Research Design

This study used an online survey design to obtain baseline information from a widely dispersed population of adolescent girls who are active users of the Internet. The results highlighted in this paper are intended to explore the relevant issues and lay the groundwork for future research on youth and cyberspace. This is considered an exploratory study which introduces the issues and will need to be supplemented with ongoing research on specific characteristics of risk and preventative intervention. Furthermore, the generalizability of the study results to the larger population of adolescent girls needs to be considered. Due to the anonymity of the respondents, one of the limitations of the research design is the possibility that the survey respondents did not represent the experience of all adolescent girls or that responses were exaggerated or misrepresented. However, it should be noted that there is established support for computer-based survey design which shows that responses in this forum are as accurate or more accurate than face-to-face interviews, and there is evidence that researchers yield similar results with standard and web samples. Also, although this was a sample of "convenience," many samples in studies are neither random nor representative. In an article in the American Psychological Association publication Monitor on Psychology, Beth Azar (2000) quotes Scott Plous regarding his review of web-based samples vs. lab studies. As for the diversity of study participants, "most studies on the representation of Web-study participants suggest that, if anything, those populations are more representative of the public than samples from more traditional lab experiments." However, people on the Internet tend to represent higher socioeconomic groups with greater levels of education. Other issues of confidentiality and ethics for collecting data in cyberspace from children need to be confronted, and formal guidelines are evolving to assist in the future direction of online research (Boehlefeld, 1996; Frankel & Siang, 1999; Jones, 1999; King, 1996; Mann & Stewart, 2000; Reid, 1996; Thomas, 1996).

METHOD

Procedure

In a web-based study conducted in conjunction with Seventeen Magazine Online, CyberAngels, the College of Education at the University of South Florida, and the Department of Child and Family Studies at the Louis de la Parte Florida Mental Health Institute, an online survey was developed and placed on the Seventeen Magazine site from May though June 1999 to assess level of Internet use, involvement in varied at-risk behavior online, incidents involving negative interactions in Cyberspace, and perceived mechanisms to promote safety and well being.


Instrument

Participants completed a 19-item questionnaire that included multiple choice and open-ended questions. The questions had been piloted through surveys of middle school students in Baltimore County which were conducted by CyberAngels (Aftab, 2000). The online survey was developed using Cold Fusion software which allows online responses to be stored in a database. The data was then uploaded into a statistical software program for analysis.

Participants

The differences in girls' use of technology (AAUW, 2000) combined with data confirming adolescent girls as the group most likely to be targeted for assault (National Center for Missing and Exploited Children, www.missingkids.org), established the need to devote the study to investigation of girls' experiences in cyberspace. At the time when the survey was posted formal legal guidelines did not exist to specify requirements for age of consent online, so the legal opinion of Parry Aftab, a cyber-attorney and Executive Director of CyberAngels, was consulted and assisted in the selection of the age span for the research which best represented the legal age standard for obtaining voluntary agreement for participation in an online study. As a result of this discussion and feedback from the University of South Florida Institutional Review Board, adolescent girls aged 12 to 18 years of age were selected as qualified to participate in the informed consent process. Since these data collection procedures were completed, new federal statutes have been implemented which address consent of youth online. Subsequently, the designated age span would need to be modified or the consent procedure altered if the study were replicated due to the issuance of the Children's Online Privacy Protection Act of 1998 (COPPA) by the Federal Trade Commission. As of April 21, 2000, COPPA required commercial web sites and online services to obtain verifiable parental consent before collecting, using, or disclosing personal information from children under 13 (Federal Trade Commission, http://www.ftc.gov/opa/1999/9910/childfinal.htm).

Since the survey was placed on an open web site, exclusionary criteria for participation were specified on the consent page (i.e., girls from 12-18 years of age); however, adherence to the criteria could not be verified due to the anonymous nature of the survey. In order to minimize superfluous replies to the survey and capture the responses of adolescent girls online, an Internet site was selected for hosting the survey which possessed demographics that matched the study's targeted population; the survey was hidden within the site; and the study was not advertised in any forum so that self selection of the site's typical users could be achieved. After removing incomplete surveys, surveys completed by individuals not targeted in the research, and multiple surveys submitted from the same source, the number of responses totaled 10,800.

Data Analysis

Using the results from the Seventeen online survey of adolescent girls, the data have been analyzed using descriptive statistics, and ongoing analyses are focusing on development of a victimology profile based on probability of online risk. This conceptualization of a victim profile for youths at risk for crime, exploitation and subsequent trauma associated with their online activities is still in its formative stages. We have used logistical analysis to isolate variables that predict at-risk activity. The participants reported online experiences which challenged them to confront choices conflicting with the development of attitudes, values, and social functioning. These dependent variables include giving out personal information online, agreeing to meet with someone, receiving or sending photos, receiving and sending suggestive or threatening email, and participating in chatrooms where the content resulted in discomfort. They also identified factors which may moderate risk (independent variables), and these variables have been used to develop odds ratios and subsequent log of the odds through a logistic regression model. These include preventative activities (supervision, education, discussion) by significant adults (parents and teachers). Initially the log odds are being modeled as a linear function of the predictors, and then more advanced measurement analysis will result in consideration of multiple predictors simultaneously.