Content uploaded by Wynne W. Chin
Author content
All content in this area was uploaded by Wynne W. Chin on Apr 04, 2014
Content may be subject to copyright.
The Partial Least Squares Approach to Structural Equation Modeling
Abstract
Provides a nontechnical introduction to the partial least squares (PLS) approach. As a logical base for comparison, the PLS approach for structural path estimation is contrasted to the covariance-based approach. In so doing, a set of considerations are then provided with the goal of helping the reader understand the conditions under which it might be reasonable or even more appropriate to employ this technique. This chapter builds up from various simple 2 latent variable models to a more complex one. The formal PLS model is provided along with a discussion of the properties of its estimates. An empirical example is provided as a basis for highlighting the various analytic considerations when using PLS and the set of tests that one can employ is assessing the validity of a PLS-based model. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
... El instrumento se aplicó directamente a los alumnos que cumplen con los requisitos mencionados en esta investigación con ayuda de los Responsables de Programas de cada universidad se ubicaron rapidamente. Para analizar los resultados y comprobar la hipótesis estadísticamente de la investigación, se utilizó el método de Ecuaciones Estructurales, ya que la bonanza de utilizarlo es que no exige la normalidad en los datos recolectados, Chin (1998) ...
... Tabla 3. La fiabilidad compuesta se considera que es un indicador más preciso que al alpha de cronbach (Chin, 1998), ya que no todos los indicadores reciben la misma ponderación. También se visualiza en la tabla 3, que las variables del modelo estructural todas explican más del 50% de su varianza (AVE). ...
... Los índices a considerar en este cálculo de las cargas deben ser ≥ 0.7 para considerarse confiables en el modelo estructural, en la figura 2 se puede apreciar que las cargas de cada items por variable cumplen con el índice según Chin (1998). Con respecto a los valores de Beta (β), o Rutas Path, en el modelo estructural de la figura 2, se determina que la influencia que pueden tener las variables independientes (X) en la variable (Y), no cumplen con los índices establecidos por Chin (1998), que indica que las cargas deben ser β ≥ 0.2 donde las variables Norma Subjetiva y Control del Comportamiento están por debajo de este índice y no ofrecen una explicación confiable de su relación con la variable dependiente (intención emprendedora). ...
The purpose of the study wasto analyze the relationship between the Theory of Applied Behavior and Entrepreneurial Education with Entrepreneurial Intention. This study is quantitative, descriptive and correlational, with the application of an instrument adapted from Fernández and Quintero (2017), to a sample of 325 students of careers for Administration purposes who have taken the entrepreneurship course. Of the variables that were studied, the one that has the strongest impact on Entrepreneurial Intention is the variable of Entrepreneurial Education with β0.290, Studen’s T 3.558 and P value of 0.000, this being positive and direct; Therefore, the role of the university in future entrepreneurs is very important and they must commit to reviewing their entrepreneurship courses for evaluation and updating.
... First, we examined measurement models. In the second stage, we evaluated the structural model by testing the reliability and validity to determine the relationships between constructs in the path model (Chin, 1998;Hair et al., 2016;Ringle and Sinkovics, 2009). ...
... According to Table 2, all reflective internal consistency reliability constructs have values larger than 0.785 and 0.861 for CA and CR, respectively. These scores exceed the recommended 0.7 required levels (Chin, 1998). To determine the reliability of an indicator, it is necessary to ensure that a construct explains at least 50% of the variation in each of its related indicators (Chin, 1998). ...
... These scores exceed the recommended 0.7 required levels (Chin, 1998). To determine the reliability of an indicator, it is necessary to ensure that a construct explains at least 50% of the variation in each of its related indicators (Chin, 1998). With reflective constructs, the indicator loadings are generally higher than the required threshold of 0.7 and statistically significant (Chin, 1998;Gefen et al., 2000;Henseler et al., 2015) This provides evidence of the indicators' reliability and confirms that they meet the criteria for internal consistency reliability. ...
The previous research emphasized the importance of service quality, brand image, and customer loyalty in the context of the hospitality industry, However, there seems to be a need for more in-depth investigation into the specific mechanisms and strategies of service quality within the nature tourist camps that contribute to building a positive brand image and fostering customer loyalty. To cover this gap in knowledge, this study examined customer loyalty and brand image perception about nature tourist camps service quality in Jordan. This research is crucial due to the rapid growth of the outdoor accommodation industry and the importance of service quality. Partial Least Squares Structural Equation Modeling (PLS-SEM) was used to analyze the data collected from tourists who stayed at Wadi Rum. The results demonstrate that brand image significantly mediates the relationship between SERVRQUAL model dimensions and customer loyalty. Theoretical and practical implications were discussed.
... This serves to substantiate the discriminant validity of the model, aligning with the criteria outlined by Fornell and Larcker (1981) as well as Hair et al. (2017). The results outlined in the previous table affirm the discriminant validity of the model, as endorsed by Chin (1998). This is because all elements of the variables exhibit stronger loading on their respective constructs, surpassing their loading on any other variables within the model's constructs. ...
... The coefficient of determination test (R²) was employed to estimate the model's predictive capacity in explaining the percentage of variance observed in the dependent variable using the independent variable. Results indicated that the variance attributed to the independent variable "SL" on the dependent variable "EP" was high, while on the dependent variable "AC" it was moderate, based on the threshold established by Chin (1998) (Table 4). ...
This research endevors to investigate the interrelationships among spiritual leadership (SL), affective commitment (AC) and employee performance (EP) in the Egyptian hotel industry. Employing a quantitative research approach, the study distributed self-administered surveys to a randomly chosen group of employees. Analyzing 623 valid survey responses using structural equation modelling (SEM), the research revealed significant positive correlations between SL and both EP and AC, with AC also showing a noteworthy positive effect on EP. Moreover, the study found that AC acts as a mediator in the relationship between SL and EP.
... See Tables 2 and 3, and Figure 2. The results at Table 2 show that each variable explain the variance of its items better than other variables that that build the model, which support the discriminant validity of the model, as recommended by Fornell & Larcker (1981) and Hair et al. (2017). Also results of data analysis show that the items loadings of each variable is greater than its loading on all other variables of the model, which support the discriminant validity of the model, as recommended by Chin (1998). ...
... In order to measure the ability of the model to predict the variance ratio occurring in the dependent variable by the independent variable, the coefficient of determination (R² ) was utilized. Table 3 shows that the variance occurring by the independent variables "GPT-PEU"; "GPT-PU"; "GPT-SQ"; "GPT-PT"; "GPT-PH"; and "GPT-AN" on the dependent variable "TS", and the variance occurring by the independent variable "TS" on the dependent variable "GPT-UI" are high, and exceed the cut-off point which recommended by Chin (1998), as both results were over 0.67. Table 3. R 2 of the endogenous latent variables. ...
This study aims at predicting the interrelationship between among Chat GPT with its six dimensions, tourist’s satisfaction and Chat GPT usage intention as perceived by tourist, and as well as to examine the moderating effect of traditional tour operator services on the relationships between all the variables. Data were collected from 624 tourists. The study hypotheses were tested and the direct and indirect effects between variables were examined using the PLS-SEM. The SEM results showed that Chat GPT’s six dimensions have a positive and significant direct impact on tourist’s satisfaction, and emphasis the moderating role of Traditional Tour Operator Services “TTOS” on the relationship between GPT’s six dimensions and “TS”, and on the relationship between ‘TS” and Chat GPT usage intention. These findings yield valuable insights for everyone interested in the use of IT in the tourism industry, and provide effective strategies for optimizing the use of technological applications by traditional tour operators.
... We obtain and acquire cross-loading by correlating the component scores of each latent variable with other items. We can infer that the model has discriminant validity if each indicator loads higher on its assigned construct than on any other construct, and each construct loads the highest on its own item [22]. The loading values for the bolded portions of the table are always greater than the cross-loadings, shown in Table 2, thus indicating that discriminant validity has been established. ...
... The value of the R square ranges from 0 to 1, and the larger the value, the higher the accuracy of prediction. If this value is around or above 0.67, the impact is considered substantial; when the value is near 0.33, the impact is considered moderate, and if the value is near 0.19, the impact is considered weak [22]. Table 5 shows the R square value of consumer trust is 0.591, that is, around 0.67, stating the effect of effort expectancy is 59.1% on consumer trust. ...
This study is primarily aimed at examining the influence of effort expectancy on consumers’ purchase intention within a Business-to-Consumer (B2C) cross-border e-commerce (CBEC) consumption market in China. With the improvement of the living standards for the Chinese consumers, the B2C imported cross-border e-commerce platforms have become the choice of consumers for online shopping. On the basis of the seminal UTAUT Model, consumer trust is proposed as a potential mediator for the relationship between effort expectancy and purchase intention. By using a non-probability sampling technique, an online questionnaire survey distribution was employed with an overall return rate of 108 usable feedback. Data were analysed using the SmartPLS Version 4.0.9.9 statistical software. Results revealed a positive and significant influence of effort expectancy on consumers’ purchasing intentions. Furthermore, we found that consumer trust significantly mediates the relationship between effort expectancy and purchase intention. These findings could assist imported cross-border e-commerce platform enterprises towards enhancing the convenience of online shopping on these platforms and improving consumers' trust in online shopping activities. This study contributes to promoting the platform enterprises to improve the convenience of the whole process of platform operation and purchase, and is conductive to paying attention to the consumer trust to enhance the enterprise’s reputation, which will help develop customer stickiness and cultivate loyal customers.
... This choice is informed by several advantages associated with PLS in comparison to alternative analytical techniques. First, PLS does not necessitate the assumption of a normal distribution of data, as highlighted by Chin (1998). Second, PLS is particularly well-suited for investigations about theoretical development, as emphasized by Hair et al. (2011). ...
Underling the key motivational factors of tourists are very significant in tourism destination marketing. The current study aimed to introduce a new insight in destination management through exploring the main motivational factors behind foreign tourists to visit Saudi. Self administrative questionnaire was distributed among a random sample of tourists visited Saudi. Structural Equation Modeling (SEM-PLS) was applied to analysis the data. The results indicated that five motivational factors influence on behavior intention to visit Saudi. However, the model applied in the study has the potential to be used in similar studies conducted in various destinations worldwide.
... An LF value above 0.70 indicates that the variable is realized by the indicator more than 50%. Chin (1998) stated that the LF value should be at least 0.60 to be acceptable, so correlations with a value of less than 0.60 were eliminated from the measurement model. Meanwhile, the AVE indicates the correlation of indicators (measurement items) with variables. ...
In forming an audit opinion, some aspects need to be considered by state auditors. This study aims to investigate the impact of integrity, independence, professional skepticism, and audit situation on audit opinion in the context of auditing state institutions by state auditors. Cluster proportional sampling was used to sample 52 auditors who worked at the Directorate General of Audit I, BPK RI. The analysis was conducted using structural equation modeling and partial least squares with SmartPLS 4 software. The results revealed that integrity and independence positively affected audit opinion. Professional skepticism and audit situations do not affect audit opinion. The results will provide input to the BPK RI in guiding the development of auditor competencies to improve the quality of audit results. This study concludes that auditors believe that when they demonstrate integrity and independence, the accuracy of audit opinion will increase. This is because, with these qualities, an auditor will have courage, honesty, and responsibility and will not be unduly influenced. Although previous studies have extensively examined the impact of these factors in the context of audit firms auditing commercial companies, this is a novel study in the context of audits conducted by state auditors for state institutions.
... A higher R-squared value indicates a better fit of the model to the data. According to Chin (1998), an R-squared value of 33 to 67% is considered moderate, while a value of 67-75% is considered moderate-strong. In this research, the predictors of customer satisfaction have a moderate explanatory power, accounting for 62% of the variation in customer satisfaction. ...
Recently, driven by the spread of advanced digital platforms and social media, many traditional brick-and-mortar retailers have ventured into the online marketplace to expand their sales channels. This research explores the crucial factors shaping customer satisfaction in local online retail. Furthermore, it examines how this level of satisfaction influences customers' intentions to repurchase within the online sales channel. This study collected primary data by administering questionnaires to 410 customers. The research employs both Maximum Likelihood and Bayesian Structural Equation Modeling to measure the causal pathways of the outlined in the research model. The study's findings indicate that information availability, perceived price, perceived quality, customer support, time-saving, and delivery services all have a positive and significant influence on customer satisfaction, with statistical significance at p < .05. Together, these exogenous variables account for 63% of the variation in customer satisfaction. Furthermore, the results demonstrate that customer satisfaction plays a significant and positive role in influencing customer repurchase intention in the online sales channel, explaining a substantial 74% of the variance in repurchase intention. The study's findings highlight the importance for local retailers to acknowledge and prioritize the factors contributing to customer satisfaction. To achieve this, retailers must ensure that all product information, images, and terms and conditions are not only accurate but also clear and comprehensive. This way, customers have a precise understanding of what products they are purchasing, the quantity, and the details of the delivery process, including when and where they can expect to receive their orders.
... Structural equation modeling (SEM) is a common method for testing models with latent variables (Chin, 1998). Based on similar previous studies (Wan et al., 2018;Li et al., 2023), this study used partial least squares structural equation modeling (PLS-SEM) for statistical tests. ...
Cultural heritage is an important contributor to sustainable urban development, and its conservation is considered a global task. Previous research has primarily focused on the conservation of heritage buildings within architectural contexts and world heritage sites within geographical contexts. However, there is a notable gap in conservation of green cultural heritage (GCH) within a landscape-based framework. This study used the extended theory of planned behavior to explore the factors and pathways that influence the public's behavioral intentions toward GCH conservation. Based on 1075 questionnaires collected in Tokyo, Japan, the hypotheses of the newly constructed theoretical model were tested using partial least squares structural equation modeling. The results show that perceived behavioral control has the greatest positive influence on citizens' GCH conservation intentions and behaviors, while social norms do not influence people's willingness to conserve. In the extended factors, the environmental awareness which including heritage awareness and cultural attachment, significantly influences the range of public attitudes towards conservation, while environmental perception, including perceived usefulness and perceived quality, significantly affects the public's conservation intentions and behaviors, respectively. Our findings have implications for local governments and policymakers to enhance public participation in GCH, as well as some several new theoretical interests for further studies.
The integration of technology has demonstrated its capacity to enhance organizational operations and foster economic development. Just as in various other sectors, the maritime industry also demands the digitalization of its operations through an efficient and seamless IT system – known as the Port Community System (PCS). The PCS serves as a secure technological platform that facilitates the exchange of data and information flow among the myriad of public and private entities involved in seaport operations. This research endeavors to explore the disparities between the existing and proposed procedures at Port Qasim, with a particular focus on the implementation of the PCS within the framework of Port Qasim Authority. Employing structural equation modeling via PLS-SEM, the study collected data through a Likert questionnaire featuring a 5-point scale. This questionnaire was administered to 123 respondents, all of whom were esteemed port professionals at Port Qasim. The findings unveiled that the perceived impact of the introduction of the PCS holds a positive and significant influence over overall process efficiency and workflow. Correspondingly, the pre-requisite knowledge required is found to have a positive and significant relationship on perceived impact of PCS introduction at the port. Nonetheless, the correlation between the prerequisite knowledge and process efficiency was deemed insignificant. This implies that possessing prior knowledge of PCS doesn't inherently translate to increased process efficiency. Interestingly, this gap can be bridged effectively through targeted training sessions, enabling users to harness the full potential of the PCS without adversely affecting process efficiency. Nevertheless, with streamlined data exchange and efficient information flow, the PCS can reduce port congestion, enhancing trade operations and positioning Karachi as a competitive trade hub. Economically, this translates into increased trade throughput, improved logistics, and heightened business attractiveness.
Canonical correlation analysis is commonly considered to be a general model for most parametric bivariate and multivariate statistical methods. Because of its capability for handling multiple criteria and multiple predictors simultaneously, canonical correlation analysis has a great deal of appeal and has also enjoyed increasing application in the behavioral sciences. However, it has also been plagued by several serious shortcomings. In particular, researchers have been unable to determine the statistical significance of individual parameter estimates or to relax assumptions of the canonical model that are inconsistent with theory and/or observed data. As a result, canonical correlation analysis has found more application in exploratory research than in theory testing. This paper illustrates how these problems can be resolved by expressing canonical correlation as a special case of a linear structural relations model.