Research Methods

Research Methods

Research question: Is there a difference between the group who get bullied more in social media?

Hypothesis: Do male or female get bullied more in social media?

Prediction: Females are more likely to get bullied more through social media.

Research Method

Participants

Three hundred and fifty university students drawn from various campuses across the United States participated in the experiment.  Most of the students were predominantly Americans though other nationalities were included. The students included both males and females and were of different races.  One hundred and forty students were male, and one hundred and ninety students were female. More details on age, race, or ethnic background and gender are provided in the appendix section at the end. According to Bourne et al. (2016), “Information on participants helps other researchers understand how your study was performed, how generalizable the result might be, and allows other researchers to replicate your results with other populations to see if they might obtain the same results.” The method used to select the students was a random sampling technique where students were chosen by chance, and each member of the included campuses had an equal chance of being selected. Cohen, Manion, and Morrison (2011) suggests that “the quality of research depends on ‘the suitability of the sampling strategy that has been adopted’ (p. 97).” The participants volunteered themselves to be interviewed, and no sort of compensation was included in the process. The study had been advertised on various campuses, so most of the selected students already knew about and willing to freely participate.  All the chosen students had at least used one social media platform within the last three months of the study. They came from different nationalities and socio-cultural backgrounds. Social engagements and involvement with activities in the university proved an essential aspect in detection on the level of active participation and the rate of using social media. The participants were expected to give accurate information without lying, or else the study would be a waste of time and resources since false data cannot provide credible results.

Material

The study material used in the collection of data used quantitative methodologies to get better insights into the cases and incidents of cyberbullying experiences. The instructional material used was a questionnaire. The questionnaires were designed in such a manner so as to explore the students’ understanding of cyberbullying and whether they had ever been victims of cyberbullying. The questionnaires were made of short questions with limited choices to avoid collecting extensive unrelated data. The questions were open-ended and had scaled responses. Piloting test had been conducted in the verification of the reliability of the questionnaires for the actual survey. Existing literature on cyberbullying and the theoretical framework provided a trustworthy guide in the development of the study. Ferris et al. (2019) state that “The credibility of the responses you receive from questionnaires depends on the structure of your questions.” He further emphasized that poor structuring will result in inadequate responses. This helped in examining the reliability of our questionnaires from the pilot test.

Design

The hypothesis “Do males or females get bullied more in social media” had to be confirmed in the study or rejected. Data collected from the students in the questionnaires had been analyzed quantitatively. The control in the study was that all the sampled participants had access to social media, so the questions provided in the questionnaire were at least relevant to them. The independent variable was having or not having knowledge about cyberbullying. The dependent variable was being a victim a cyberbullying what depended on whether you understood what it was in the first place. There are several manipulations to be carried out in order to test the hypothesis. These manipulations include carrying out calculations on the percentage or fraction of the sample that understood cyberbullying. Also, calculating the percentage of those who understood cyberbullying and have been victims of it. The final step is evaluating the percentages of the victims of bullying who were males or those who were females. Then you can run the hypothesis to confirm whether it was a reasonable hypothesis and accept it, or whether it was a weak hypothesis and reject it. The factors that are varied are the knowledge of cyberbullying and whether you have had an experience in cyberbullying, a victim. In order to have good control of the study, the questions are structured in a hierarchical manner such that you have to answer question one before you proceed to the next question. If you skip one question, then the response in the following question will be irrelevant. For example, if you do not answer the question about knowing or knowing whether cyberbullying exists in social medial, then it is impossible for you to provide a reliable response on whether you are a victim of cyberbullying or not. Therefore, through structuring the questions in a specific manner, one is able to rule out or control the apparent effects of the study.

Procedure

The printed out questionnaires were distributed to the various universities that participated in the study. The selected sampled students were issued with the questionnaires and given a period of one month to fill the questionnaires. The duly filled questionnaires were returned to the collection center at the university. During the exercise, the participants were required to provide clear responses by themselves, not to include another person in their exercise. They were encouraged to be honest and provide reliable data; if they do not understand a question, they would better leave it instead of guessing.

The questionnaire had four parts to be answered. The first one was collecting the participant’s demographic data such as gender, age, and ethnic background. An important question considered in this part was their membership in any club at the university so as to try to determine the level of social engagement. The second part of the questionnaire focused on the participant’s social platform exposure and involvement. Their behavior on social media and their perception concerning cyberbullying, in general, was also considered in this section. The third part of the questionnaire explored their social media personal experience or contact with cyberbullying and whether they reported such cases. The last part of the questionnaire examined the participant’s usage of social media platforms. It also included their personal views and experiences in relation to both traditional bullying and cyberbullying.

The survey ended with four open-ended questions on which participants reported personal incidents as well as their opinions regarding cyberbullying. Unfilled questionnaires were returned and accepted as an indicator of not willing to take part in the survey and were later separated from the other questionnaires to ensure total anonymity.  Data analysis was carried out using Cronbach’s alpha or coefficient alpha, which was developed in 1951 by Lee Cronbach (Bonett, D. G., & Wright, T. A. 2015). Tests were done using the Cronbach’s alpha to determine whether multiple-question Likert scale surveys are reliable. The tests measure the internal consistency of variables that are extremely challenging to measure in real life. Analysis of the collected data was guided by the initial research question raised at the beginning of the study.

Appendix

Figure 1. The distribution of the sample according to gender.

Figure 2. The distribution of samples according to ethnic background.

Figure 3. The distribution of sample according to age.

References

Bonett, D. G., & Wright, T. A. (2015). Cronbach’s alpha reliability: Interval estimation, hypothesis testing, and sample size planning. Journal of Organizational Behavior, 36(1), 3-15.

Ferris, M., Cohen, S., Haberman, C., Javalkar, K., Massengill, S., Mahan, J. D., … & Phillips, A. (2015). Self-management and transition readiness assessment: development, reliability, and factor structure of the STARx questionnaire. Journal of pediatric nursing, 30(5), 691-699.

Bourne, M., Parker, S., Terry, S., & Free, R. (2019). The impact of a new recruitment strategy to improve recruitment of participants to research studies.