PU565 Unit 8 Codebook Excerpt Example

One Definition Rewritten Three Times Until Two Coders Agreed: A Unit 8 Codebook Excerpt Showing Five Codes With Their Inclusion and Exclusion Rules, Boundary Cases and the Agreement Each Version Produced

[Student Name]

Purdue University Global

PU565: Qualitative and Quantitative Methods

Unit 8 Codebook Excerpt

[Instructor Name]

[Date]

The transcripts, the coders and the agreement figures are composite; the codebook structure and the research cited are real.

Code 1: Price as a Closing Move

Version one: passages where the participant gives cost as the reason for not buying. Two coders, six transcripts, 41 flagged passages, agreement on 27. The disagreements were all of one kind: one coder flagged every mention of a dollar figure, the other only those where the figure ended the discussion. Version two: passages where cost is given as the reason and the participant does not then discuss any way around the cost. Agreement rose, and a new disagreement appeared where participants mentioned a price and then changed the subject, which one coder read as closing and the other as ordinary conversational drift. Version three, current: Apply where the participant names a cost or affordability and the passage contains no subsequent consideration of financing, saving, insurance, a cheaper model or help from family, anywhere in the transcript. Do not apply where cost appears inside a comparison of models, or where the participant raises cost and then asks the interviewer a question about it. Agreement held at 38 of 41 passages, and the three remaining disagreements are listed in the codebook as unresolved rather than averaged away. Boundary case: I saw them, 300 dollars, and my daughter said she would pay half, and I still said no. Coded as not price, since the financing appears and the refusal survives it. This is the case that produced version three. A code that two people apply differently is not a measurement problem to be reported as a coefficient; it is a definition that has not been written yet.

Code 2: Not Yet

Apply where the participant places action in the future without naming a condition that would trigger it. Do not apply where a condition is named, however vague, such as when I retire or if it gets worse at church, which belong to code 3. Boundary case: maybe next year is coded not yet, since a date is not a condition. Agreement 34 of 36.

Code 3: Named Trigger

Apply where the participant names an event, a threshold or a person that would move them to act. Includes clinical thresholds, social occasions and another person's insistence. Do not apply to statements about price bands, which belong to code 1. Boundary case: when I can't hear my granddaughter is coded named trigger, since the event is specifiable even though it is not measurable. Agreement 31 of 33.

Code 4: Managing the Room

Apply where the participant describes altering seating, position, venue, volume or company to hear better without disclosing hearing difficulty. Includes pretending to have heard, nodding along and changing the subject, which were separate codes until the second reading showed them doing one job. Do not apply where the participant asks for repetition openly, which belongs to code 5. Boundary case: I sit at the end so I can turn my good ear in is coded managing the room. Agreement 44 of 47.

Code 5: Asking and Its Cost

Apply where the participant describes requesting repetition, or avoiding the request, and attaches a social consequence to it. The consequence is what distinguishes this from a simple account of asking. Do not apply to concealment without a request, which is code 4. Boundary case: I ask twice and then I let it go is coded here, because the giving up carries the consequence. Agreement 29 of 34, the lowest of the five, and the codebook records that the remaining gap is about whether irritation counts as a social consequence.

How Agreement Was Calculated and What It Means

Agreement is reported as passages agreed over passages flagged by either coder, which is a proportion and not a chance-corrected statistic. A chance-corrected coefficient was calculated for code 1 and is reported alongside, but the codebook treats the proportion as the working number during definition development, since the purpose at this stage is to improve the definition rather than to certify it (McHugh, 2012).

Rules That Apply to Every Code

1. Code the passage, not the participant; one speaker may carry several codes and contradict themselves. Nothing in this excerpt was imported from an outside framework; each label grew out of the transcripts themselves, so the definitions are allowed to change as later interviews arrive (Hsieh & Shannon, 2005).

2. A passage may take two codes where it does two things, and the codebook names the three pairs that legitimately co-occur.

3. Any new code added after this version carries the date, the transcript that prompted it, and the initials of the coder who proposed it.

4. Disagreements are recorded and resolved by discussion, with the resolution written into the definition rather than into a coder's memory, which is the practice the thematic approach this study follows asks for (Braun & Clarke, 2006).

References

Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77-101. https://doi.org/10.1191/1478088706qp063oa

Hsieh, H. F., & Shannon, S. E. (2005). Three approaches to qualitative content analysis. Qualitative Health Research, 15(9), 1277-1288. https://doi.org/10.1177/1049732305276687

McHugh, M. L. (2012). Interrater reliability: The kappa statistic. Biochemia Medica, 22(3), 276-282. https://doi.org/10.11613/BM.2012.031