Qualitative Data Analysis Help for Thematic, Content and Grounded Theory Research

What qualitative data analysis help covers ¶

The most common problem I see is a method section that names one approach and a findings section that follows another.

Qualitative data analysis is the systematic process of organizing, coding and interpreting non-numerical data, such as interview transcripts, focus groups, open-ended survey answers and documents, to produce themes, categories or theory. Our qualitative data analysis help is consulting: we work alongside you on approach, coding, software and write-up, and you remain the analyst.

Most researchers come to us at one of four points:

• Before coding: choosing between thematic analysis, content analysis, grounded theory or framework analysis, and matching the choice to your research question and epistemology.
• During coding: a codebook that keeps growing, codes that overlap, or themes that read like topic summaries.
• In the software: setting up an NVivo, ATLAS.ti or MAXQDA project so cases, attributes and queries actually answer your questions.
• At write-up: turning coded data into a findings chapter or article that meets reviewer expectations.

If your study mixes interviews with survey data, the quantitative strand sits with our statistics help team, and we coordinate the two.

Thematic analysis: Braun and Clarke’s six phases ¶

Braun and Clarke’s thematic analysis is the most widely used approach in qualitative research, and also the most often misapplied. We review your process phase by phase.

Their later work renames some phases (for example, “generating initial themes”) and stresses that the process is recursive: you move back and forth between phases.

What is reflexive thematic analysis?

Reflexive thematic analysis is Braun and Clarke’s own position: themes are actively generated by the researcher, and the researcher’s subjectivity is an analytic resource to be reflected on. It sits beside two other families, coding reliability approaches and codebook approaches such as template and framework analysis. Mixing their assumptions, for example reporting intercoder reliability for a reflexive analysis, is a frequent reviewer objection.

Content analysis, grounded theory and codebook development ¶

Agree the unit of coding before calculating kappa. Most poor agreement scores trace back to how segments were cut.

Content analysis

Content analysis is a research method for systematically categorizing text, and sometimes counting categories, to describe patterns in communication. Hsieh and Shannon (2005) distinguish conventional (categories derived from data), directed (categories from prior theory) and summative (counting and interpreting keywords) approaches. We help you define the unit of analysis, build the coding frame and pilot it.

Grounded theory

We support classic (Glaserian), Straussian and constructivist (Charmaz) grounded theory. Each uses different coding stages: open, axial and selective coding for Strauss and Corbin; initial, focused and theoretical coding for Charmaz. We check that constant comparison, memo writing and theoretical sampling are visible in your audit trail, and that “saturation” is justified rather than simply asserted.

Codebook development and intercoder reliability

A usable codebook entry has a code name, a definition, inclusion and exclusion criteria, and a typical example extract. Where your approach calls for reliability testing, we help you choose the statistic: Cohen’s kappa for two coders, Krippendorff’s alpha for more coders or missing codes. Percentage agreement alone ignores chance agreement and rarely satisfies reviewers.

NVivo help, ATLAS.ti help and MAXQDA support ¶

We work in all three major packages, including NVivo 15. The concepts are similar; the names and menus differ.

Typical requests include restructuring a flat list of 300 codes into a hierarchy, linking transcripts to cases so you can compare groups, and exporting a codebook or matrix for an appendix. If your recordings still need typing up, our research transcription service formats transcripts for direct import into NVivo or ATLAS.ti.

How qualitative analysis consulting works ¶

1. Send your brief. Your research question, methodology section, a sample of anonymized transcripts and, if you have one, your project file or codebook.
2. We match an analyst to your field. Health, education, management and social science studies each have their own reviewer expectations.
3. Review and feedback. We return written feedback on your coding and documents with Track Changes and margin comments that explain the reasoning, or work through your project with you in live sessions.
4. Follow-up. You can ask your analyst questions about their comments after delivery, so you can apply the feedback to the rest of your dataset.

Turnaround runs from 24 hours to 7 days depending on the volume of material.

Reporting qualitative findings ¶

Strong qualitative reporting interprets the extracts it quotes. We review your findings and method sections against the standard your journal or committee expects: COREQ (32 items, for interviews and focus groups), SRQR (21 items) or APA’s JARS-Qual.

• Each theme opens with its central claim, then uses extracts as evidence.
• Extracts carry consistent participant identifiers, such as P07 or FG2-P3.
• Your analytic approach, reflexivity and quality procedures are described in enough detail to be evaluated.
• Tables or thematic maps match the text exactly.

For language and flow once the analysis is settled, see our dissertation editing service, or read our guide to thematic coding.

Your analysis, your authorship ¶

You keep full authorship. We explain, check and coach; we do not code your dataset for you to submit as your own work. Universities differ on how much analytic support is allowed for theses, so check with your supervisor and read our academic integrity policy before you book.