Technology Acceptance Model (Perceived Usefulness and Perceived Ease of Use)
In contrast with the other entries of the Consumer Health Informatics Research Resource website (which cover psychosocial variables that contribute to cognitive, affective, and ultimately behavioral effects), this entry is about technology, or a channel of communication. Specifically the technology acceptance model assesses end-user acceptance of a technology for a health communication purpose. The ultimate success of health communication interventions via web based and other communication technologies is dependent on the use of technology by the target audiences for intended purposes.
Davis's Technology Acceptance Model (TAM) provides a valid and reliable measure that predicts the acceptance or adoption of new technologies by end-users (Davis, 1989; Davis, Bagozzi, & Warshaw, 1989). It also is a commonly used model to measure technology acceptance (King & He, 2006).
Davis's (1989) original TAM predicts acceptance based on the end-user's perceived usefulness (PU) and perceived ease of use (PEOU) of the technology for a specific purpose. Davis applied the model to work settings, and defined PU as "the degree to which a person believes that using a particular system would enhance his or her job performance." (Davis, 1989, p. 320). In non-work settings, the goal becomes personal objectives instead of enhancing job performance. For example, a PU goal or "job" of a health care consumer in selecting a physician (using web-based data) hypothetically is to identify a highly qualified medical practitioner. In this circumstance, web based information is useful to the extent it helps achieve the latter goal. Davis (1989) defined PEOU as: "the degree to which a person believes that using a particular system would be free of effort" (p. 320). Here, PEOU's definition, unlike PU's, transcends work settings as well as goals or purposes.
Davis' (1989) TAM model's validity and reliability were established in two studies. Davis' instrument included six items to measure PU and six items to measure PEOU. The first study compared workers' perceptions of usefulness in addition to the ease of use of implemented technologies from workers' self-reported use of these technologies. Some users of computer-based electronic mail (containing file editor software) rated their PU and PEOU respectively and reported the extent each was used.
Using pooled data for both technologies, the instrument's Chronbach alpha reliabilities were high for both PU (.97) and PEOU (.91) in the first study. Both PU and PEOU were significantly correlated with self-reported use of these systems (r = .63 and .45, respectively; p < .001).
The second study included a different group of workers in a non-work setting. In this study workers were asked to rate the PU and PEOU of computer graphics systems that were not in current use and to prospectively predict future adoption.
The instrument from the second study (Davis, 1989), for one of the two computer graphics systems (called CHART-MASTER), is included below. There are six questions for 'Perceived Usefulness' and six questions for 'Perceived Ease of Use.' Each question for both constructs included a seven-point Likert scale from extremely likely to extremely unlikely.
Scale for Perceived Usefulness:
- Using CHART-MASTER in my job would:
- enable me to accomplish tasks more quickly
- increase my productivity
- enhance my effectiveness on the job
- make it easier to do my job
- I would find CHART-MASTER useful in my job.
Scale for Perceived Ease of Use:
- Learning to operate CHART-MASTER would be easy for me
- I would find it easy to get CHART-MASTER to do what I want it to do
- My interaction with CHART-MASTER would be clear and understandable
- I would find CHART-MASTER to be flexible to interact with
- It would be easy for me to become skillful at using CHART-MASTER
- I would find CHART-MASTER easy to use
Reliability and Validity
Pooling the data, the Chronbach alpha reliabilities of this second instrument were high for both PU (.98) and PEOU (.94). Significant correlations with self-predicted use were found for PU (r = .85, p < .001) and PEOU (r = .59, p < .001). In addition, Davis reported high convergent, discriminant, and factorial validity for both scales in both studies.
The association between PU and use was more pronounced in both studies than the association between PEOU and use. Additional analyses suggested that PU mediated the effect of PEOU on use.
Originally, TAM was applied to adoption of technology in the workplace -- with a focus on computer technology. The focus on work settings has continued and has included perceptions of adopters within health care work places (Holden & Karsh, 2010), such as: physicians (e.g., Paré, Sicotte, & Jacques, 2006), nurses (e.g., Tung, Chang, & Chou, 2008), and other clinicians (e.g., Schaper & Pervan, 2007). Research also has focused on medical professionals acceptance of electronic health records (EHRs) and other health information technology (HIT), such as Blumenthal and Glaser's (2007) findings that to providers EHRs represent an HIT that potentially enable systemic improvement in quality and cost. Similar research assessed the acceptance of HITs, such as personal digital assistants (Yi, Jackson, Park, & Probst, 2006), as well as radiological picture archiving and communication systems (Duyck et al., 2008), and telemedicine technology (Hu, Chau, Liu Sheng, & Kar Yan, 1999).
Holden and Karsh (2010) urge additional study of TAM's application within health care settings, including distinguishing to what extent PU is based on perceived gains in patient outcomes or productivity.
Related research also focuses on consumer health behaviors and their adoption of medical technologies. For example, Wilson and Lankton (2004) assessed consumer acceptance of health information technology to help patients better manage chronic disease. TAM also has been integrated with motivational theory (Davis, Bagozzi, & Warshaw, 1992), resulting in an extended model in which PU is paired with external motivation (EM) in a PU-EM scale (Venkatesh, Speier, & Morris, 2002) -- that is provided below. The PU-EM and PEOU scales (immediately below) are similar in wording to the scales used in measuring non-health related, workplace technology.
Perceived Usefulness - Extrinsic Motivation:
- Using (e-health) will support critical aspects of my heath care.
- Using (e-health) will enhance my effectiveness in managing my health care.
- Overall, (e-health) will be useful in managing my health care.
Perceived Ease of Use:
- My interaction with (e-health) will be clear and understandable.
- (E-health) will be easy to use.
- I will find it easy to get (e-health) to do what I want it to do.
Both PU and PEOU were found to be highly reliable (α = 0.96 and 0.91, respectively) (Wilson & Lankton, 2004).
TAM also has been used to measure consumer acceptance of consumer health informatics technologies, such as m-health, e-health and telemedicine research (Or et al., 2011). A recent review urged the application of the TAM in more consumer health informatics studies (Keselman, Logan, Smith, Leroy, & Zeng-Treitler, 2008).
TAM has been evaluated within broader behavioral theories such as the Theory of Reasoned Action (Legris, Ingham, & Collerette, 2003).
Some extensions of TAM include TAM 2 (Legris et al., 2003) and the Unified Theory of Acceptance and Use of Technology (Or et al., 2011). These extended models include additional explanatory variables. Among the variables added to these extended models is subjective norms in TAM 2 and social influence in the Unified Theory of Acceptance and Use of Technology. The questions used to measure subjective norms and social influence include:
- Pediatricians who are important to me think I should use [the] system (Chismar & Wiley-Patton, 2002).
- Colleagues who are important to me think I should use [the] system (Yi et al., 2006).
Studies of TAM within expanded frameworks have found that the association between PU and behavioral intention is partially mediated by attitude, further suggesting the primary effect of PU on behavior and the secondary or supporting effect of PEOU. Or et al. (2011) found a subjective norm's impact on behavioral intent was mediated by PU.
Rationale for Selection
While the original TAM instrument is recommended for use, its weakness (in comparison to TAM2 -- or the Unified Theory of Acceptance and Use of Technology) is the original does not account for additional explanatory variables such as subjective norms. Nevertheless, the original TAM is parsimonious and includes PU, which has consistently explained more variance in technology acceptance than other variables (Holden & Karsh, 2010). The original TAM also has contributed to explaining the association between PU and the important technology design variable of PEOU.