Evaluating Structural Equation Models with Unobservable Variables and Measurement Error

Published on Feb 1, 1981in Journal of Marketing Research
· DOI :10.2307/3151312
Claes Fornell50
Estimated H-index: 50
(UM: University of Michigan),
David F. Larcker81
Estimated H-index: 81
(NU: Northwestern University)
The statistical tests used in the analysis of structural equation models with unobservable variables and measurement error are examined. A drawback of the commonly applied chi square test, in addit...
📖 Papers frequently viewed together
1,821 Citations
1 Author (Rex B. Kline)
28.7k Citations
32.1k Citations
Factor analysis, path analysis, structural equation modeling, and related multivariate statistical methods are based on maximum likelihood or generalized least squares estimation developed for covariance structure models. Large-sample theory provides a chi-square goodness-of-fit test for comparing a model against a general alternative model based on correlated variables. This model comparison is insufficient for model evaluation: In large samples virtually any model tends to be rejected as inade...
13.2k CitationsSource
#1Robert I. Jennrich (UCLA: University of California, Los Angeles)H-Index: 31
#2D. B. Clarkson (MU: University of Missouri)H-Index: 1
A jackknife-like procedure is developed for producing standard errors of estimate in maximum likelihood factor analysis. Unlike earlier methods based on information theory, the procedure developed is computationally feasible on larger problems. Unlike earlier methods based on the jackknife, the present procedure is not plagued by the factor alignment problem, the Heywood case problem, or the necessity to jackknife by groups. Standard errors may be produced for rotated and unrotated loading estim...
19 CitationsSource
#1Ronald D. Anderson (IU: Indiana University)H-Index: 12
#2Jack L. Engledow (IU: Indiana University)H-Index: 5
Last. Helmut Becker (UP: University of Portland)H-Index: 11
view all 3 authors...
A simple structural model relating attitudes toward business to product satisfaction, experience in shopping, and search effort is proposed and tested by use of confirmatory factor analysis (COFAMM...
88 CitationsSource
#1Claes Fornell (NU: Northwestern University)H-Index: 50
Although much progress has been made in clarifying the properties of canonical correlation analysis in order to enhance its applicability, there are several remaining problems. Canonical variates do not always represent the observed variables even though the canonical correlation is high. In addition, canonical solutions are often difficult to interpret. This paper presents a method designed to deal with these two problems. Instead of maximizing the correlation between unobserved variates, the s...
18 CitationsSource
#1David A. Aaker (University of California, Berkeley)H-Index: 64
#2Richard P. Bagozzi (University of California, Berkeley)H-Index: 117
Structural equation models which include unobservable variables permit theoretical constructs to be represented by multiple indicators. The use and evaluation of such models are illustrated in an i...
91 CitationsSource
Factor analysis in several populations, covariance structure models, three-mode factor analysis, structural equation systems with measurement model, and analysis of covariance with measurement model are all shown to be specializations of a general moment structure model published previously in this journal. Some new structured linear models are also described; they may be considered either generalizations or special cases of existing models. Simple representations are developed for complex linea...
50 CitationsSource
#1Elliot M. Cramer (UNC: University of North Carolina at Chapel Hill)H-Index: 15
#2W. Alan Nicewander (OU: University of Oklahoma)H-Index: 15
A distinction is drawn between redundancy measurement and the measurement of multivariate association for two sets of variables. Several measures of multivariate association between two sets of variables are examined. It is shown that all of these measures are generalizations of the (univariate) squared-multiple correlation; all are functions of the canonical correlations, and all are invariant under linear transformations of the original sets of variables. It is further shown that the measures ...
123 CitationsSource
#1Sik-Yum Lee (CUHK: The Chinese University of Hong Kong)H-Index: 53
#2Robert I. Jennrich (UCLA: University of California, Los Angeles)H-Index: 31
Several algorithms for covariance structure analysis are considered in addition to the Fletcher-Powell algorithm. These include the Gauss-Newton, Newton-Raphson, Fisher Scoring, and Fletcher-Reeves algorithms. Two methods of estimation are considered, maximum likelihood and weighted least squares. It is shown that the Gauss-Newton algorithm which in standard form produces weighted least squares estimates can, in iteratively reweighted form, produce maximum likelihood estimates as well. Previousl...
122 CitationsSource
Using the analysis of covariance structures methodology, the construct validity of the tripartite classification of attitudes is investigated. Convergent and discriminant validity were achieved for the reanalysis of Ostrom's (1969) data examining attitudes toward the church, while convergent, but not discriminant, validity was attained for the reanalysis of Kothandapani's (1971) data investigating attitudes toward birth control. Reasons for the difference in findings are examined including an an...
163 CitationsSource
#1Charles E. WertsH-Index: 20
#2Donald A. RockH-Index: 32
Last. Karl G. Jöreskog (Uppsala University)H-Index: 51
view all 4 authors...
The psychometric application of Joreskog's (Note 1) procedure for simultaneous factor analysis in several populations is illustrated. Using Scholastic Aptitude Test data from two samples, procedures are shown for checking test construction assumptions about units of measurement and error variance, within and between samples.
12 CitationsSource
Cited By45611
#1Biswajit Acharjya (Entrepreneurship Development Institute of India)
#2Simanchala Das (K L University)
The outbreak of the novel coronavirus disease (COVID-19) has resulted in the complete disruption of the learning ecosystem across the world. The sudden shift from the class room learning to the use of virtual platforms has not only made an unprecedented impact on the learning style of the students, but has also resulted in the problem of adoption of the same. Thus, with the significant surge in the usage of e-learning mechanism, the researchers even tend to predict the continued usage of the dig...
#1Tuyet-Mai NguyenH-Index: 4
#2Ashish MalikH-Index: 15
Last. Pawan BudhwarH-Index: 66
view all 3 authors...
The COVID-19 pandemic has caused organizational crises leading to shutdowns, mergers, downsizing or restructuring to minimize survival costs. In such organizational crises, employees tend to experience a loss or lack of resources, and they are more likely to engage in knowledge hiding to maintain their resources and competitive advantage. Knowledge hiding has often caused significant adverse consequences, and the research on knowledge hiding is limited. Drawing upon the Conservation of Resources...
#1Megan C. Good (CPP: California State Polytechnic University, Pomona)H-Index: 1
#2Charles H. Schwepker (UCM: University of Central Missouri)H-Index: 19
Abstract null null Previous research has explored the role of political skill as an intra-organizational skill but not in the context of business-to-business seller’s customer relationships and sales performance. In this paper, we: (i) develop a framework to explore the relationships between political skill, customer-oriented selling, customer-relationship-building competence, and customer-directed deviance; and (ii) investigate whether these relationships improve sales performance. The empirica...
Abstract null null Despite its importance, for micro, small, and, medium-sized enterprises (MSMEs), the impact of new service development performance (NSDP) on organizational performance (OP) and the significance of formal or standardized service development processes (SSDPs) remains unknown. This study proposes that NSDP is positively associated with OP and that SSDPs positively moderate this direct effect. These and other hypotheses are tested with data collected from 801 Irish service MSMEs u...
#1Xiao-Ling Jin (SHU: Shanghai University)H-Index: 14
#2Xiaoyue Chen (SHU: Shanghai University)
Last. Zhongyun Zhou (Tongji University)H-Index: 12
view all 3 authors...
Abstract null null Entitled cover images are used widely in social shopping communities to induce content reading, a prerequisite for product purchasing. However, it is still unclear which features of cover images affect users’ product-knowing willingness and content-reading willingness and how. To address this research gap, this paper draws upon the stimulus-organism-response framework and takes a mixed-method approach by conducting 1) a Delphi study to identify two image features (authenticity...
#1A. K. M. Najmul Islam (Lappeenranta University of Technology)H-Index: 18
#2Matti Mäntymäki (UTU: University of Turku)H-Index: 16
Last. Ofir Turel (University of Melbourne)H-Index: 40
view all 4 authors...
Abstract null null This study explores how using social networking sites (SNSs) to cope with stressors induced by a global pandemic (in this case, COVID-19) can have negative consequences. The pandemic has imposed particular stressors on individuals, such as the threats of contracting the virus and of unemployment. Owing to the lockdowns and confinements implemented to limit the spread of the pandemic, SNS use has surged worldwide. Drawing on Lazarus and Folkman’s theory of stress and coping, we...
#1Mahlagha Darvishmotevali (PolyU: Hong Kong Polytechnic University)
#2Levent Altinay (Oxford Brookes University)H-Index: 47
Abstract null null Given the numerous environmental issues facing the hotel industry, there is growing pressure to respond to them by implementing sustainable strategies such as green human resource management (HRM) practices. Thus, there is a need to examine how green HRM enhances environmental performance. Accordingly, this study explores the causal relationship between green HRM, employees’ pro-environmental performance (P-EP), environmental awareness, and servant leadership. The data were co...
#1Minhao Zhang (UoB: University of Bristol)H-Index: 9
#2Yichuan Wang (University of Sheffield)H-Index: 22
Last. Hossein G.T. Olya (University of Sheffield)H-Index: 18
view all 0 authors...
Abstract null null Although social media analytics has been extensively adopted by the hotel industry, how hotels leverage social media analytics capability (SMAC) to create business value remains unclear. Underpinned by real options theory (ROT), this study presents a framework to examine the mechanisms by which organisational agility indirectly influences hotel performance through the mediating role of SMAC. We tested the framework by drawing on a dataset of matched data from hotel practitione...
#1Mingmin Zhang (SCNU: South China Normal University)
#2Ping Xu (URI: University of Rhode Island)H-Index: 8
Last. Yinjiao Ye (URI: University of Rhode Island)H-Index: 7
view all 3 authors...
Abstract null null Chinese social media platforms such as WeChat, TikTok (Douyin in Chinese), and Weibo have become increasingly popular, attracting large amounts of loyal users in and outside of China. Borrowing theories on brand trust and perceived brand values from Marketing and Management, this study examines how perceived media values influence trust in Chinese social media brands such as Weibo and WeChat. Utilizing original survey data collected from Chinese social media users, our study f...