarstedt ◽  
Christian M. Ringle
Keyword(s):  
Structural Equation ◽  
Statistical Power ◽  
Goodness Of Fit ◽  
Model Comparison ◽  
Latent Class ◽  
Evaluation Criteria ◽  
Secondary Data ◽  
Equation Modeling ◽  
Content Type ◽  
Rules Of Thumb

Purpose The purpose of this paper is to provide a comprehensive, yet concise, overview of the considerations and metrics required for partial least squares structural equation modeling (PLS-SEM) analysis and result reporting. Preliminary considerations are summarized first, including reasons for choosing PLS-SEM, recommended sample size in selected contexts, distributional assumptions, use of secondary data, statistical power and the need for goodness-of-fit testing. Next, the metrics as well as the rules of thumb that should be applied to assess the PLS-SEM results are covered. Besides presenting established PLS-SEM evaluation criteria, the overview includes the following new guidelines: PLSpredict (i.e., a novel approach for assessing a model’s out-of-sample prediction), metrics for model comparisons, and several complementary methods for checking the results’ robustness. Design/methodology/approach This paper provides an overview of previously and recently proposed metrics as well as rules of thumb for evaluating the research results based on the application of PLS-SEM. Findings Most of the previously applied metrics for evaluating PLS-SEM results are still relevant. Nevertheless, scholars need to be knowledgeable about recently proposed metrics (e.g. model comparison criteria) and methods (e.g. endogeneity assessment, latent class analysis and PLSpredict), and when and how to apply them to extend their analyses. Research limitations/implications Methodological developments associated with PLS-SEM are rapidly emerging. The metrics reported in this paper are useful for current applications, but must always be up to date with the latest developments in the PLS-SEM method. Originality/value In light of more recent research and methodological developments in the PLS-SEM domain, guidelines for the method’s use need to be continuously extended and updated. This paper is the most current and comprehensive summary of the PLS-SEM method and the metrics applied to assess its solutions.

Measurement scales for technology-generated customer contact

2017 ◽  
Vol 37 (5) ◽  
pp. 534-556 ◽  
Author(s):  
Raghunath Rudran ◽  
Ajith Kumar J.
Keyword(s):  
Structural Equation ◽  
Online Shopping ◽  
Service Providers ◽  
Equation Modeling ◽  
Measurement Scales ◽  
Customer Contact ◽  
Content Type ◽  
Measurement Models ◽  
The Government

Purpose The purpose of this paper is to develop measurement scales for customer contact in a technology-generated context. Design/methodology/approach The authors adapted the scales of Froehle and Roth (2004), by following a systematic scale adaptation and development process. The adapted scales were tested for psychometric properties and refined by building measurement models using partial least squares structural equation modeling. Findings The authors found it necessary to revise Froehle and Roth’s (2004) original items in most of the scales. After testing, the “attitude towards the episode” scale was dropped and remaining nine scales were retained. Research limitations/implications The scales will be useful to future researchers on online shopping to advance their research. The scales can be tested and validated with data from multiple empirical contexts and adapted to those contexts as necessary. Future studies must examine path relationships between belief, attitude, and intention constructs. Practical implications The adapted scales can be useful to practitioners in the domain of online shopping to measure the beliefs, attitudes, and intentions of their customers. Potential beneficiaries include service providers, service designers, industry associations as well as regulators in the government. Originality/value The overarching contribution of this paper lies in developing scales pertaining to the online shopping context of technology-generated customer contact. The paper has simultaneously addressed two relatively less attended areas of research on service operations – the role of technology in customer contact and measurement of customer contact.

Leadership and quality management measurement models: an empirical study

2019 ◽  
Vol 33 (1) ◽  
pp. 52-66
Author(s):  
Natkamol Chansatitporn ◽  
Vallerut Pobkeeree
Keyword(s):  
Factor Analysis ◽  
Longitudinal Study ◽  
Quality Management ◽  
Structural Equation ◽  
Equation Modeling ◽  
Future Research ◽  
Content Type ◽  
Leadership Attributes ◽  
Staff Members ◽  
Measurement Models

Purpose The purpose of this paper is to explore, confirm and verify leadership with regards to quality management measurement models. This research focused on identifying individual staff members’ leadership attributes at the Thai National Institute of Health in relation to quality management. Design/methodology/approach The research instrument used in this study was a modified questionnaire on self-leadership and quality management that was distributed to the institute’s staff. Leadership and quality management construct variables were observed and measured through staff perceptions, attitudes, practices and existing facts at the institute. Exploratory factor analysis (EFA), confirmatory factor analysis (CFA) and structural equation modeling (SEM) were used to examine the data. Findings The questionnaire had a 65 percent response rate. EFA revealed six factors from 27 questionnaire items and CFA was used to confirm the measurement models that were fitted to the data. The leadership attributes of staff members at the institute were statistically associated to and impacted on quality management by SEM analysis. Research limitations/implications In-depth understanding of leadership and quality management could be done through a longitudinal study because the two factors would change over time. Even though this model is not a longitudinal study, it could help the institute facilitate and manage quality in practice through leadership. Originality/value A cross-sectional study is used to examine the effect of leadership on quality management through factor analysis and SEM, which provided empirical evidence for future research. Leadership and quality management measurement models have statistically proven to be appropriately, technically and theoretically correct by design for observing variables used in the leadership measurement model that affects quality management.

scholarly journals HTMT2–an improved criterion for assessing discriminant validity in structural equation modeling

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Ellen Roemer ◽  
Florian Schuberth ◽  
Jörg Henseler
Keyword(s):  
Structural Equation Modeling ◽  
Latent Variables ◽  
Structural Equation ◽  
Discriminant Validity ◽  
Empirical Studies ◽  
Geometric Mean ◽  
Equation Modeling ◽  
Content Type ◽  
Measurement Models ◽  
Loading Patterns

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.

Engaging leadership in the job demands-resources model

2015 ◽  
Vol 20 (5) ◽  
pp. 446-463 ◽  
Author(s):  
Wilmar B. Schaufeli
Keyword(s):  
Structural Equation ◽  
Work Performance ◽  
Online Survey ◽  
Job Demands ◽  
Well Being ◽  
Organizational Outcomes ◽  
Job Resources ◽  
Equation Modeling ◽  
Cross Sectional ◽  
Content Type

Purpose – The purpose of this paper is to integrate leadership into the job demands-resources (JD-R) model. Based on self-determination theory, it was argued that engaging leaders who inspire, strengthen, and connect their followers would reduce employee’s levels of burnout and increase their levels of work engagement. Design/methodology/approach – An online survey was conducted among a representative sample of the Dutch workforce (n=1,213) and the research model was tested using structural equation modeling. Findings – It appeared that leadership only had an indirect effect on burnout and engagement – via job demands and job resources – but not a direct effect. Moreover, leadership also had a direct relationship with organizational outcomes such as employability, performance, and commitment. Research limitations/implications – The study used a cross-sectional design and all variables were based on self-reports. Hence, results should be replicated in a longitudinal study and using more objective measures (e.g. for work performance). Practical implications – Since engaged leaders, who inspire, strengthen, and connect their followers, provide a work context in which employees thrive, organizations are well advised to promote engaging leadership. Social implications – Leadership seems to be a crucial factor which has an indirect impact – via job demands and job resources – on employee well-being. Originality/value – The study demonstrates that engaging leadership can be integrated into the JD-R framework.

Age and gender differences in socially aversive (“dark”) personality traits

2021 ◽  
pp. 089020702098843
Author(s):  
Johanna Hartung ◽  
Martina Bader ◽  
Morten Moshagen ◽  
Oliver Wilhelm
Keyword(s):  
Personality Traits ◽  
Structural Equation ◽  
Age Groups ◽  
Dark Triad ◽  
Equation Modeling ◽  
Model Parameters ◽  
Age And Gender ◽  
Measurement Models ◽  
And Gender ◽  
Latent Means

The strong overlap of personality traits discussed under the label of “dark personality” (e.g., psychopathy, spitefulness, moral disengagement) endorses a common framework for socially aversive traits over and beyond the dark triad. Despite the rapidly growing research on socially aversive traits, there is a lack of studies addressing age-associated differences in these traits. In the present study ( N = 12,501), we investigated the structure of the D Factor of Personality across age and gender using local structural equation modeling, thereby expressing the model parameters as a quasi-continuous, nonparametric function of age. Specifically, we evaluated loadings, reliabilities, factor (co-)variances, and means across 35 locally weighted age groups (from 20 to 54 years), separately for females and males. Results indicated that measurement models were highly stable, thereby supporting the conceptualization of the D factor independent of age and gender. Men exhibited uniformly higher latent means than females and all latent means decreased with increasing age. Overall, D and its themes were invariant across age and gender. Therefore, future studies can meaningfully pursue causes of mean differences across age and between genders.

Sign in / Sign up

Export Citation Format

Share Document