Weiss Selecting Interview Respondents
Weiss, R. S. (1994). Learning from strangers: The art and method of qualitative interview studies. New York: The Free Press.
Chapter 2: Respondents: Choosing Them and Recruiting Them
Aims and Substantive Frame of the Study
- The substantive frame of the study is the set of topics the study is designed to explore.
- The narrower the substantive frame, the more focus and the easier it is to identify participants, but the more difficult it is to be generalized.
Panels and Samples
- A Panel of Knowledgeable Informants defined: "...people who are uniquely able to be informative because they are expert in an area or were privileged witnesses to an event (p. 17)."
- A Representative Sample defined: "...people who, taken together, display what happens within a population affected by a situation or event (p. 17)."
The Panel of Informants
We use a panel when we want to study:
- "A study of an organization requires that the investigator succeed in obtaining informants without being perceived as an intrusive foreign presence (p. 19)."
- Surviving in the field is dependent upon social grace, self-confidence, institutional awareness, and persistence. Social grace includes sensitivity, considerateness, and tact. Institutional awareness is knowledge about the politics of the organization or institution.
- We use a panel when we want to study:
- A loose collectivity
- A social institution
- A Key Informant
- Defined: "A good person to start with in any study requiring a panel is a knowledgeable insider willing to serve as an informant on informants. But others who might help include a knowledgeable marginal or disaffected figure within the system (p. 20)."
- Begin interviews with these individuals who are easily accessible and/or those individuals who are marginal. Beginning with such individuals allows for a margin of error in that making mistakes with these individuals is less critical than making mistakes with a key informant.
- How Large a Panel?
- You know when to stop when you have reached saturation.
- "The best answer is that you stop when you encounter diminishing returns, when the information you obtain is redundant or peripheral, when what you do learn that is new adds too little to what you already know to justify the time and cost of the interviewing (p. 21)."
- Representational Samples
- Probability Sampling for Qualitative Research
- "if the people who make up a probability sample are chosen in such a way that each choice is independent of every other choice, and the sample includes at least 60 respondents, then the sample is likely to be a fairly good representation of the population (pp. 21-22)."
- A probability sample must be selected randomly.
- Samples That Attempt to Maximize Range
- "Random sampling will provide us with a picture of the population as well as of particular instances, and sampling for range will ensure that our sample includes instances displaying significant variation (p. 23)."
- When determining the sample to maximize range, consider looking for contrast in the following ways:
- Independent variables
- Dependent variables
- Convenience Sampling
- Snowball sampling locates informants through referrals - one referral leads to another, which leads to another and so on. Other ways to find participants include advertising for volunteers and locating a congregating place where individuals whom you want to study would gather.
- Arguments for the Generalizability of the Findings of Convenience Samples
- Respondents' Own Assessments of Generalizability
- Similarity of Dynamics and Constraints
- Theory Independent of Qualifiers
- Corroboration from Other Studies
- An Invalid Argument for Generalization from Convenience Samples
- Convenience samples do not often have enough characteristics to represent the population at large, therefore making generalizability difficult.
- Comparison Cases
- Having comparative cases strengthens the validity and generalizability.
- Conceptually Important Cases
- Unique cases can add to the conceptual understanding of the phenomena.
- An N of 1
- "However, case research can absorb as much data-gathering effort and analytic time as would research based on larger samples. Case research is different primarily because it anchors its potential for generalization in the welter of detail of the single instance (p. 33)."
- Recruiting Respondents