In this phase, the Safety 7 youth researchers developed hands-on skills in cleaning, organizing, and analyzing qualitative data. They gained experience working with transcripts from interviews and focus groups, creating a codebook, applying codes using Dedoose, and identifying themes through collaborative analysis. This section describes what the team did and how these activities were structured to support learning and skill development.
Preparing the data
The qualitative data came from 31 interviews (15 youth, 16 adult family and community members) and 16 focus groups with a total of 61 participants (40 youth and 21 adult family and community members). All sessions were audio recorded and professionally transcribed.
Before beginning analysis, each transcript was reviewed for accuracy by comparing it against the original audio. Then, all transcripts were carefully de-identified to protect participant privacy. Personally identifiable information, such as names, schools, street names, parks, businesses, transit stops, and dates, was redacted and replaced with [MASKED] text.
Youth researchers were trained to complete this process through a hands-on data cleaning activity. Using printed transcripts and highlighters, they practiced identifying and masking sensitive information. This training emphasized the importance of confidentiality in research and introduced youth to the practice of reviewing raw data—a foundational skill in qualitative analysis.
Example facilitation guides, activities, and other resources
- Data Cleaning Presentation – This presentation offers guidance for cleaning qualitative data and includes an interactive activity designed to help youth researchers practice cleaning their own transcripts.
Introduction to Dedoose
To support qualitative coding and analysis, the team used Dedoose, a web-based platform designed for mixed-methods research. Youth were given step-by-step instructions for setting up accounts and accessing the project workspace. Training sessions guided them through how to upload and open transcripts, apply codes to excerpts of text, and use memos to track observations and questions.
Each youth had the opportunity to practice in Dedoose with sample transcripts. They learned how to highlight text, apply and create codes, group codes into categories, and edit or merge codes as needed. Youth also learned to distinguish between a “code” (a brief label that describes the content of a passage) and a “memo” (a note capturing thoughts, questions, or patterns). This hands-on experience gave youth a practical understanding of how software tools can support qualitative research.
Example facilitation guides, activities, and other resources
- Guide for Understanding and Coding Qualitative Data – This guide offers an introduction to qualitative research and provides practical instructions for using Dedoose to analyze data.
Building the codebook
The Safety 7 played a central role in creating the codebook that guided the analysis. The process began with a set of deductive codes developed from the research protocols and early impressions of the data. These initial codes reflected key topics the team expected to find, such as community safety, experiences with violence, trusted adults, and safe spaces.
Youth researchers then reviewed sample excerpts and proposed new codes based on what they saw in the data. These inductive codes captured ideas or patterns that emerged organically, ensuring the codebook reflected real participant experiences. Adult researchers facilitated discussions to clarify code meanings and organize related codes into broader categories.
Through this process, youth learned how to translate qualitative insights into systematic labels, how to define codes clearly, and how to identify when multiple excerpts were expressing the same underlying idea. Their participation directly shaped the final structure and language of the codebook.
Example facilitation guides, activities, and other resources
- Qualitative Analysis Codebook – This document includes the final version of the codebook.
Coding and review
With the finalized codebook in place, a team of four Child Trends’ researchers coded the full set of transcripts. Every transcript was double-coded to ensure consistency. For each file, two researchers independently applied codes, then met to review and reconcile any differences. This step ensured reliability across coders and strengthened the validity of the findings.
Due to the volume of data and limited availability stemming from other commitments, such as school and other employment, youth researchers could not code every transcript. However, their perspectives remained central and continued to inform and guide the coding process throughout. Throughout coding, Child Trends’ researchers noted new ideas or discrepancies using memos, which were then reviewed during team discussions. Youth were also consulted to review how certain codes were being applied and to determine whether any updates to the codebook were needed.
This approach modeled rigorous qualitative practices, such as maintaining inter-rater reliability, using memos to track analytic decisions, and refining code definitions in real time.
The Safety 7 includes seven youth researchers (Zionnah Garnett, Elijah Jones, London Mclean, Jermese Perkins, Ife Tobechi, Ike Tobechi, and Meagan Tutt) who make all key research decisions in partnership with staff from Child Trends and Sasha Bruce Youthwork. The Safety 7 is supported by Quiana Lewis Wallace and Deja Logan from Child Trends, Courtney Gibbs, and Bianca Faccio from Sasha Bruce Youthwork. Principal investigators for the study are Samantha Holquist and Quiana Lewis Wallace (Child Trends) with support from Kelly Murphy. Jennifer Widstrand (Child Trends) served as the Project Director. Additional contributions came from Ja’Chelle Ball, Kenya Downing, Jessica Conway, Olivia Reyes, and Alyssa Scott (Child Trends), and Deborah Shore, Donnell Potts, Ivana Gutierrez, and Jane McGhee (Sasha Bruce Youthwork).
If you have questions about the Centering Youth Experiences to Address Community Violence project, please contact Principal Investigator Quiana Lewis Wallace at qlewiswallace@childtrends.org.
This project was supported by Award No. 15PNIJ-22-GG-01420-RESS, awarded by the National Institute of Justice, Office of Justice Programs, U.S. Department of Justice. The opinions, findings, and conclusions or recommendations expressed in this publication/program/exhibition are those of the author(s) and do not necessarily reflect those of the Department of Justice.