OVERVIEW OF WEB-BASED RESEARCH
This section first introduces Web-based research, and briefly points out some advantages and disadvantages or concerns. Then, Active Server Pages are introduced.
ADVANTAGES & DISADVANTAGES
The advantages and disadvantages of Web-based research are covered more in-depth in several of the articles cited on the first page. Some of the frequently-cited advantages of Web-based research (in comparison to paper-and-pencil versions of research) include that Web-based research is:
The most frequently-cited advantage is that Web-based research projects are accessible by a larger population. For running projects only with undergraduates, this is not very important. Most students now have computers at home or in the dorms, and there are many accessible computer labs throughout universities. And the undergraduate population will still be limited to those who are required to participate in research because of their psychology or other course. However, projects directed at adult community members can reach a large population with greater ease. Similarly, undergraduate samples at universities other than the researcher's own can be reached over the Web with greater ease. This allows greater access to research for those undergraduates who are required to do research for a course, but the university does not offer enough opportunities within its own research projects.
Also, projects that would otherwise be limited to undergraduates could expand to include community members without the effort of compiling addresses and the costs of mailing materials. Of course, while some adults might complete the research project without compensation, a higher response rate can be obtained by offering payment. This may not always be affordable to the researcher. Also, the population is limited to those who have access to the Web; while now that is typically a young, white male, the population is homogenizing daily as more people gain access.
The costs of Web-based research are reduced because there is no need to copy hundreds of pages of research materials. Also, there is no need to occupy researchers' or their assistants' time overseeing undergraduate participants filling out the paper-and-pencil versions. Web-based research is faster in the sense that data is automatically entered into a database by the Web server. Of course, it may take longer to complete data collection because participants can choose when to participate, and thus may procrastinate.
Finally, Web-based research is interactive in two senses. First, Web pages can be programmed to give immediate feedback to participants upon completion of the experiment. Second, the content of Web pages can be manipulated by certain programming to give different material depending on participants' prior responses. For example, one prior experiment required that participants be presented with arguments to change their initial sentencing verdict. These arguments therefore had to correspond to the participant's initial verdict, which was recorded on a previous Web page. Over the Web, those who voted for death were able to read an argument supporting a life verdict, and vice versa. To do this with paper and pencil would require greater (and quick) effort by the researcher.
Of course, there are also several issues that must be addressed when doing research over the Web:
As noted above, it is possible to reach a larger population of adult community members over the Web. Besides the obvious demographic differences between them and undergraduate samples (which actually could be a benefit in the sense of greater generalizability), this also raises the issues of self-selection and motivational differences. While undergraduates are "required" to participate in research for credit, adults are not. Those who participate without compensation may be participating because of an interest in research (unfortunately probably lacking in many undergraduates) or an interest in the subject matter of the research.
Also, researchers should have a way of controlling against one person completing the experiment more than once. As described in another section of this guide, it is possible to record or ask for participants' email addresses and IP addresses. Researchers will have to monitor this information to ensure that they do not repeat within one study.
Although several authors (listed on the first page) have concluded that, for their project, Web-based research is valid (i.e., it does not effect the results of the study), further investigation is needed. Both the different samples reached and the different medium involved could effect several aspects of research: the means and variability of dependent or measured variables; the strength of the effect of manipulated variables (i.e, sample or method would interact with the manipulated variable); and the factor structure of any scales or measures included in the research. If one project is conducted using both the Web and paper and pencil, validity can be assessed. Otherwise, validity will remain an issue to be investigated.
USING ACTIVE SERVER PAGES FOR WEB-BASED RESEARCH
All the other sections of this guide are directed toward using a specific method for doing Web-based research. That method employs Active Server Pages to manipulate variables (for a study), to control a participant's movement through the Web pages, and to collect the participant's data and write it to a database. Certainly there are other methods for accomplishing these tasks; a popular one is CGI (Common Gateway Interface). The advantages and disadvantages of ASP and CGI could be debated here but will not be. For the simple fact that I learned how to use ASP for studies over the Web and do not know CGI and Perl programming, this guide focuses on ASP. I continue to believe that ASP is easier to use (and to learn), although it can only be used on Microsoft servers (for the time being).
Both ASP and CGI operate on the same distinction between client-side and server-side scripting. Client-side scripting, such as JavaScript, is code that is written into Web pages and that is executed by the client's (the participant's) Web browser, so that the browser performs the proper function or task. In contrast, both ASP and CGI involve code that is read and executed by the server, before content is sent to the client's browser over the Web. Generally, the ASP or CGI scripts produce output in HTML form, which is then written into the Web page that is sent to and received by the client's browser. The main differences that I perceive between ASP and CGI for our purposes here are: (1) ASP is written either in VBScript (a version of Visual Basic) or JScript (a version of JavaScript) while most CGI is written in Perl or C++, which I consider more confusing, although more widely applicable, programming languages; and (2) ASP is written right into a Web page, while CGI scripts are files or programs separate from the Web pages.
IF YOU WISH TO RETURN THE HOME PAGE, CLICK HERE
IF YOU HAVE COMMENTS OR QUESTIONS, EMAIL ME AND LET ME KNOW