Developing a usable website is pivotal for e-business success. Researchers have devoted effort to develop metrics, guidelines and theories of website usability, yet there still is a lack of consensus on the multifaceted dimensions of website usability and lack of investigation of the nomological networks among website usability constructs. This study first investigated the common dimensions of website usability by integrating the findings of previous studies and a focus group study with website usability experts. Instruments to measure the constructs were developed and empirically validated. Then nomological networks between website usability constructs and between those constructs and online purchase intention and purchase were examined. Three field studies including two questionnaire surveys and a causal mapping analysis were conducted. The research identified ten website usability constructs with strong psychometric properties. A number of nomological networks between usability constructs were discovered, contributing to identification of sources of variances of purchase intention and purchase behavior. Findings of this study are expected to provide useful insights for practitioners to develop more usable websites and for researchers to better assess the effect of website usability on online customer behavior.
Within the e-commerce environment, the impact of a web site design on customer purchases cannot be fully understood without an evaluation of the usability of the web site
– DeLone and McLean [19, p. 25]
Developing usable websites is pivotal for e-business success since online consumers touch, feel, search, and experience products (or services) mainly through websites [18] and ⁎[68]. A usable website has been known to provide a positive attitude toward online stores, increases stickiness and revisit rates, and eventually stimulates online purchase [13], [32], ⁎[50] and [70]. To make comparisons with competitors' websites, benchmark performance, and plan improvements for designing a more usable website, HCI researchers have devoted effort to define website usability constructs, develop instruments to measure the multifaceted website usability constructs, assess their influences on online customers' attitude or behaviors including satisfaction, loyalty, purchase or revisit intention, and actual purchases [23], [30] and ⁎[46]. This has led to a number of metrics, guidelines, and even theories of website usability.
However, even with this knowledge and contrary to our expectation, current websites still contain a number of usability problems ⁎[9], [14], [67] and ⁎[68]. Difficult-to-understand content, inconsistent formats, difficulty in navigation, disorientation, lack of interaction and reliability, inefficient search capabilities, and ill-defined help functions are usability problems frequently identified with commercial websites. Regarding the poor usability, Becker and Mottay [9] mentioned that “online business failures are increasing as consumers are turning away from unusable sites. The ‘build it and they will come’ attitude has led to the demise of several e-commerce sites which are too slow, too buggy, or too complex for ease of use” (p. 54).
There are numerous reasons for poor website usability, but this study focuses on the lack of consensus on the multifaceted dimensions of website usability and lack of investigation of the nomological networks among website usability constructs. Precise articulation of all dimensions of website usability facilitating particular design assumptions enables the design process to become more tractable for developers by focusing their attention and restricting their options, thereby improving design outcomes. In addition, designing standardized instruments to measure those constructs can facilitate direct comparisons across time, culture and geographical regions [30]. However, there is a lack of consensus on the constructs of website usability and how to measure them. Thus, previous studies were conducted based on different sets of constructs measuring website usability. Using different instruments to measure the same usability construct is another problem with previous studies, making direct comparisons of research findings difficult. Researchers have lamented on the absence of general measures and instruments of web usability and called for research regarding this matter ⁎[46] and ⁎[68].
HCI researchers and practitioners have speculated about potential causal relationships among multi-dimensional usability constructs [e.g., ⁎[50] and [52]] and found high correlations between some of those constructs [e.g., 46]. Previous usability studies assumed independent relationships among usability constructs and thus only examined their direct influence on online customers' perceptions and behaviors. Scientific investigation is needed to reveal the relationships among those constructs to precisely assess overall influence on website usability, but few attempts have been made to examine the relationships.
Given this background, this study has two primary goals: (1) to identify common dimensions of website usability by integrating the findings of previous studies and through a focus group study with website usability experts. The instruments to measure the constructs are developed and empirically validated. (2) to investigate the nomological networks among website usability constructs by use of a causal mapping approach and structural modeling analysis. Through empirically testing and validating the networks, the effect of each website usability construct on online customers' purchase intention (and purchases) can be more precisely captured.
A critical understanding of perceptions and purchase behavior of consumers cannot be done without a suitable appreciation of website usability. Arguing that the usability of a website is a fundamental component of the entire user experience, this study develops website usability constructs and measurement instruments and investigates their nomological networks useful to both researchers and practitioners. Findings of this study are expected to be used for successfully measuring the design quality of websites and be used as building blocks for developing a strong theoretical model of website usability.
