In dissertation and thesis qualitative methodologies, such as appreciative inquiry, case study, delphi method, grounded theory, phenomenology, and others, you can establish or improve the validity of your findings/results if you use one of various forms of triangulation.
Here is an excellent definition of “triangulation” from Thomas A. Schwandt in The Sage Dictionary of Qualitative Inquiry (2007): “Triangulation is a means of checking the integrity of the inferences one draws. It can involve the use of multiple data sources, multiple investigators, multiple theoretical perspectives, and/or multiple methods.” (p. 298) As an example, the multiple data sources could be responses to interview questions and responses to survey questions. These would be compared and contrasted. A study could include two or more evaluators, who would independently score answers to open-ended survey questions. Another option is for the principle investigator to code the transcripts of interviews and have an outside evaluator/external auditor validate the identified meaning units and assigned themes. Schwandt continues: “The strategy of triangulation is often wedded to the assumption that data from different sources or methods must necessarily converge or be aggregated to reveal the truth.” (p. 298) I appreciate that Schwandt places great importance on discovering the truth in research studies. It has been my experience that some dissertation students do not care about the truth and are willing fabricate their findings based on their own biases. These individuals have little integrity and are not interested in triangulation. Without an audit trail of the coding (analysis) procedures, there can be no confidence that the findings are valid. This is why triangulation is so important.

Schwandt’s definition is the most comprehensive description of triangulation available. It easily supports the following three-step approach I suggest to my clients, who conduct interviews in their appreciative inquiry, case study, delphi method, grounded theory, phenomenological, and other qualitative designs.

Step One – Interviews are conducted and audio-recorded. Note-taking is strongly discouraged during the interview process since it distracts the interviewer from what the participant is saying and contributes to missed opportunities for more probing questions for clarification. Three common issues that may need more description are the participants’ use of empty or fluff words like “tough.” What does tough mean? You will want to ask for more information and/or description. When they say “tough,” exactly what do they mean? A second example comes up from time to time. A participant may be somewhat vague, such as “That was no fun.” This comment has little real meaning to you as a researcher. When you hear these types of comments, you will want to interrupt them and ask for the statement or phrase in question to be described in more detail. A third issue is when the interviewee refers to second-hand information, such as “The teacher in room 4 said that she heard a parent say that ‘all the children are doing fine.'” Here you would want to remind the participant that you only need to know their personal perceptions, thoughts, and experiences because you are researching what they think and feel.

Step Two – Member Checks: Audio-recordings are transcribed and the transcriptions returned to participants for their review, editing (if necessary), and approval.
This is what W. Paul Vogt in the Dictionary of Statistics and Methodology: A Nontechnical Guide for the Social Sciences (2005) calls “member check (or validation).” His definition of member check is “The practice of researchers submitting their data or findings to their informants (members) in order to make sure they correctly represented what their informants told them. This is perhaps most often done with data, such as interview summaries; it is less often done with interpretations built on those data.” (p. 190-191)

Step Three – Researcher collaborates with an “outside evaluator” or “external auditor” during data analysis (coding).
This individual has three main responsibilities: (1) confirming the researcher’s accurate identification of meaning units in the transcripts; (2) validating the conversion of meaning units into appropriate idiographic themes, and (3) assessing the accuracy of nomothetic themes derived from recurring idiographic themes. Schwandt (2007) states, “This is a procedure (of auditing) whereby an independent, third-party examiner systematically reviews an audit trail maintained by the inquirer…. An audit trail is a systematically maintained documentation system.” (p.12)
You will find that, with the inclusion of this form of triangulation in your data analysis procedures, your Chair will appreciate your efforts to use a validation strategy. This process lessens the concerns many Chairpersons have over possible subjectivity or discovering what the researcher already knew or wanted to find. Triangulation contributes to objectivity and validates that your findings are true. My clients, who used this approach in their dissertations and theses, have never had the authenticity of their results questioned by their committee.
Key Words: audit trail, coding, external auditor, idiographic, member checks, outside evaluator, nomothetic, qualitative methodologies, triangulation.

 

William G. Wargo, Ph.D.
© 2016

 

References

Schwandt, Thomas A. (2007). The Sage Dictionary of Qualitative Inquiry. 3rd ed. Thousand Oaks, CA: Sage Publications

Vogt, W. Paul. (2005). Dictionary of Statistics & Methodology: A Nontechnical Guide for the Social Sciences. (3rd ed.). Thousand Oaks, CA:
Sage Publications.

Recommended citation for this article:

Wargo, W. G. (2016). Triangulation to Establish Validity of Qualitative Data. Menifee, CA: Academic Information Center.

*For additional information, please contact me at wgwargo@academicinfocenter.com or (951) 301-5557.