Introduction
The article by Hinduja and Patchin (2013) is the outline of the author’s research into the matter of the impact that cyberbullying has on the youth. This article’s title is “Social influences on cyberbullying behaviors among middle and high school students,” which reflects the scope of the study. This paper will critique the research design and methods Hinduja and Patchin (2013) selected for their research and offer a research strategy for an extension of this study.
Article Critique
The central idea of this research is that the contemporary youth faces a new type of bullying behavior — cyberbullying and that the social influences impact a person’s likelihood of engaging in this behavior. For example, if an individual watched a cyberbullying behavior, will they be motivated to report it to parents or educators, or will this affect their perception of cyberbullying differently.
The theory of socializing agents that guide the choices of adolescents is the basis of this research. As children grow up, they watch and learn from the behaviors of their parents, and this becomes a norm, later allowing for social conformity (Hinduja & Patchin, 2013). The hypothesis is that adolescents who know peers engaged in cyberbullying are more likely to engage in this behavior. In contrast, those who know that their parents or educators will punish them for cyberbullying are less likely to cyberbully. The results show that the initial hypotheses are correct, and cyberbullying behaviors are influenced by the social agents surrounding the adolescents. Hence, the conclusions support the hypothesis and show that students cyberbullying behavior is impacted by their peers, educators, and parents.
Strengths and Weaknesses
First and foremost, this study has a large sample size of 4441 individuals from a southern school district in the United States. Moreover, the researchers report having 49% of female respondents and 69% of non-white individuals, which helps gain a better comprehension of cyberbullying’s impact on minorities and vulnerable populations (Hinduja & Patchin, 2013).
A significant weakness of this study is the potential biases in responses because the researchers aim to gain a comprehension of a person’s malicious behavior. More specifically, the questionnaires aimed to determine if these adolescents engaged in bullying and their attitudes towards this behavior, but it is possible that the results were affected by the unwillingness of these individuals to admit their adverse behavior.
Research Design
In this study, the students were asked to respond to a web-based survey with nine questions, reflecting on their experiences within the past nine days. In summary, the researchers conducted their study in 2010 by distributing their questionnaire to a sample of 4441 students from 33 schools in one district from 6th to 12th grade (Hinduja & Patchin, 2013). The researchers distributed surveys to schoolchildren on a school day, and the rate of completion was 99% for those students who attended classes on the day of the study. The dependent variable is cyberbullying behavior, and the independent variables are peer involvement and informal sanctions.
The potential reasons why the researchers choose a survey as a method, and therefore qualitative design include the large sample size and the type of results they expected. According to McLeod (2019), qualitative methods allow one to confirm or reject the hypothesis and test the relationship between variables. Hence, qualitative methods are appropriate, since, in this study, the premise was to understand whether the social agents around students affect their cyberbullying behavior, or if simplified, the researchers aimed to understand if the answer to the hypothesis is either “yes” or “no” and not gain a better comprehension of related factors, such as experiences, motivation, perceptions, and others.
Hence, the justification for the selected research method is the scope of the research and the hypothesis that the authors chose — the goal was to either confirm or reject the hypothesis. Another justification for the research design is the sample size since 4441 respondents require a method of data collection and analysis that is tailored to be used on large quantities of data. From this perspective, an online survey is an excellent method because the researchers can export the responses right away into the software, such as SPSS, and analyze it right away (“Questionnaire design and analyzing the data using SPSS,” n.d.). With quantitative data, for example, interviews, the resources and time required to collect and analyze the responses from this population would be immense, making it impossible to complete the study. Thus, the research methods in this study match its purpose, scope, and hypotheses. In contrast, if the researchers choose to use qualitative methods, they would have a lot more information for analysis.
Sampling
The sample of 4441 represents a larger population of middle and high school students, and the researchers concluded that it is sufficiently large and diverse, based on the results of a t-test (Hinduja & Patchin, 2013). Based on the description of the selection of the sample, the researchers applied the random sampling technique. According to Khan Academy, under simple random sampling, each representative of a population has an equal chance of being selected for the study (“Sampling methods review,” n.d.). In this case, the researchers defined the general characteristics of the population they were interested in and allowed the school administration to select the participants.
The researchers selected the school districts and asked the educators to choose classes that will be responding to the survey. Hence there was no direct involvement of none of the researchers in the process of selecting the specific individuals for this study. Considering the large sample size, this method is appropriate because it allows for saving time and effort while the results are still representative of the broader population.
Another way of sampling that could be applied for this research is voluntary sampling, where the students would be asked to participate. Khan Academy Article describes this method as the researcher explains the aim of their study and “request for members of a population to join the sample, and people decide whether or not to be in the sample” (“Sampling methods review,” n.d., para. 5). In this case, the researchers would avoid bias or fear of disclosing harmful behaviors that these students might have had when answering the survey questions in a classroom with their peers.
The resources required to implement this study design include school representatives who would hang posters and explain this study to students and the materials needed to create these posters. Hypothetically, this sampling method would require a slightly greater budget but would help remove bias and incorrect responses because students would participate voluntarily. However, no changes to data collection and analysis would be necessary in this case.
The Trustworthiness of the Results
The results appear to be trustworthy since the authors used statistical analysis to test the validity and reliability of these answers. Moreover, since the researchers found that many students reported that they bullied others after seeing their peers do the same, it appears that the students were not afraid to report their adverse behavior. The authors argue that they used SPSS software to process the responses of the students, and based on the outcomes, for example, the number of students who claimed that they witnessed bullying behavior and them reporting that they cyberbullied others, made conclusions (Hinduja & Patchin, 2013). Descriptive statistics methods allow summarising data and drawing meaningful insights from it (“Descriptive and inferential statistics,” n.d.; “Descriptive statistics,” n.d.). In addition, they report calculating the variation of responses, which suggests that the results are trustworthy, and the researchers completed the necessary steps to verify them. The researchers also used other methods, such as a 2×3 table to compare the bullying behavior and response of adults. Finally, they created “two ordinary least-squares (OLS) regression models” to detect the patterns in students’ answers (Hinduja & Patchin, 2013, p. 720).
Extension of this Study
To conduct a study for IDA about the differences in drug use among men, women, and adolescents who live in poor areas, examine different patterns of drug use, and the impact of anxiety on drug use methods, one should apply quantitative methods. Similar to Hinduja and Patchin (2013), the use of a survey will allow collecting a large sample of data since IDA wants to draw insights from three distinct groups — men, women, and youth. To construct a questionnaire, the researcher will have to develop a hypothesis, and the first step is to conduct a literature review to find the basis for this hypothesis. For instance, the assessment may suggest that men are more likely to use drugs, while women who use drugs report having more anxiety. Hence, the hypothesis would be that men are more likely than women and adolescents to use drugs and that women report higher levels of stress.
The survey questions that would help find answers include the assessment of demographic characteristics and “yes” or “no” questions, for example: “I have used drugs within the last 14 days” and “In the past two weeks I felt extreme anxiety.” Next, after collecting the responses, the researcher would use SPSS to summarize the answers and see if anxiety is linked to drug use and if there are other detectable patterns. To assess the lack of adequate services in the area, question such as “I perceive access to public services in the area as adequate” can be included.
For the sample selection, the researcher could partner with social services and use random sampling. To account for the generalization issue, the researcher should use a sufficient sample for each of the three groups. For example, in line with the critiques study, collecting 4400 responses from representatives of the three groups would be ideal. However, the exact sample size suitable to make generalized conclusions should be calculated based on the number of individuals who use drugs and live in impoverished areas.
Improving this Research
As was discussed in the section about the weaknesses, there are some areas of improvement within this project. For one, there are concerns about the potential bias of responses due to the sampling strategy and the way in which the researchers conducted their assessment. The students were recruited by allowing their parents to send a form declaring they do not wish their children to participate, which, as opposed to willing full participation, creates some issues with the responses. Hence, by selecting a different sampling strategy, the researchers would be able to get more significant results. In addition, by using mixed-method research, for example, surveys and interviews, they would be able to build on the quantitative data with some insights from the information collected from qualitative methods. In this way, the researchers would be able to gain a better comprehension of how and why parents’ and educators’ attitudes towards cyberbullying impact students’ engagement in this behavior.
These improvements would help achieve better accuracy of results and gain a better comprehension of the results. However, one must admit that this approach has a limit on the number of interviews the researchers can conduct since this method is time and resource-consuming, and analysis of qualitative data is more intricate when compared to statistical analysis. In terms of the budget for these improvements, depending on the number of interviews, the researchers would have to dedicate more resources to complete this study.
Conclusion
Overall, this paper evaluates the article by Hinduja and Patchin (2013) and the research methods these authors used. The researchers aimed to detect patterns of cyberbullying behavior using the social agent’s theory. The research design, methods, and analysis of the data suggest that the study’s conclusion is reliable. An extension of this study aimed at examining patterns of drug use among adolescents can be conducted with the same methods, such as quantitative design and survey.
References
Descriptive and inferential statistics. (n.d.).
Descriptive statistics. (n.d.).
Hinduja, P. & Patchin, J. W. (2013). Social influences on cyberbullying behaviors among middle and high school students. Journal of Youth Adolescence, 42, 711–722.
McLeod, S. (2019). What’s the difference between qualitative and quantitative research? SimplyPsycology.
Sampling methods review. (n.d.).
Questionnaire design and analyzing the data using SPSS. (n.d.).