Defining and Measuring Bullying in Adolescents
Description
This study seeks to consolidate existing literature on bullying in adolescents and develop a standardized inventory that future bullying researchers could use to assess and study bullying. It will involve first creating a measure, and asking adolescents to give input on the questions formulated by researchers. A final measure will be distributed to adolescents via Qualtrics, and assessed for validity and reliability.
Objectives and Hypothesis
While there is a great deal of research on bullying in adolescence, much of the literature relies on a standard definition of bullying given by Olweus (1993). While this definition is intuitive, it has not led to any standard measure being used by researchers.
The former is problematic because students’ individual definitions of bullying tend to be inconsistent (Cornell et al., 2012). No person wants to feel that they are engaging in negative behaviour. The latter is a problem because these behaviours are not consistent across studies, and often do not consider cyber-related bullying, which is one of the most prevalent types of bullying happening among youth.
This inconsistency in the literature has caused confusion. Because researchers are unable to agree on a standardized measure for bullying, it is difficult to talk about how data compares across studies. To help address this inconsistency, our objective for this project is to put some work toward developing an inventory of bullying traits and behaviours that could be used to test bullying in Canadian high schools. A good inventory is important for being able to talk in academic circles and provide a clear message to educators.
This research is important to the wider community because understanding how adolescents define bullying, and developing an inventory to test bullying will disambiguate the literature. This in turn will benefit future efforts to research and prevent bullying in high schools—both in the classroom and online.
We hypothesize that our final measure for bullying will be valid, reliable and provide a more accurate analysis of bullying behaviour in adolescents than measures used in the past. The final produced measure will able to be used to research bullying behaviour in subsequent studies.
Sources:
Cornell, D., & Cole, J. C. M. (2012). Assessment of bullying. In S. R. Jimerson, A. B. Nickerson, M. J. Mayer & M. J. Furlong (Eds.), 2nd ed.; handbook of school violence and school safety: International research and practice (2nd ed.) (2nd ed. ed., pp. 289-303, Chapter xxiii, 645 Pages) Routledge/Taylor & Francis Group, New York, NY.
Olweus, D. (1993). Bullying at school: What we know and what we can do. Malden, MA: Blackwell Publishing, 140 pp.
Methodology and Proposed Procedures
Once the proposed measure has been approved by the ethics committee, the researchers will first take the preliminary measure to a group of Ontario adolescents who participated in a prior online survey with Dr. Dumas and who agreed to be contacted for future research projects. These teens will be sent an online questionnaire via e-mail. The questionnaire will be completed using the survey software, Qualtrics. In addition to the preliminary measure (see attached), the participants will also complete an open-ended question asking them to define bullying. They will also be asked specific questions about our bullying measure, namely: (1) what questions do you think do a good job measuring bullying behaviour? (2) what questions do not do a good job of measuring bullying behaviour? Why? (3) What questions need to be reworded so they make more sense to teens? How? and (3) Are there any topics or questions that are missing from this questionnaire? What are they?
Data from participants will be taken into consideration for a revised version of the bullying questionnaire. Questions will be added or omitted based on the comments of participants. Next, the researchers will compile a questionnaire for online participation via the survey software, Qualtrics. This survey will ask participants to complete the bullying measure, as well as a version of the HEXACO personality inventory, the Basic Empathy Scale, and a bullying scale—this is because an accurate bullying assessment should also correlate with certain personality traits. Data will also be collected on age, gender, and ethnicity. The resulting data will be analyzed.
Data Analysis
For the first part of the study, data will be analyzed using qualitative methods. The researchers will identify common themes in participants’ responses (e.g., questions to be removed or language to be altered) and this will be used to edit and improve the bullying questionnaire.
For the second part of the study, data will be analyzed primarily using exploratory factor analysis and correlations. We will conduct an exploratory factor analysis to identify potential subscales of our bullying measure (e.g., cyber, physical). To examine reliability of our bullying questionnaire, we will examine item-total correlations, the internal consistency coefficient (alpha) and split-half reliability. Additionally, we will run correlations with other measures in the online survey to assess convergent and discriminant validity. Finally, we will examine differences on our bullying questionnaire across gender, age and ethnicity. These tests will be performed using SPSS.
