Book•
Structural Equations with Latent Variables
28 Apr 1989-
TL;DR: The General Model, Part I: Latent Variable and Measurement Models Combined, Part II: Extensions, Part III: Extensions and Part IV: Confirmatory Factor Analysis as discussed by the authors.
Abstract: Model Notation, Covariances, and Path Analysis. Causality and Causal Models. Structural Equation Models with Observed Variables. The Consequences of Measurement Error. Measurement Models: The Relation Between Latent and Observed Variables. Confirmatory Factor Analysis. The General Model, Part I: Latent Variable and Measurement Models Combined. The General Model, Part II: Extensions. Appendices. Distribution Theory. References. Index.
Citations
More filters
21 citations
TL;DR: In this article, the authors discuss classical test theory and the theory of coefficient $$\alpha $$>>\s>>\s, and argue that lower bounds are useful when assessing product quality features, such as a test-score's reliability.
Abstract: Prior to discussing and challenging two criticisms on coefficient $$\alpha $$
, the well-known lower bound to test-score reliability, we discuss classical test theory and the theory of coefficient $$\alpha $$
. The first criticism expressed in the psychometrics literature is that coefficient $$\alpha $$
is only useful when the model of essential $$\tau $$
-equivalence is consistent with the item-score data. Because this model is highly restrictive, coefficient $$\alpha $$
is smaller than test-score reliability and one should not use it. We argue that lower bounds are useful when they assess product quality features, such as a test-score’s reliability. The second criticism expressed is that coefficient $$\alpha $$
incorrectly ignores correlated errors. If correlated errors would enter the computation of coefficient $$\alpha $$
, theoretical values of coefficient $$\alpha $$
could be greater than the test-score reliability. Because quality measures that are systematically too high are undesirable, critics dismiss coefficient $$\alpha $$
. We argue that introducing correlated errors is inconsistent with the derivation of the lower bound theorem and that the properties of coefficient $$\alpha $$
remain intact when data contain correlated errors.
21 citations
Dissertation•
01 Jan 2004
TL;DR: In this article, the authors investigate factors that encourage and discourage purchase intention of consumers when buying health foods online in Thailand, including perceived usefulness, perceived risk, customer experience, and product and company attributes.
Abstract: The advent of the Internet, accompanied by the growth of related technologies, has
created a significant impact on the lives of people around the globe. For marketers,
one of the most significant impacts has been the emergence of virtual stores that sell
products and services online. Consumers can now purchase goods and services
virtually anywhere, 24 hours a day, 7 days a week, without geographical and
temporal boundaries.
While many marketers acknowledge the importance of using the Internet in their
marketing mixes, little research has empirically tested the critical factors that
influence an individual's decision when buying products or services online. Based on
the gaps found in the literature, this study was designed to investigate factors that
encourage and discourage purchase intention of consumers when buying health
foods online in Thailand. The study also examined the relative importance of such
factors. Thus, the research problem investigated in this thesis is:
What are the important factors influencing consumer's online purchase
intention of health foods in Thailand?
The specific objectives of this research were not only to identify and explore the
relative importance of factors, but also to develop a model to investigate the factors
influencing purchase intention of consumers when buying health foods online in
Thailand. This research was designed in three stages covering both exploratory and
explanatory research. The exploratory stage covering stage one and two, started by
reviewing the existing literature relating to behaviors and attitudes when consumers
buy products online. The Technology Acceptance Model (TAM) developed by Davis
(1989) was selected as a theoretical framework to build a conceptual model for this
study. In addition to the two original constructs in the TAM model, namely,
perceived usefulness (POU) and perceived ease of use (EOU), the literature review
suggested three additional constructs. These were perceived risk (PR), customer
experience (CE), and product and company attributes (PCA). Four focus groups
were conducted in stage two to gain consumer insights in order to understand, refine
and develop the final model and hypotheses to be confirmed and tested in the
explanatory research. Finally, the modified TAM model and eleven hypotheses were proposed to explain and test the behavioral intention of health food consumers when
buying health foods online in Thailand.
In the explanatory stage, which forms the major portion of this research, an online
survey was conducted with responses from 786 consumers taken from the Cerebos
customer database. All respondents had used both health foods and the Internet
during the past 12 months. Exploratory factor analysis (EFA) was used to explore
and test the suitability of data collected from the survey. Structural equation
modeling (SEM) was chosen to confirm the measurement model in this study
because it offered a mechanism to validate the relationships between constructs and
indicators by using confirmatory factor analysis (CFA) and tested the relationships
among constructs by using path analysis. All five constructs in the model exhibited
high levels of reliability, validity and produced the final measurement and structural
models. Nine out of eleven hypotheses were accepted and two were rejected. In
addition, six propositions were also found in this study.
Similar to prior research, the results in this study indicated that perceived usefulness
(POU) was a powerful determinant and the strongest predictor of behavioral
intention. It was the only construct that showed a significant positive and direct
effect on purchase intention with no indirect effect involvement. Customer
experience (CE) was the second most important factor in this study. The customer
experience itself, did not have any direct effect on purchase intention but
demonstrated a significant positive and indirect effect on purchase intention.
Perceived risk (PR) was the third most important factor in this study. Similar to
customer experience, perceived risk did not have any direct effect but it
demonstrated a significant negative and indirect effect on purchase intention.
Perceived ease of use (EOU) and product and company attributes (PCA) were found
to have little effect on behavioral intention in this study. Similarly to previous
studies, the two original constructs in the TAM model, perceived usefulness (POU)
and perceived ease of use (EOU), were found to be mediating factors of other
constructs in influencing purchase intention (PI), in this study.
In summary, forty effective measurement items were identified and confirmed to be
associated with fourteen factors under these five constructs in the structural model.
Variety of choices was the most effective item of measurement for perceived usefulness (POU), while modern personality, product assurance, trusted company
and simple order procedure were found to be the most effective measurement items
for customer experience (CE), perceived risk (PR), product and company attributes
(PCA), and perceived ease of use (EOU), respectively. All of these factors
demonstrated statistically significant high factor loadings on the relevant constructs.
The findings from this research provide useful information for corporate
management, and marketers in prioritizing and allocating their resources in terms of
manpower, investment, marketing effort, and time to improve the impact of these
constructs, all of which will ultimately enhance the possibility of consumers buying
health foods online. Results from this study are beneficial to Web developers in
designing attractive and effective Web sites or homepages to draw consumer's
attention when buying products online. Cost of using the Internet should be reduced
to make it more competitive and affordable to wider population. In addition, these
findings are also useful to the Thai government in designing and drafting an Internet
policy to enhance the scope and development of e-commerce and online business in
Thailand.
This dissertation concluded by identifying opportunities for future research. These
were addressing the delimitations of scope, further testing and validation of the
measurement scales, measurement of actual buying behavior, adding demographic
and psychographics variables into the model, conducting longitudinal observation,
and last but not least, the application of the modified TAM model to other consumer
products in the Thai context.
21 citations
TL;DR: Given that compliance measures are rarely, if ever, error-free indicators of exposure it is argued that both the designs for the collection of compliance data and the statistical models for their resulting analysis should be changed to take the possibility of measurement error into account.
Abstract: This paper explores the implications of measurement error in the analysis of compliance-response relationships in data from randomized trials. Given that compliance measures are rarely, if ever, error-free indicators of exposure it is argued that both the designs for the collection of compliance data and the statistical models for their resulting analysis should be changed to take the possibility of measurement error into account. An analysis which ignores measurement error in the compliance measurements will provide biased estimates of compliance-response relationships. Provided that one has two or more indicators of compliance for each subject, more appropriate models can be fitted using covariance structure modelling software. If one wishes to explore interactions from repeated measures data on both compliance and response then it is also important that one recognizes that the response measures are also error-prone and that they too are dealt with appropriately.
21 citations