Beyond the Qualitative Interview

7/30/2019 Beyond the Qualitative Interview

1/13

Eleanor McLellan, Kathleen McQueen, JudithNeidig

Ysabel Carlos


2/13

The inapproriate or inadequate preparation of transcripts from audioor digital recordings can delay or negatively affect the analysisprocess. (MacQueen and Milstein 1999)

This article focuses on the preparation if transcribed, audiotaped,structured, unstructured, or semistructured interviews for QDA.

Mergenthaler and Stinsons 7 Principles for developing transcription

rules:1. Preserve the morphologic naturalness of transcription

2. Preserve the naturalness of the transcript structure

3. The transcript should be an exact production

4. The transciption rules should be universal

5. The transciption rules should be complete

6. The transcription rules should be independent

7. The transciption rules should be intellectually elegant

Beyond the Qualitative Interview: DataPreparation and Transcription


3/13

Researchers must decide whether their analysis is best supported by transcription or

by researchers notes derived from or supplemented by a review of the audiotapes.

(Patton 2002:380-84)

Essentially researchers undertake their first data reduction step when they decide

what will be transcribed and what will be left out. (Miles andHuberman 1994)

For example, a researcher must make choices whether a textual document should

include non-linguistic observations, be transcribed verbatim and identify speech

patterns, vernacular expressions, intonations, or emotions.

In this transcription protocol, an audiotape should be transcribed in its entirety andprovide a verbatim account of the interview. (Including elisions, mispronunciations,

slang, grammatical errors, nonverbal sounds)

Assist transcribers by specifying what you want included, to ensure that transcripts are

prepared in a standardized manner and is consistently prepared.

DATA PREPARATION



4/13


For ease of readability, transcripts are formatted identically and support either manual or

computer-assisted coding:

Arial, size 10, 1-inch margins (top, bottom, right left), left-justified text

**For some analyses, it may not be necessary to transcribe the whole interview. Only

those related to the research question or theory.

The level of transcription should complement the level of the analysis. (Drisko, 1997)

What to include should always be driven by the research question that an analysis

attempts to answer.

Always identify the interviewer and interviewee. Assign a speaker label (i.e., source ID) to

further assist in navigating through hard copy or electronic versions of the transcript.


5/13


TEXT FORMATTING

1. Arial, size 102. One-inch margins

3. All text shall begin at the left hand margin (no indents)

4. Entire document should be justified

LABELLING FOR INDIVIDUAL INTERVIEW TRANSCRIPTS

Example:

Participant ID:

Interview Name:

Interviewee Category/Subgroup:

Site/Location:

Date of Interview:

Interviewer ID:Transcriber:

LABELLING FOR FOCUS GROUPTRANSCRIPTS

Site:

#Participants:

Focus Group Sample: (e.g., women, men)

Focus Group Interview No.Date of Interview:

Facilitator ID:

Recorder ID:

Transcriber:


6/13


The transcribers should insert a single blank line between the file labelling information and

the actual interview transcription. A double pound sign ## shall precede and follow each

participant identification label (i.e., source ID)

A single hard return shall be inserted after the Source ID. The individualsresponse/comment shall begin on the next line.

SOUR

CE ID


7/13


Audiotape changes

Transcriber should inidicate when the interview is recorded on a new tape and include info verifying thatthe second side of the tape is blank as well as the total number of audiotapes associated with theinterview.

Type in UPPERCASE LETTERS.

End of interview

Type END OF INTERVIEW in uppercase letters on the last line of the transcript along with info

regarding the total number of audiotapes associated with the interview and verification that the

second side of the type is blank. A double space should precede this information.


8/13


SOURCE LABELLING

INDIVIDUAL INTERVIEWS

Source IDs shall begin with the alpha character that designates data collection site/location followed by theindividuals three-digit identification number

Example; F100 = Fenway interviewee #100

Interviewer IDs shall begin with alpha character Ifollowed by the location and interviewers three-digit identification number

Example; IF002 = Fenway Interviewer #002

GROUP INTERVIEWS

Group participants and facilitators shall be assigned a unique source ID.

Example;R500 = Rhode Island focus group participant #500

Group facilitator source IDs must begin with alpha character F followed by a three-digit identification number

Example; FR101 = Rhode Island focus group facilitator #101

Focus group recorder (notetaker)s alpha character shall begin with R

Example;RR002 = Rhode Island focus group recorder #002


9/13


content

Audiotapes shall be transcribed verbatim (i.e. Recorded word for word) including nonverbalbackground sounds (e.g. Laughter, sighs, coughs, claps)

Nonverbal sounds shall be typed in parentheses, for example, (short sharp laugh), (group laughter),

(siren in the background)

If theres mispronunciation of words, these shall be transcribed as how the individual said them.

Transcript should not be cleaned up by removing foul language, slang, grammatical errors, or

misuse of words or concepts.

If the incorrect word results in difficulties in comprehension of text, type the correct word within square

brackets.

Example: I thought that was pretty pacific [/specific/], but they disagreed.

Filler words such as uh, hm, huh, mm, mhm, yeah, yuhuh, huh, ugh shall be transcribed.

Word or phrase repetitions shall be transcribed.

If a word is cut off, a hyphen shall be inserted at the end of the last letter or audible sound.

Example: He wen- he went and did what I told he shouldnve.


10/13


INAUDIBLE INFORMATION The transcribers shall identify portions of the audiotape that are inaudible or difficult to decipher.

If the inaudible part is relatively small, the transcriber shall call this inaudible segment. This

information shall appear in square brackets.

Example: The process of indentifying missing words in an audiotaped interview of poor quality is

[inaudible segment]/

If the part is lengthy, or there is dead air, transcriber should record this information in square

brackets plus the time estimate for information that could not be transcribed.

Example: [Inaudible: 2 minutes of interview missing]

OVERLAPPING SPEECH When individuals are speaking at the same time and it is not possible to distinguish what each

person is saying. Place crosstalk in square brackets after the last identifiable text and pick up

with the next audible speaker.

Example: People may contribute to the conversation, making it difficult to differentiate between one

persons statement [cross talk]. Resulting in loss of information.


11/13


12/13


STORAGE OF AUDIOTAPES

If not actively being transcribed, secure in a locked cabinet.

REVIEWING FOR ACCURACYCheck all transcriptions against the audiotape and revise the transcript accordingly. Listen to each tape against

the transcript three times before it is submitted.

SAVING TRANSCRIPTS

Save each file as an individual MS-DOS ASCII text file with a .txt extension or a rich text file with an .rtf

extension.

Individual interview transcript files shall be assigned the interview name followed by the participant ID (e.g.

VaxC071.txt = Vaccine Interview for UIC participant #071)

For focus groups, the second character shall be a number designating the focus group number for thesite/location. Remaining characters shall designate the sample population. (e.g. CIWOMEN.RTF = UIC

focus group #1, women)


13/13


BACKUP TRANSCRIPT FILES

Back up files on a diskette or CD. Store in the same location as theaudiotapes.

Documents

Beyond the Qualitative Interview