Posted by arlene

While such approaches may be perfectly feasible (and we do not wish to discourage creative use of technology), it is always worth considering whether the target population has access to the software required. People can only participate in a study if they have the appropriate software installed and configured, so an effective approach is to use systems that are already highly widespread and widely used (say, email or web browsers). If participants are required to install special software and learn new systems, many will be discouraged from participating, either because of the time and difficulty involved, the download costs (several megabytes for most contemporary client packages), or even because they do not trust unknown software on their systems. Thus the use of more exotic technologies will inevitably reduce the size of the accessible population.

Even in more controlled situations, such as departmental Intranets, participants‘ levels of knowledge and experience can cause difficulties. In studies using web browser technology and undergraduate student samples, some participants have little idea how to use a web browser, even when considerable efforts have been expended to make the interface as simple and transparent as possible. Sometimes they seem to worry that they should not trust the instructions the program is giving them, as if they assume that the software contains an error, or that the instructions on the screen are not intended for them. For example, at the end of one study, the browser window displayed a message saying ‘The experiment is over. You may now close this window’, which prompted several people to check with the researcher whether this was true? Such difficulties are generally easy to resolve when the researcher is present, but could cause substantial disruption when help is not immediately available, such as when the materials are delivered across the Internet.

In the Intranet case, it is even dangerous to assume basic familiarity with computers, or with the particular platform you are using — a friend has confessed that when he recently upgraded to a Windows machine, after years using only MSDOS, he had no idea how to use a mouse, and needed detailed instructions; we have also encountered some undergraduates with the same difficulty. Participants will usually be uncomfortable when using technologies of which they know little, and such factors can influence the results, for example by depressing performance levels on the computer-based task relative to a task using more conventional technologies such as paper.

Ease of use is always an important consideration when writing software, but especially so when the user is likely to encounter it only once. While in the general case, a user can overcome a difficult interface if necessary by repeated exposure, in the research case the interface needs to be transparent and to work immediately, first time, with minimal effort on the user’s part. Interface problems can cause the sample to be skewed, when some classes of users are unable to complete the study for technical reasons, or, worse, they may return meaningless data that bias the results.

Possibly related posts: (automatically generated)
Web Searching, I do not use Software and Protocols

• The Active Directory Schema continue...
• What Is a Domain Tree? What Is a Forest?
• New Rules of Engagement
• Create a Smooth Path to Buying (Navigation), Successful Websites Rule
• Windows Internet Name Service part 2
• Sender Authentication Systems
• What Is a Domain Tree? What Is a Forest? continue...
• The Active Directory and Dynamic DNS continue...
• Installing Active Directory on a Windows Server 2003 Computer
• The Active Directory Service and Windows Server 2003
• Inter-domain Trust Relationships