V. DISCUSSION
The results of the GEE assessment indicate that there is no statistically significant relationship between the number of females who were cyberbullied and the number of females who seriously considered suicide. The results are important because an earlier research study conducted by Peng and Davis (2017) used much of the same data, and their findings indicated that female students who were cyberbullied were more likely to have seriously considered suicide. However, unlike the current study, Peng and Davis assumed that the observations were independent. When the observations were not considered independent, as in the current study, female students who were cyberbullied were not more likely to have seriously considered suicide. This study is important because it shows the importance of assumptions, and how those assumptions can impact the findings.
Limitations
There are several limitations in this study. First, because the sample is limited to female students in grades 9-12, the findings cannot be generalized to other populations. Second, it is not possible to know the actual number of female students who were electronically bullied because many students who are bullied never officially report it to authorities (Loveless, 2020). Third, because cyberbullying is a recent issue, there are only four youth risk behavior surveillance surveys of available data. Thus, the amount of data available is less than optimal. Fourth, because the data used in the study were second-hand and collected for a different reason, the data values cannot be more clearly defined. Finally, because the study is quantitative in nature, it does not provide an in-depth understanding of the meanings that the participants have associated with their lived experiences (Berg, 2007).
Further Research
Following are ideas for additional research. First, to eliminate the problem of using different sample sizes from each state, a researcher could collect original data using a fixed sample size from each state. In addition, if the data were collected at one point in time, then the researcher could help ensure that the data values are independent. By doing this, the researcher may be able to use a parametric statistic to assess the relationship between the number of females who were cyberbullied and the number of females who seriously considered suicide. This is important because a parametric statistic will usually have more statistical power than will a nonparametric statistic (Field, 2005; Su, 2020). Second, even if a study indicates that there is a relationship between the number of females who were cyberbullied and the number of females who seriously considered suicide, it will not indicate why they seriously considered suicide. A qualitative study needs to be performed to address the reasons why individuals feel as they do. Finally, specific theories that are believed to be linked to cyberbullying can be studied. Whatever theory is used to explain the problem, the same theory should be used to resolve the problem. To do this, data must be collected in a manner that is in alignment with the theory of interest. For example, if it is believed that the deterrence theory can be used to reduce cyberbullying, then data on deterrence and cyberbullying need to be collected. If the fear of punishment is quantified and is shown to be inversely related to the amount cyberbullying, then a resolution based on the celerity, certainty, and severity of punishment can be implemented to address the problem (Barkan, 2006).