This study was conducted using a qualitative descriptive approach (Sandelowski, 2000). A focus group strategy was used to gain a comprehensive description of youths' perceptions of the roles that health care providers should play in addressing youth bullying. Because bullying occurs within a social context, the focus group methodology was determined to be well aligned with exploring this phenomenon and has previously been shown to be highly effective for this purpose (Horowitz et al., 2004, Krueger and Casey, 2009). The study was approved by the institutional review boards at Boston College, Boston Children's Hospital, and Maine Medical Center.
Study Setting and Recruitment
Purposive sampling was used to select participants who could provide in-depth information about the phenomenon of interest. Youths in middle or junior high school from diverse racial, ethnic, socioeconomic, and geographic backgrounds were recruited from two states through fliers posted in the clinic areas, announcements on the hospital Web page, or nomination by clinic staff.
For youths expressing interest in participating in the focus groups via telephone or e-mail, eligibility initially was confirmed through a telephone discussion with their parent(s). Inclusion criteria required that the youths were between 10 and 16 years of age, fluent in spoken English, willing to share their expertise and experiences in a focus group, and had participated in school or community bullying prevention educational programs. Youths with significant cognitive or behavioral impairments resulting in disabling language comprehension or expressive problems as determined through the parental interview (e.g., use of Individualized Educational Plans, parental report, etc.) were excluded because they would be unable to participate in group focus group activities. Data were collected from September through November 2015.
Two measures, a demographic form and moderator guide, were used during the study. The investigator-created demographic form collected key information about the study participants. The moderator guide was developed based on group process principles and the study's purpose and objectives (Stewart & Shamdasani, 2014). Semistructured interview questions and probes were designed in accordance with the study's overall aims and from information culled from published multidisciplinary research findings and were informed by bullying theory and expert clinical opinion. Example questions can be found in the Box.
An evidence-based approach for conducting successful youth focus groups on health-related topics was used in the focus group design (Horowitz et al., 2003, Morgan et al., 2002). The groups, 1.5 hours in length, took place in private conference rooms within regional health care facilities. Before the start of the focus group, parental consent and child assent were obtained. The participants were then asked to complete the demographic form, were provided with snacks, and engaged in facilitated informal conversation. The focus groups were led by experienced moderators and were conducted in a semistructured manner because this approach allowed for the collection of data from both individuals and the individual as part of a larger group. The moderator initially began by introducing herself and the recorder and by orienting the participants to the group's purpose and ground rules. Participant introductions followed. Participants were told that they did not have to answer any questions with which they were uncomfortable. Members of the research team served as recorders who audiotaped the sessions and took field notes.
During the focus groups, questions were explained, elaborated, redirected, and introduced as necessary to encourage dialogue while addressing the concepts of interest. Open-ended questions for each of the major constructs initially were explored, moving from concrete to more abstract constructs. Follow-up probes were used to clarify or gain additional information on topic areas that emerged from the initial responses to the open-ended questions. After participants had an opportunity to offer their personal views, the moderator shifted to a more directive style of questioning by asking participants what they thought about specific topics. Congruent with focus group methodology, no push toward ensuring conformity or consensus was made, because the aim was to derive as comprehensive a picture of the participants' views as possible and to ensure that there was opportunity for contrary opinions to arise (Krueger & Casey, 2009). As we will describe, data analysis began as soon as the first focus group transcript was available, and additional focus groups were held until saturation was reached and no new categories emerged from the data.
Descriptive statistics were used to describe the characteristics of the study population. For the qualitative analysis, focus group audiotapes were transcribed verbatim at the conclusion of each session by a professional Collaborate Institutional Training Initiative-certified transcriptionist. The accuracy of the transcription was ensured by having the moderator and recorder listen to the tape and compare it with the transcript text. The transcripts were entered into Hyper-RESEARCH 3.03 (Researchware, Inc., 2011), a qualitative software computer program, to facilitate the analysis.
Field notes about the environment and participant observation, context, and tone were added to facilitate interpretation.
The primary data analytic technique used was conventional qualitative content analysis (Krippendorff, 2013). To be consistent with an inductive approach to generate codes, the initial goal was to extract those verbatim text statements that could help explain the youths' views of bullying and their health care providers' role in addressing it. There were no preconceived codes, but the data analysis was guided by the research purpose.
Three members of the research team, working independently, read each of the transcripts in its entirety to obtain a general sense of participants' perceptions of their nurse practitioner's or physician's role in assessing bullying. They then re-read the transcripts, and word-by-word and line-by-line analyses were conducted. In vivo codes, consisting of significant words, phrases, and statements, were identified, and similar codes were clustered together into categories.
After first-level coding, the referential adequacy of the findings was determined. This was done by having the research team members compare their preliminary codes against the raw data. In keeping with content analysis processes, the groups and topics were reviewed multiple times by the research team to identify key themes existing within and across the focus groups. Areas of disagreement were resolved through discussion, further transcript review, and consultation with focus group participants. Lastly, coding redundancies were removed. A final set of themes was identified through the process of constant comparison.
An iterative, deductive, analytic process then was used to aid in data interpretation while fully illuminating the emergent themes (Sandelowski, 2000). First, categories were developed from codes; broader themes then were developed from the categories capturing the essence of youths' perceptions of their health care providers' role in addressing bullying (DeSantis & Ugarriza, 2000). Care was taken to ensure that the new categories and themes described related, but differing, phenomena at each level of abstraction (Guba & Lincoln, 1985). At each stage of the process, the team discussed discrepancies in coding and conducted further transcript review to reach consensus. Data analysis was completed when no further information emerged from the transcripts.
Maintenance of Rigor
Strategies to maintain rigor were be used throughout the study. Rigor in qualitative research is evaluated by credibility, fittingness, and auditability (Morse et al., 2002, Sandelowski, 1986). During the focus groups, the moderator attempted to make participants feel comfortable to ensure that participants spoke freely about their experiences. The emergence of data was driven by the participants; this was achieved by allowing participants to express what was important to them rather than leading them with notions of what the study investigators assumed would be important. To ensure credibility in this study, selected focus group participants reviewed the findings and verified that they represented their thoughts, feelings, and statements. They validated that youths who experienced bullying would immediately recognize the information that was been reported. Purposive sampling, where the participants were representative of youths at risk for bullying exposure, helped ensure that the findings would be seen as meaningful to the audience or readers (fittingness). For example, participants were selected from the middle through high school age groups, when youth bullying is known to peak. In addition, flyers were placed in settings known to include a diverse range of socioeconomic statuses and racial and ethnic backgrounds. Auditability was achieved by having the research team ensure the accuracy of the focus group transcripts, keep clear notes, and track all of the decisions regarding data analysis so that others could replicate the study and arrive at the same conclusion based on the original data and decision trail.
J Pediatr Health Care. 2017;31(5):536-545. © 2017 Mosby, Inc.