Literature review of our research

Suggested article:
Chao M. C. & Eric T.G. (2008). Understanding Web-based learning continuance intention: The role of subjective task value. Information & Management, 45,194-201"

Impact Factor of the Journal: 2.119 (for 2006)
Journal Citation Reports® 2007, published by Thomson Scientific

1. What is the problem for the study?

Context
“Web-based learning helps to provide a cost-effective way to widen access to higher education for the society and to reduce the cost of and increase availability of training for organization. Its success depends mainly on learners’ loyalty (continuance).The study is desired to explore individuals intentions to continue using Web-based learning in a voluntary setting.” (p. 194)

There has been existing researches and theories, e.g. Unified Theory of Acceptance and Use of Technology, on accounting for factors affecting user acceptance and usage of IT. These theories are extended to account for the factors affecting individuals’ volition to continue the web-based learning,

The problem
1. Are there any other factors affecting the individuals’ perseverance in web-based learning? 2. And if so, are they in any way stronger than those which are already identified in existing theories?
3. How can we use these factor variance to predict the learners’ continued intention in web-based learning?
4. What will be the practical implication to future enhancement of the design of web-based learning to entice users to stay loyal?

2. What procedures did the experimenter use for the study?

2.1. Develop assumptions and hypotheses for a research model
1. Adapt from literature review - Unified Theory of Acceptance and Use of Technology (UTAUT) as user acceptance to new IT, extending to evaluation on web-based applications
2. Incorporate new items from other literature – expectancy-value model
3. Determine the independent variables:
- Performance Expectancy, Effort Expectancy, Social Influence and Facilitating Conditions (from UTAUT)
- Computer self-efficacy (from UTAUT / Expectancy-value model)
- Attainment value, Utility Value, Intrinsic Value (positive subject task value adapted from expectancy-value model)
- Social Isolation, Anxiety, Delay in Responses, Risk of Arbitrary Learning (cost component of subject task value adapted from expectancy-value model)

2.2 Sample: Part-time students who took one or more Web-based courses offered by a university in Taiwan
Procedures:
l Involving critical advisors: A pretest of questionnaire was conducted using four experts to assess logical consistency, ease of understanding, sequence of items and task relevance.
l Inviting member checking: A pilot study with 20 part-time Master’s degree students who had taken Web-based course was also conducted.
l Actual survey: Sent three thousand e-mails providing a hyperlink to the Web survey to part-time students who had registered for at least one Web-based course

2.3 Instrument
The questionnaire developed by the authors based on the set hypothesis developed from merging the factors in existing theories and adaptation of subject task values from other theories

2.4 Validity and Reliability testing
- Adequacy of measurement model (Confirmatory factor analysis)
- Criteria model fit (Chi-square value, AGFI, NNFI, CFI)
- Reliability (composite reliability values)
- Convergent validity (AVE)
- Discriminant validity (square root of AVE)

3. What were the major conclusions for the study?

Conclusions
3.1 That a ‘new’ research model to predict user behaviour on the intention to continue web-based learning is established based on the authors’ newly incorporated factors as hypothses (most of the hypotheses are valid). However, some of the hypotheses proposed are not supported by the research.

3.2 “Performance expectancy, effort expectancy, computer self efficacy, attainment value, utility value and intrinsic value were significant predictors of individuals’ intentions to continue using Web-based learning while anxiety had a significant negative effect.”

3.3 Social influence and facilitating conditions which used to be identified as positive factors for continuance of web-based learning, and are so hypothesized by the authors, are proved to be not so significant in this survey, while social isolation, delay in responses, risk of arbitrary learning, identified in the authors’ hypothesis as negatively related to the web-based learning, are not detrimental as imagined.

3.4 Intrinsic value exhibited as the strongest factor motivating an individual’s intention to continue on web-based learning.

Implications
- Future design of web-based learning should put more focus on enhancing the features that will promote users’ intrinsic values (playfulness) in the web-based learning process to sustain their interest in learning.

- Continued efforts should be spent on strengthening the design of web-based learning systems in the light of the other factors identified as positive predictors in the research.

4. How would you classify the study, according to the six types of research studies we looked at in this lesson?

Dependent variable – continued intention in web-based learning
Independent variable (not under experimenters’ control) – the 14 attributes listed in para 2.1.3
è Causal-comparative research

Limitation / Additional thoughts
From the author
This survey only targeted on a single web-based system for part-time adult learners -> not representative enough to other types of learners.
The return rate of the survey (286 out of 3,000, representing around 10% of the population in the system, presumably fairly active) -> contextual interpretation applies.
The survey only measures the subjects’ intention at a moment in time and does not represent the ongoing change of intention.
The research is applicable to represent a voluntary setting in a CE course (vs a mandatory setting, e.g. an organization)

Additional thoughts
The demographic information of the non-respondents can also be tracked as a reference to compare with the ‘active population’ in the web-based system. Such information can become other variables in extension.

Also in relation, the reason behind the non-response is an unknown factor – the non-respondent may have been inactive in the web-based system either because he is really disinterested or he may be a high-achiever in web-based learning who switched to other web-based learning programmes due to the unsatisfactory performance of the web-based learning in question.

Other than the application in improving the web-based learning system, the model may be applied to assist the administrative decision of programme admission or promotion if the background of the learners with low intention to continue can be tracked – that may form another research interest.