Interviewing strategy and qualitative sampling

As an interviewing strategy, open-ended questions leave the responses open to participant experience and categories and don’t close down the discussion or allow the participant to answer the question in one word (Snow et al, 2005).  Though in the past it was rejected because it did not involve a precise measurement, sometimes data that may not be easily measurable or counted, have value because of its intrinsic complexity and showcase of the “conditional nature of reality” (Rubin, 2012).  A whole field of text-analytics is aiming to prove that this data, considered as unstructured data, is an important part of knowledge discovery and knowledge sharing. Thus, Rubin (2012) says that open-ended questions grant the participant the chance to respond to the question in any way they choose, as elaborated on a response, allow participants to raise issues that are important to them, or even raise new issues not thought of by the interviewer.  Creswell (2013), further states that the more open the questions the better because it will allow the interviewer to listen to what people say and how they say, which can allow the participants to share their own views.  Usually, there are a few open-ended questions.  Finally, open-ended questions are used primarily in qualitative studies, but a mixture of both close-ended and open-ended questions could be asked in mixed methods studies.

One thing is to have the right questions as part of your interviewing strategies, it is another thing to have the right qualitative sampling plan.

Sampling Plans {purposeful/judgmental sampling, maximum variation sampling, sampling extreme or deviant cases, theoretical sampling, snowball/chain-referral sampling, cluster sampling, single-stage sampling, random sampling} (Creswell, 2013, Rubin, 2012, & Lofland et al, 2005). Here are just three of the many sampling plans listed in the sampling plan space.

  • Purposeful/judgmental sampling: In order to learn about a selective character, group, or category or their variations, you group the population into different characters, groups, or categories to collect data from with the participants now representing those divisions. (Creswell, 2013 & Lofland et al, 2005)
  • Maximum variation sampling: Allows for an analysis of error and bias in a phenomenon, through sampling and discovering the widest range of diversity in the phenomena of interest. (Lofland et al, 2005)
  • Snowball/chain-referral sampling: Asking your initial set of contacts with characteristics X, if they can refer to you their network that has the same characteristics X that you are studying. This is a means to enlarge your sample size and break down barriers to the entrance of your future participant. (Lofland et al, 2005). Depending on the characteristic X, like domestic violence, sexual assault, etc., this technique may run into IRB issues (Rubin, 2012).  Rubin (2012), stated that the way to avoid IRB issues if you have the current participants contact the future participants on your behalf to participate in the interview process, but this can drastically reduce the number maximum number of participants you could have gotten.

References:

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: