Internet Resources
for Clinical Psychologists,
Part II
Shawn R. Currie, Ph.D.
Addiction Centre, Foothills Hospital, Calgary, AB
Part I of this article [Canadian Clinical Psychologist, 10 (1)] looked at the Internet as a clinical tool. Part II will examine the potential of the Internet to assist in research with a specific focus on the pros and cons of online data collection.
Finding Scholarly Information
Most researchers make use of online access for searching Medline, PsychInfo or other scholarly databases. Choice of the web-based versus text-based services is a matter of personal preference. Since the last article, I have tried to find a site offering free, unrestricted access to PsychInfo but must admit to being unsuccessful. Searching this database from home or office will require a subscription or library privileges at a university or college. There seem to be fewer restrictions placed on access to Medline. One of the advantages of the "information superhighway" is that the boundaries of literature searching have been expanded beyond the limitations of Medline/PsychInfo searching. There are numerous other databases now available via the user-friendly web. For example, there is the Cochrane Database of Systemic Reviews which is a collection of meta-analyses and comprehensive reviews on health care interventions. A similar database is Best Evidence, an index of scientifically rigorous outcome studies with expert commentary on the value of each study for clinical practice. Both of these databases are available through most online university library networks. For those in the addiction field, ETOH (etoh.niaaa.nih.gov) is a bibliographic database maintained by the U.S. National Institute of Alcohol Abuse and Alcoholism which is specific to articles on alcohol abuse and treatment.
In addition to peer-reviewed journal articles, the web can also be helpful in finding other sorts of useful information. For example, it is easy to obtain government documents over the Internet. Both the NIH (www.nih.gov/health) and Health Canada (www.hc-c.gc.ca/english/research.htm) web sites provide listings of most of their publicly available documents such as research monographs, government reports, research updates and other publications. Many reports can be downloaded directly from the web. You can get the first-ever Surgeon-General Report on Mental Health at www.surgeongeneral.gov/library/mentalhealth/home.html.
Finally, if you are ready to publish your own research, the Internet provides quick access to the instructions to authors for most psychology journals. The Psych Web site provides an excellent list of journals on the web with direct links to each journal's publisher (see www.psychwww.com/resource/journals.htm). Check out the same site for a list of resources on APA style.Statistical Information on the Web
There are more than a hundred web sites offering statistical advice and even online calculation of popular statistical tests. All one needs to do is input the means, standard deviations, counts, etc. into an interactive set of pages and the program will instantly do the analysis you request and give you the results in a matter of seconds (useful if you can't afford, or don't have access to SPSS). The AmoebaWeb site (www.sccu.edu/programs/academic/psych/webstatistics.html) is a good starting point with many links to other statistical sites. Online power analysis for calculating sample size needs is also available from a number of different sites. One maintained by the Mississippi State University Music (yes, music!) Department at www.msstate.edu/dept/musiced/daniel/poweranalysis.html contains a great list of web sites dedicated to power analysis. A power calculator is available at www.stat.ucla.edu/~jbond/HTMLPOWER/index.html.
Research Grants
All of the major granting agencies have web sites with detailed information on their funding programs including, in many cases, the actual applications ready for downloading. Unfortunately, electronic submission of grant applications is still not available. Most sites will also publish the results of recent funding competitions, a helpful feature for getting an idea of the kind of research being supported by a particular agency. The NIH site has it's own search engine CRISP (commons.cit.nih.gov/crisp/owa/CRISP.Generate_Ticket) which allows the user to search current and previously funded projects and training awards by investigator name, topic or other key words. The Internet can be a useful tool for finding smaller funding agencies, both private and government supported, that are potential sources of research dollars. Most universities provide a list of external granting agencies on their web site. My old alma mater Dalhousie University has a good list (www.dal.ca/~research/) and so does McMaster (www.mcmaster.ca/ors/funding).
As our own health research funding infrastructure changes from MRC/SSHRC to the new Canadian Institutes of Health Research (CIHR), it is nice to get the updates over the Internet (see www.mrc.gc.ca/cihr-icrs/cihr.html). Research institutes that function "without walls" such as the CIHR somehow seem more real when given a home on the Internet.
Recruiting Subjects on the Web
Web-based subject recruitment is a new trend. Researchers can place notices on consumer-oriented web sites, or within 'newsgroups' and 'listservers' in order to advertise research studies. This form of recruitment is really no different from using consumer groups' newsletters to solicit subjects. Assuming you have permission of the group or agency, and the notices are not misleading in any way or obscure the objective of the web site or newsgroup, Internet recruitment is generally sanctioned. The same ethical principles that bind conventional subject recruitment apply to Internet recruitment. Some researchers have taken it upon themselves to propose specific guidelines for conducting Internet research (Childress & Asamen, 1998; Michalak & Szabo, 1998). Because these guidelines have not been adopted by any psychological association they are best viewed as sound advice on the topic. For example, one recommendation is that the recruitment of subjects for mental health interventions be restricted to the investigator's immediate geographic location to avoid any legal/jurisdictional issues associated with cross-border research. Furthermore, investigators should always provide a physical address and telephone number in addition to an e-mail address for subjects to respond or contact the investigator for further information.
Conducting Research Studies Online
The latest Internet phenomenon is conducting research studies online. Michalak (1997) administered a web-based questionnaire to study symptoms of seasonal affective disorder (SAD) to a sample of 425 internet users across 35 countries. Szabo, Frenkl, and Caputo (1997) examined addiction to running using a similar methodology. Both of these studies demonstrated the potential of the Internet for conducting survey research on the web. Foremost, there is the low cost and ease of administration. A questionnaire can be easily constructed for the web, which can then be accessed by literally thousands of users within a fraction of the time that a traditional mail out questionnaire can take. A web-based study could attract subjects who would not normally participant in a mail-out survey. Moreover, the return of the information is virtually instantaneous, and can easily be downloaded into a database or statistical program without any hand keypunching required. In addition to surveys, computer programming advances have made it possible to conduct and even design learning experiments involving facial recognition, memory tasks, etc., on an interactive web site (Murray, 1998). The Internet is also being looked at as a tool for delivering and evaluating psychoeducational types of interventions (Childress & Asamen, 1998). It is thought that the anonymity afforded by the web medium will ultimately attract more participants than traditional face-to-face or even telephone interventions.
Despite the advantages, there are many ethical and scientific concerns associated with conducting this form of research. One major concern is the security of the information being transferred electronically. Researchers should take steps to ensure any data is properly encrypted to prevent unauthorized access by computer "hackers." In addition, there are copyright issues to contend with when researchers plan to put published instruments on the web. Another concern is how to obtain proper informed consent, and ensuring the opportunity to debrief participants if necessary. Although written information about the study purpose, risks, etc. can easily be provided over the web, there should still be the opportunity for potential subjects to ask the investigator questions directly via a telephone contact or interactive e-mail.
A large segment of the Internet population is composed of teenagers and older children. With this expanding consumer group comes the potential to recruit large numbers of young subjects into research studies without having to contend with the red tape involved in going through the school system. However, there is a concern that such an approach would constitute exploitation of a vulnerable population. Certainly, one would takes steps to verify that parental consent has been obtained in an informed and ethical manner. Nevertheless, researchers need to prepare for all contingencies (e.g., disclosure of child abuse) when working with a vulnerable population in a largely anonymous format.
In addition to ethical concerns, the scientific validity of Internet-collected data may be questionable. Any information obtained via the Web is subject to the same standards of validity and reliability as conventional survey data. Michalak and Szabo (1998) recommend that researchers confirm that a web-based questionnaire is comparable to the same items administered in a paper-and pencil format (i.e., conduct an alternative forms reliability check). Even if this can be established, there is no way to control the circumstances or environment in which a study instrument is completed over the Internet. A large number of users log on during off-peak hours when the availability of phone-in circuits is greater and rates are sometimes lower. A questionnaire on depression may elicit quite different responses from a subject who fills it out at 2:00 a.m. compared to 10:00 a.m. One would also need some basic verification that the respondent is from the intended population of subjects.
The geographic reach of the Internet provides both advantages and disadvantages for online research. While it is possible to recruit subjects from diverse, even international, geographic regions, investigators need to consider the impact of such heterogeneity on the validity of the results. For example, the Internet is predominantly an English service, but the Web is accessed by a large sector of the population for whom English is a second language. Verifying competency in English becomes more of a challenge in cyberspace.
Although the percentage of the population surfing the web is growing every day, it is safe to say that an Internet-recruited sample would not generalize to all persons. Internet users are not representative of the entire socioeconomic spectrum, and persons who volunteer to complete an online survey or participate in an Internet intervention may reflect an even smaller segment of the population. The impact on a study's ecological validity will depend largely on the topic of the research. A study on 'Internet addiction' may not be affected, whereas a primary prevention program for depression among single mothers would be certainly. As with all research, investigators need to acknowledge the limitations of their sample and conclusions derived from the research.
Final Thoughts
There is no doubt that the Internet has expanded the boundaries of research for those who feel comfortable with the technology. Furthermore, the speed, ease, and coverage of the web medium will likely attract more psychologists to consider online data collection. Nevertheless, I imagine most clinical researchers will continue to conduct their studies "the old fashioned way," and probably with good reason. As noted above, there is still a large segment of the population who do not have the financial resources to access the Internet and many of these individuals are the focal group for mental health interventions. I will end by mentioning an interesting study by Bier, Sherblom, and Gallo (1996) that demonstrated the potential of the Internet to serve as an intervention on its own. The investigators provided low-income families with free computers and a subscription to an Internet service. They found self-esteem and confidence increased among family members. Many participants reported greater self-efficacy in being able to find relevant health information with the web. Though not a controlled trial, this simple study suggests the Internet can have an influence on persons who are generally considered at risk for mental health problems like depression. However, it may also suggest that as more and more human activities (e.g., banking, shopping, communication) get transferred to an Internet format there is potential for persons who are not connected to feel left out and become marginalized even further. Exploring the social impact of the Internet and the role that clinical psychology will play in this evolving technology will certainly be a fertile topic of future research.
References
Bier, M. C., Sherblom, S. A., & Gallo, M. A. (1996). Ethical issues in a study of Internet use: Uncertainty, responsibility, and the spirit of research relationships. Ethics and Behavior, 6, 141-151.
Childress, C. A., & Asamen, J. K. (1998). The emerging relationship of psychology and the Internet: proposed guidelines for conducting internet intervention research. Ethics & Behavior, 8, 19-35.
Michalak, E. E. (1997). The use of the internet as a research tool: The nature and characteristics of seasonal affective disorder (SAD) amongst a population of users. Interacting with Computers, 9, 349-365.
Michalak, E. E. & Szabo, A. (1998). Guidelines for Internet Research: An Update. European Psychologist, 3, 70-75.
Murry. B. (1998, November). New tool makes online research easier for students. APA Monitor, p. 1.
Szabo, A., Frenkl, R., & Caputo, A. (1997). Relationships between addiction to running, commitment to running and deprivation from running: A study on the internet. European Yearbook of Sport Psychology, 1, 130-147.