Content area
Results from a field study of individuals with disabilities who used voice recognition technology are discussed. Twenty-three individuals who were successful in the use of VRT and 17 who were unsuccessful were interviewed. Qualitative results indicated that task-technology fit, training, the environment, and the disability limitations were the differentiating items. The ability to use the VRT for a trial period may be the major factor resulting in successful adoption of this technology.
Keywords
Adaptability, Technology, Disabled people, Voice recognition
Abstract
In this paper, results are presented from a field study of individuals with disabilities who used voice recognition technology (VRT). Twenty-three individuals who were successful in the use of VRT and 17 who were unsuccessful were interviewed by the researcher. Qualitative results indicated that task-technology fit, training, the environment, and the disability limitations were the differentiating items. The ability to use the VRT for a trial period may be the major factor resulting in successful adoption of the technology.
Introduction
Organizations are faced with the problem of providing suitable support when they employ a person with a disability or when a current employee becomes disabled. An individual with a disability is defined as "a person who has a physical or mental impairment that substantially limits one or more major life activities, a record of such an impairment, or is regarded as having such an impairment" (Equal Employment Opportunity Commission - the US Department of Justice, 1991, p. 1). In many cases, computer technology is being adapted for use by individuals with disabilities in order to provide a means to be employable. However, virtually no information exists explaining why the same adaptive technology will be successful for one person and unsuccessful for another.
For the purposes of this study, adaptive technology is any computer software or hardware tool that assists individuals with disabilities to be more effective in performing job related or daily activities (Brown, 1992). For example, a person with blindness may use a screen reader in order to hear rather than see what is displayed on the computer monitor or may use a printer that prints in Braille.
Voice recognition technology (VRT) was the type of adaptive technology studied in this research. The purpose of VRT is to allow the input of data without the use of a keyboard, thus enabling the individual to speak into the microphone instead of typing in commands. Voice recognition technology was developed for use by fighter pilots and has been targeted for hands-busy applications. Few companies have marketed their product directly to the disabled consumer; however, voice recognition technology is being used as an adaptive technology by individuals with disabilities. For example, VRT may be used by individuals with motoric disabilities, such as people with quadriplegia (total paralysis of the body from the neck down), by allowing the computer to be accessed without the use of one's hands.
Most individuals with disabilities praise adaptive technology and say that they could not imagine life without it (Lazzaro, 1990; Sheldon, 1990). However, a national survey on abandonment (Phillips and Broadnax, 1992) concluded that almost one-third of assistive devices are abandoned. As not all attempts to use adaptive technology conclude favorably (Bowe and Little, 1984; Brown, 1989), individuals and organizations search for guidelines to assist in the adoption of adaptive technology. Individual case studies of successes and abandonments of adaptive technology at the High Tech Center for Individuals with Disabilities have resulted in guidelines for the selection and use of adaptive computer technology (Brown, 1989). However, empirical evidence documenting the differences between successful and unsuccessful outcomes is lacking.
This investigation answers some fundamental questions: What factors contribute to successful adoption and use of adaptive technology by individuals with disabilities? Which factors are likely to be associated with failure? How can employers increase the chances that their employees with disabilities will be successful users of adaptive technology? Do the models used for the adoption of information technology (IT) innovations by individuals within an organization apply to the adoption of adaptive technology by individuals with a disability working in an organization?
Research methodology
To provide insights into the research question what are the factors that contribute to success in the adoption and use of an adaptive technology innovation by an individual with a disability working in an organization? - a field study was conducted (Goette, 1995, 1998). The field study consisted of a survey questionnaire and standardized open-ended interview questions. All of the questions were asked by the researcher in a personal interview with each respondent.
Individuals with physical or sensory disabilities using or having used voice recognition technology were selected to participate in this study. Information about individuals fitting these descriptions was obtained from organizations and vendors involved with individuals with disabilities. The researcher cold-called the organizations. Individuals from the organizations contacted people that they knew fitting the requested parameters to receive permission to pass the name and address information on to the researcher.
A total of 23 people with successful adoptions and 17 people with unsuccessful adoptions were interviewed for this study. This number was substantial enough to conduct statistical analysis but small enough for the researcher to locate and visit the required number of individuals. The researcher contacted each participant to arrange a time for the interview. The researcher traveled to the home or office of each participant over a period of seven months and drove more than 12,000 miles to conduct the interviews.
It is expected that the same models used for the adoption of an IT innovation by an individual within an organization may be applied to individuals with disabilities within organizations as well. Expectancy theory (Vroom, 1964), innovation-diffusion theory (Rogers, 1983), and task-technology fit theory (Goodhue, 1992) form the basis for this research. The independent variables are categorized as expectations, innovation characteristics, and task-technology fit while the dependent variable is success.
Theoretical foundations
When the need arises to adopt an innovation, the individual develops expectations about using the innovation and the results of using the innovation. Expectancy theory suggests that one's predisposition to perform a specific task is a function of the perceived probabilities and consequences of success and failure of task performance (Bradt, 1991; Vroom, 1964). This means an individual with a disability considers his or her probable success or failure with VRT and develops expectancies about the VRT before using the innovation. Since these expectancies may influence the actual success with the VRT innovation, it is necessary for researchers to understand expectancies in order to explain the influence of expectations on the outcomes of adoption decisions.
Much of innovation-diffusion theory was developed for potential adopters making voluntary choices to adopt or not adopt an innovation based on the benefits they expect from the personal use of the innovation (Fichman, 1992). The research conducted in this study fits the model of innovation-diffusion theory because voice recognition technology is an adaptive technology innovation that may be used by individuals with disabilities to improve their quality of life.
A classification scheme developed by Rogers (1983) can be used to describe five different attributes of innovations. This general classification scheme was developed to allow comparison of different innovations. Rogers' categories of innovation attributes are as conceptually distinct and as general as possible. The innovation characteristic categories are: relative advantage, compatibility, complexity, trialability, and observability.
In the field of rehabilitation engineering, studies have also examined the characteristics of the adopted device. In studies of consumer perceptions, the four most important characteristics have been found to be relative advantage, cost, complexity, and dependability (Phillips and Broadnax, 1992; Ward, 1990). Dependability is the reliability of the adopted device. Ward (1990), in the second year of a five-year national study on technology abandonment, found cost to be a major obstacle in obtaining assistive devices. Phillips and Broadnax (1992), in a national survey on technology abandonment, found poor performance (i.e. relative advantage and dependability) to be a significant reason for the abandonment of assistive devices.
When information technology or adaptive technology may be used for a variety of tasks, the characteristics of the task it is used for must be considered. Based on models presented by Cooper and Zmud (1990) and Goodhue (1992) the task and information technology characteristics should be compatible to have a positive effect on adoption and implementation. Because VRT is an adaptive technology tool that can be used for different purposes, the task must be considered when investigating adoption outcomes.
Innovation-diffusion theory, task-technology fit theory, and expectancy theory form the basis of the research used in this study to investigate how VRT can be used successfully and why VRT is sometimes abandoned.
Independent variables
Based on prior research by Vroom (1964), Ginzberg (1981), and Scherer (1988), the independent variable of expectations consists of four constructs. The first three of these are the expected benefits from using VRT, the expected success of using VRT, and the expected complexity of using VRT. The fourth construct is the expectations that the adopter has about the expectations of others concerning the success of the adopter in using the voice recognition technology.
The perceived characteristics of cost, relative advantage, compatibility, complexity, result demonstrability, visibility, and trialability are seven characteristics of the innovation studied (Rogers, 1983; Tornatzky and Klein, 1982; Moore and Benbasat, 1991). Voluntariness was also included to verify that the use of VRT was voluntary for the individuals studied because the level of use is an acceptable dimension of success only when use is voluntary (Lucas, 1978; Welke and Konsynski, 1980).
Task-technology fit consists of the tasks of computer access, environment control, word processing, and using VRT for all computer access. Environment control is using the VRT to use the phone or control the lights, TV, etc. without needing assistance. Through discussions with people involved with individuals with disabilities using VRT and the prior experience of the researcher with VRT, the conclusion was reached that the two most common functions of the VRT for individuals with disabilities are environment control and word processing. Therefore, it is proposed by the model that these two tasks be critical to VRT users with disabilities.
Furthermore, to encompass the myriad other possible tasks for which the VRT could be used, it should be required that individuals with a disability use a computer to perform their work. There should be a suitable level of task-- technology fit when VRT is used for environment control, word processing, and all computer access.
Dependent variable
In order to evaluate success adequately, multiple constructs should be utilized to measure success. The constructs of user satisfaction, system quality, and use are commonly studied dimensions of success (DeLone and McLean, 1992; Conrath and Sharma, 1993; Ives and Olson, 1984). For the purpose of this study, use and user satisfaction as perceived by the user are utilized to form the dimensions of success. Use comprises the level of use and achievements through use. Achievements through use is the percentage of tasks that are being accomplished through the application of VRT while the level of use is how often the VRT is employed. Use is an important dimension because the use of adaptive technology by an individual with a disability should be voluntary in nature. Using actual use as a measure of success only makes sense for voluntary or discretionary users (Lucas, 1978; Welke and Konsynski, 1980). If a user is forced to use the system when it does not work successfully then actual use will be high even though the system is not a success. The number of hours per week that the VRT is utilized as reported by the individual was recorded as level of use. Although individual reports of time may be inaccurate, no objective measure of recording the hours of VRT use per week was available to the researcher.
Achievements through use is concerned with what is being accomplished through the application of the information technology. An individual who is not using all the functions that an innovation has to offer may rank low in achievements through use. In other words, if the individual does not use the innovation, but instead utilizes other methods to accomplish certain functions, then achievements through use will have a lower ranking. For example, microwaves may be used to defrost, reheat, or cook food. Someone who only uses a microwave to reheat, and employs alternative methods to defrost and cook, will have a lower achievements through use ranking than someone who utilizes the microwave for all three tasks. Achievements through use is the percentage of tasks accomplished of possible tasks performed by the VRT as reported by the user.
User satisfaction is utilized in this study because, if the individual is not satisfied with the VRT, it is less likely the VRT will be successful. Also, user satisfaction is a widely employed surrogate for success (Conrath and Sharma, 1993; DeLone and McLean, 1992; Ives and Olson, 1984). Because questions concerning system quality are represented by a dimension of user satisfaction, system quality will not form a separate measure in this research.
Success maybe seen as a continuum. A success rating was calculated from the dimensions employed to measure success (Bailey and Pearson, 1983; Emory and Cooper, 1991). The scores for the user satisfaction constructs, level of use, and achievements through use were equally weighted to arrive at a score for the successful or unsuccessful use of VRT. A cluster analysis was performed to verify that the successful/ unsuccessful grouping matched the self-reported successful or unsuccessful use as indicated by the individual interviewed.
Data gathering
During the interview the researcher read each question to the participant and recorded the verbal response. This was necessary because the variety of disabilities precluded a common format for the questionnaire that could be completed independently by the participants. In the case of successes, the researcher observed the individual using the VRT. The interviews took from one to two hours of the individual's time. The time difference was due to the length of the demonstration and the amount of extra information volunteered by the participants. All the individuals interviewed were extremely generous with their time and very open with their experiences regarding VRT.
Although interviewer error is a major source of response bias (Emory and Cooper, 1991), the response bias was kept to a minimum because the researcher personally conducted all the interviews and the interview was structured with the open-ended questions last.
Result
As discussed earlier, 23 individuals with disabilities who were successful in their use of voice recognition technology were interviewed. The data gathered were compared to information collected from 17 individuals with disabilities who were unsuccessful in their use of VRT. Two individuals considered themselves successful, but they abandoned the VRT for reasons beyond their control. Therefore, their data are not included with the remaining 15 unsuccessful users of VRT.
Demographic information
The average length of time the individuals in the study have had debilitating disabilities is 20.33 years. The majority of the disabilities are due to intervertebral disk disorders (spinal problems causing paralysis), multiple sclerosis, muscular dystrophy, cerebral palsy, arthritis, and/or tendinitis (inflammation of a tendon). In the selected individuals, all of these conditions had led to paralysis or muscle control problems rendering them unable to type on a keyboard in a conventional manner. Thirty of the subjects have quadriplegia (total paralysis), and two have blindness.
The ages of the subjects ranged from 19 to 64 with an average age of 39.4. Seventy-three percent of the individuals studied were male, and over 88 percent of the individuals had attended college. Fifty-eight percent of the individuals in the study were employed. However, there were no significant differences in the percentage of successful adoptions for the individuals who were employed versus the percentage of successful adoptions for the individuals who were not employed.
According to LaPlante et al. (1992), 9.4 percent of men (18-64) have a disability limiting work activities compared to 8.5 percent of females. Thus, both the sample and the actual population of working age individuals with disabilities reflect a male majority. The sample used in this study is well educated and does not necessarily reflect the educational characteristics of the population of individuals with disabilities.
Computer experience
Thirty-one percent of the individuals studied did not access a computer on a consistent basis before adopting VRT but, instead, relied on others to do the work for them by dictating information for input into the computer. The individuals who did access a computer used various adaptations prior to using voice recognition technology.
Twenty-three percent of the individuals studied had no computer experience before adopting voice recognition. However, the lack of computer experience did not necessarily increase the amount of training received. Training on the VRT system ranged from none (26 per cent) to 70 hours (3 per cent).
Individuals who were unsuccessful in using VRT attempted to use the system for one month to three years with the majority of the unsuccessful users (53 percent) discarding VRT within three months. This is consistent with a national survey on the abandonment of assistive devices which showed that most abandonment occurs after five years (largely due to obsolescence) or within the first three months of use (Phillips and Broadnax, 1992). Individuals successfully using VRT had been using the system for five months to nine years. Thus, these people had enough time to evaluate the VRT system and determine if they could use VRT successfully.
System type
Two basic types of VRT systems were used by the individuals in the study. The first type is an older technology which requires that an entire vocabulary be repeated by the user several times in order to train the software. The vocabulary or a subset of this vocabulary may be made active at a particular time. This type of system is very accurate (due to the limited number of valid words at any one time); however, this system is not designed for use with word processing software.
The second type of system has a large built-in vocabulary. Command words and selected words are said by the user to train the software, and the system adapts to the user's voice over time. The accuracy rate is not as high because of the number of valid words available, but the system is designed for use with word processing software.
The VRT systems used by all the individuals in the study were speaker dependent systems. This means that the user had to train the VRT system to recognize his or her voice. To do this, the user said system selected words two or three times each. The sounds from these words were broken apart into digitized patterns and stored in the system. After completing this initial system training, whenever the user speaks into the microphone, the analog sound waves are digitized and matched to the stored patterns. The resulting word appears on the computer screen as input to the computer as if the user had typed the word.
Qualitative findings
Expectancies
Individuals who were successful in the use of VRT had realistic expectations about the adoption process. This means that the expectations they had prior to adopting the VRT matched their actual adoption experience. They expected to have to devote more time and effort than the vendor stated was necessary to become proficient with the voice recognition technology. Employers who understood the VRT and gave their employees the necessary support aided in the successful adoption process. Some individuals stated that their employers had unrealistic expectations about how quickly and effectively they could succeed with VRT and that the employers were disappointed with the results of the adoption. Typically, these users could not meet their employers' expectations and abandoned the VRT. Most of the individuals returned to whatever method or device they had been using previously to accomplish their work; however, a couple of individuals no longer considered themselves to be employable.
Innovation characteristics
Rogers' (1983) classification scheme was not as useful as expected. Statistical analysis showed that relative advantage, compatibility, and complexity loaded on the same factor making these characteristics a single attribute (benefits). Individuals did not consider cost a factor because vocational rehabilitation organizations or employers purchased the majority of the VRT systems. Also, too few adopters had had an opportunity to observe or try the VRT before adopting to include these two innovation characteristics as comparison factors. On the other hand, the reasons for abandonment discussed later in this paper indicate that trialability may be the most important innovation attribute related to unsuccessful adoptions of VRT.
The question to be answered is why were the unsuccessful users unsuccessful? To organize the answers from each individual into groups a cross-case analysis was conducted (Patton, 1990). The four patterns that emerged from the cross-case analysis of the interviews with the individuals who were unsuccessful in their use of VRT are discussed in this section.
Task-technology fit
According to research by Goodhue (1992), it is important that the technology fits the task it will be used to accomplish. In the majority of the situations, whether the VRT was used in the organization and/or the home, clerical and environment control were the main tasks. Fiftythree percent of the individuals had difficulty in this area. Some tasks can only be easily accomplished by certain types of VRT systems. For example, if the individual used a VRT system that was specifically designed for use with computer aided design software then the VRT could be used successfully for that purpose. However, if the individual used a large vocabulary system (designed for use with word processing software) with computer aided design software, then the individual grew frustrated with the VRT system. Trying to use a VRT system for a purpose other than what it was intended for is like trying to use Windows without a mouse - it can be done, but most users believe it is not worth the frustration. While this is generally true, sometimes the individuals and organizations involved did not realize that a VRT system existed that was specifically designed for their task, so they were trying to make do with the VRT system with which they were familiar.
Some individuals had VRT systems that had a limited vocabulary. With this type of system, a vocabulary must be set up in advance and made active in order to be used to accomplish a specific task. For example, one vocabulary is set up for use with a database program while another vocabulary is set up for a graphics program. These limited vocabulary tasks are well suited for this type of VRT system. However, if the individual must do a lot of word processing, it soon becomes tiring and frustrating because the individual must use the international alphabet to spell almost every word. An individual who was employed as a clerical worker had to maintain records consisting of names and addresses. The majority of these names and addresses had to be spelled for the VRT system. The employee reverted back to the typing stick (a pencil-like device held in the mouth) that was used prior to the VRT because constantly spelling was too slow and frustrating.
Training
A second area that was cited by 40 percent of individuals as a reason for unsuccessful use of VRT was a lack of training. The best training seemed to occur when the individual was trained either by a consultant knowledgeable in training or by a person that concentrated solely on technology training. In the latter case, the organization that purchased the equipment employed the person conducting the training.
Twenty-six percent of the subjects received no training while 22 percent received more than 20 hours of training. Two of the individuals with no training claimed that because of their previous computer experience they did not need any training. Prior computer experience, prior knowledge of the software programs that the VRT will be used to access, and an understanding of how the VRT system works all influence the amount of training needed by an individual.
Individuals who are not computer literate usually need more training than individuals who have substantial computer experience. Organizations obtaining VRT systems for individuals with disabilities mentioned that the training time advertised by the system vendor could be a problem. When the vendor suggested that an individual only required eight hours of training to successfully use the system, the agency purchasing the system may not have been allowed to pay for any more than eight hours of training for the individual. If vendors modify their estimated training requirements to account for the difference in prior computer knowledge of their customers it may allow agencies more flexibility in funding training.
The workplace was the site of a second training problem. Although they may have had access to sufficient training, at least three of the individuals were not allowed time to get adjusted to using the VRT system at work. They did not have adequate time to become competent in using the system because there was no reduction in work load or an extension of deadlines to allow for the extra time needed to become proficient with the system. The individuals reverted to performing the tasks using the previous method, such as a typing stick or dictation, in order to complete the tasks on time rather than taking the additional time needed to complete them using the VRT. This is often the case when any new information technology is introduced and not enough time is given to allow users to adjust to the new information technology (Rahim et al., 1997).
Although, ultimately, the VRT would be faster for the individual to use, initially, it may take longer to get the task accomplished. Several subjects remarked that if they were allowed to take the system home for a period of time (perhaps a month), they would have had a chance to use the system long enough to become proficient. After that, they would have been able to bring the VRT back to work and use the system efficiently. However, due to current job procedures, these individuals considered VRT to be unsuccessful for them.
The environment
Another problem encountered by individuals with disabilities who were unsuccessful in the use of VRT was related to environmental factors at work or home. Environmental problems were cited by 20 per cent of the subjects who were unsuccessful as a reason why VRT did not work for them. Background noise could be a major problem in some work environments. This noise is picked up by the VRT system as a word to be input resulting in the individual having to constantly correct errors made because of the background noise.
Desk microphones and headsets are the two types of microphones used for input. Even though the individual may need help to put on a headset, this type of microphone picks up less background noise. Because the individual is the same distance away from the headset microphone each time the VRT system is used, headsets may allow for more accurate input. One individual who is a journalist used the headset microphone to enter his stories into the computer. His desk is located in the quietest area of the newsroom, and his fellow employees are relatively quiet whenever he is working so that his VRT system does not pick up background noise.
The disability limitations
One-third of the individuals who were unsuccessful in the use of VRT considered the limitations of their disability to be a problem. VRT is not as fast as entering data from a keyboard. Any individual who can type with one finger, one hand, or a typing stick faster than 25 words a minute may not be satisfied with VRT. A technology specialist should have the person try other typing alternatives before purchasing a VRT system. If the individual with the disability insists that VRT will do the job, a trial period could be used to discover if the system is unacceptably slow.
To work properly, VRT must have consistent input. VRT systems can understand someone with a serious speech impediment as long as the speech impediment is consistent. If the speech impediment is inconsistent then the VRT cannot be accurate. Several subjects with cerebral palsy exhibited this type of problem.
Conclusions and implications
A summary of the guidelines used to apply the results to practice are given below. These guidelines answer the fundamental questions: What factors contribute to successful adoption and use of adaptive technology by individuals with disabilities? Which factors are likely to be associated with failure? How can employers increase the chances that their employees with disabilities will be successful users of adaptive technology? These guidelines should be employed when individuals with disabilities are adopting VRT. The use of these guidelines in other situations is limited.
The first guideline is that the expectations of the adopter should be controlled to maximize the chances for success. Employers should manage an individual's expectations by allowing him or her to view the VRT and understand its potential benefits as well as its limitations before an adoption decision is made. Individuals who were successful expected benefits from using VRT, expected to be successful in using the VRT, and expected to have to work hard to understand and use the VRT. The importance of expectations has been shown in other IT adoption studies such as the adoption of Internet-based stock trading (Loh and Ong, 1998).
The second guideline is to select the correct VRT system to fit the tasks it will be used to accomplish. Actual benefits from the use of VRT must be obtained by the person with a disability in order to be a successful user of VRT. To achieve the potential benefits, the VRT should be suited to perform the tasks it will be used to accomplish. If the task to be accomplished is word processing then the VRT system should be designed for use with word processing software. A common mistake was made when an employer had heard of another similar situation in which VRT was successfully used and automatically assumed the same VRT would work for his or her employee. This same problem may be seen in the adoption of a Computer Aided Software Engineering (CASE) tool by individuals within an organization. If the CASE tool does not support the structured method used, then individuals and organizations will not achieve full benefits from the adoption (Rahim et al., 1997).
Examples like these demonstrate the importance of matching the task to the type of VRT system to be used. The subjects indicated that they believed part of this problem came from a lack of VRT product knowledge on the part of themselves and the individual they worked with to obtain the technology. In many instances, an individual from a vocational rehabilitation organization was involved with the individual, and sometimes the employer, in selecting the VRT system. In other cases, VRT product selection assistance came from a boss, coworker, relative, or friend. To assist in eliminating problems with task-technology fit, a technology specialist that oversees the task-- technology fit of each purchase could be employed by the vocational rehabilitation organization. Another solution is to allow individuals to try the product in their home or office for an extended length of time. While this requires an initial training program for individuals who may ultimately decide not to adopt, it would probably be cheaper to conduct more initial training periods than to purchase equipment that does not fit the task and ultimately is abandoned.
The third guideline is to obtain thorough training. The amount of training should correspond to the experience level of the individual adopting the VRT, the complexity level of the VRT system and the tasks, and the learning curve of the adopter. By doing these things the individual can learn enough to be able to take advantage of all the functions available from the VRT system. One employer provided an employee with the system and expected her to begin using it immediately because another employee at a different location had done so. The employer did not take into account the lack of computer experience the employee had. The employee was not allowed to take the system home to practice. She abandoned the VRT for her work because she was given no time or assistance to become proficient in the use of the system. Therefore, it may be concluded that a considerate employer or a modification of company policy is necessary for the individuals to benefit fully from the VRT system. A lack of adequate training has been a problem in other types of IT adoption as well (Kemerer, 1992; Rahim et al., 1997).
How training was accomplished presented another problem. Usually, training occurred for eight hours in a row for a day or two on a oneon-one basis. It may be better to stagger the person's training allowing time to practice over a few days or a few weeks. In this way, the individual may absorb more from the same number of hours of training. If a videotape or other alternative method of training were available, individuals may learn more on their own. Many individuals with disabilities cannot use VRT documentation manuals without extensive help, so recording these manuals on audio cassette or videotape may aid the training process.
The final guideline is to try the system in the actual user's environment before adopting it. Qualitative findings suggest that trialability could be the most important contributor in ensuring the successful adoption of voice recognition technology. The individual talking to the computer creates noise, and other employees may be distracted from doing their own work. Alternatively, the information to be entered into the computer may be of a private nature and should not be overheard. If the individual using the VRT system can be located in an office with a door that may be closed, this problem as well as problems with background noise may be solved.
By allowing individuals to try the VRT for an extended period of time in an actual setting, problems regarding the task-technology fit, disability limitations, and environmental concerns can be evaluated before adoption occurs. One individual was led by the vendor to believe the VRT would work well for him. His employers perpetuated this belief because it would be a much less expensive accommodation than the current one. (The employer was currently paying for a clerical worker to do the computer work as the individual directed. If the VRT was successful, the assistant position could be eliminated.) In addition to motoric disabilities, the individual had a serious speech impediment. After the system was purchased, it quickly became clear that the VRT would not work for this individual. A trial period using the VRT would have highlighted this problem before the system was purchased. This guideline may be much more important for an individual with a disability than for the non-disabled adopter of IT because the disability limitations make each adoption situation unique.
This research has also answered the question: Do the models used for the adoption of information technology innovations by individuals within an organization apply to the adoption of adaptive technology by individuals with a disability working in an organization? Yes, the general models of innovation theory may be applied to individuals with disabilities working in organizations. The only area of difference appears to be that when these models are used with individuals with a disability the limitations of the disability must be considered when matching the individual to the technology.
Although this study dealt only with voice recognition technology, the resulting guidelines are general enough for them to be able to be applied to other types of adaptive technology as well. While these guidelines appear to be common-sense, many employers do not permit potential adopters the freedom to follow the above suggestions. Hopefully, this research will provide the needed documentation allowing potential adopters to become successful voice recognition technology users.
References
Bailey, J.E. and Pearson, S.W. (1983), "Development of a tool for measuring and analyzing computer user satisfaction", Management Science, Vol. 29 No. 5, pp. 530-45.
Bowe, F. and Little, N. (1984) "Computer accessibility: a study", Rehabilitation Literature, Vol. 45 No. 9-10, pp. 289-91.
Bradt, J.A. (1991), "Pay for impact", Personnel Journal, Vol. 70 No. 5, pp. 76-9.
Brown, C. (1989), Computer Access in Higher Education for Students with Disabilities, 2nd ed., High Tech Center for the Disabled, Sacramento, CA.
Brown, C. (1992), "Assistive technology computers and persons with disabilities", Communications of the ACM, Vol. 35 No. 5, pp. 36-45.
Conrath, D.W. and Sharma, R.S. (1993), "Evaluation measures for computer-based information systems", Computers in Industry, Vol. 21 No. 3, pp. 267-71.
Cooper, R.B. and Zmud, R.W. (1990), "Information technology implementation research: a technological diffusion approach", Management Science, Vol. 36 No. 2, pp. 123-39.
DeLone, W.H. and McLean, E.R. (1992), "Information systems success: the quest for the dependent variable", Information Systems Research, Vol. 3 No. 1,
pp. 60-95.
Emory, C.W. and Cooper, D.R. (1991), Business Research Methods, 4th ed., Irwin, Homewood, IL.
Equal Employment Opportunity Commission and the US Department of Justice (1991), Americans with Disabilities Act Handbook (EEOC-BK-19), October, Washington DC.
Fichman, R.G. (1992), "Information technology diffusion: a review of empirical research", in DeGross, J.I., Becker, J.D. and Elam, J.J. (Eds), Proceedings of the 13th International Conference on Information Systems, pp. 195-206.
Ginzberg, M.J. (1981), "Early diagnosis of MIS implementation failure: promising results and unanswered questions", Management Science, Vol. 27 No. 4, pp. 459-78.
Goette, T. (1995), "Determining factors in the successful use of adaptive technology by individuals with disabilities: an exploratory field study", an unpublished doctoral dissertation, Georgia State University, Atlanta, GA.
Goette, T. (1998), "Factors leading to the successful use of voice recognition technology", Proceedings of the 3rd International ACM Conference on Assistive Technologies, pp. 189-96.
Goodhue, D.L. (1992), "User evaluations of MIS success: what are we really measuring?", in Nunamaker, J.F. and Sprague, R.H. Jr (Eds), Proceedings of the Hawaii International Conference on System Sciences: Vol. IV Information Systems, pp. 303-14.
Ives, B. and Olson, M.H. (1984), "User involvement and MIS success: a review of research", Management Science, Vol. 30 No. 5, pp. 586-603.
Kemerer, C.F. (1992), How the Learning Curve Affects CASE Tool Adoption, IEEE, pp. 23-8.
LaPlante, M.P., Miller, S. and Miller, K. (1992), "People with work disability in the USA", Disability Statistics Abstract Vol 4, Disability Statistics Program, Institute for Health and Aging, San Francisco, CA.
Lazzaro, J.J. (1990), "Opening doors for the disabled", Byte, pp. 258-68.
Loh, L. and Ong, Y.-S. (1998), "The adoption of Internetbased trading: a conceptual framework and empirical results", Journal of Information Technology, Vol. 13 No. 2, pp. 81-95.
Lucas, H.C. Jr (1978), "Empirical evidence for a descriptive model of implementation", MIS Quarterly, Vol. 2 No. 2, pp. 27-42.
Moore, G.C. and Benbasat, I. (1991), "Development of an instrument to measure the perceptions of adopting an information technology innovation", Information Systems Research, Vol. 2 No. 3, pp. 192-222.
Patton, M.Q. (1990), Qualitative Evaluation and Research Methods, 2nd ed., Sage Publications, Newbury Park, CA.
Phillips, B. and Broadnax, D.D. (1992), National Survey on Abandonment of Technology: Final Report, Request Rehabilitation Center, National Rehabilitation Hospital, Washington, DC.
Rahim, M.M., Khan, M.K. and Selamat, M.H. (1997), "Adoption versus abandonment of CASE tools: lessons learned from two organizations", Information Technology & People, Vol. 10 No. 4, pp. 316-30.
Rogers, E.M. (1983), Diffusion of Innovations, 3rd ed., The Free Press, New York, NY.
Scherer, M.J. (1988), "Assistive device utilization and qualityof-life in adults with spinal cord injuries or cerebral palsy", Journal of Applied Rehabilitation Counseling, Vol. 19 No. 2, pp. 21-30.
Sheldon, K.M. (1990), "An electronic bill of rights for the disabled", Byte, p. 6.
Tornatzky, L.G. and Klein, K.J. (1982), "Innovation characteristics and innovation adoption-- implementation: a meta-analysis of findings", IEEE Transactions on Engineering Management, Vol. 29 No. 1, pp. 28-45.
Vroom, V.H. (1964), Work and Motivation, John Wiley & Sons, New York, NY.
Ward, C. (1990), "Design for all: consumer needs assessment project year 2" (results of the second year of a five-year study), Electronic Industries Foundation, Rehabilitation Engineering Center, Washington, DC.
Welke, R.J. and Konsynski, B.R. (1980) "An examination of the interaction between technology, methodology, and information systems: a tripartite view", Proceedings of the 1st International Conference on Information Systems, December, pp. 32-48.
The author
Tanya Goette is based in the Information Systems and Communications Department, Georgia College and State University, Milledgeville, Georgia, USA.
Copyright MCB UP Limited (MCB) 2000