Last month Research Now and SSI, both long-time supporters of the MRII, completed a key step in their merger by rebranding as Dynata. We’re grateful to Dynata for being one of our sponsors; such sponsorships have enabled us to develop our new Principles Express line of courses. And we’re especially grateful to each of the following Dynata staff: Melanie Courtright, president-elect; Jackie Lorch, past chair of the Curriculum Committee; Keith Phillips, author of our Principles Express course Sampling in Market Research; and Pete Cape, author of Quantitative Data Collection Methods. The following is an excerpt from Pete’s text.Jeffrey Henning
The choice of the most appropriate data collection method will be a trade-off between various factors dependent on the project objectives, the client’s demands and constraints, and the real choices available to the researcher in his or her country. The factors can be arranged into three groups.
Task factors related to the objectives of the project:
- Question diversity and flexibility – more complex question types can be undertaken when there is an interviewer present to help explain the task or when the questionnaire is visible to the participant. Questionnaires may become more complex if intricate routing and survey logic is required.
- Physical stimuli – the need for this can range from the most basic (a visual representation) to touch and possibly to taste.
- Sample control – defined as the ability of the method to reach the participants specified by the sample definition. It can be usefully broken down further into:
- Theoretical coverage: is it possible, in theory, to reach all potential participants?
- Sample frames: to what extend do the lists you have available to you reduce the actual coverage?
- Efficiency: can the above two factors be mitigated by greater efficiency, that is, the ease with which qualifying participants can be found?
- Quantity of data – how much information can be gathered will depend on the interest level the participant has in the subject, how engaging the task is and the ease with which the interview can be abandoned once started.
- Response Rate – the majority of non-response is made up of refusal and ‘not-at-home’. Refusal can be reduced by actions taken by an interviewer (persuasion and follow-up/recontact). Not-at-home is reduced by repeated contacts at different times and days.
2. Situational factors related to the practical aspects of the project plus budget and time constraints:
- Environment – the situation the participant is in, both in terms of physical and psychological comfort and how conducive it is for concentrating on the survey and tasks.
- Fieldforce control – how easy it is to monitor and check the work done by interviewers, either in the selection of participants or in administering the survey itself.
- Interviewer bias – the extent to which the interviewer, however unwittingly, conveys their own opinion and thus suggests an answer which the participant is biased to agree with.
- Speed and cost. Speed is normally measured from the moment the survey questionnaire is ready to be used for interviewing to the point where data processing and tabulation can be done.
3. Participant factors related to the participants, who they are and how they will react to the survey.
- Perceived anonymity – the extent to which the participant feels the interviewer or researcher will not know who they are.
- Social desirability bias – the tendency of participants to give answers that are socially acceptable whether or not they are true.
- Sensitive information – surveys may require asking for information that participants consider highly personal, confidential or otherwise sensitive.
- Low incidence rate – how hard it is to find a qualifying participant will affect both the project speed and cost.
- Participant control – the extent to which the participant is in control of the survey taking experience, when it is done and the speed at which it is completed.
For more from Pete, please check out the Principles Express course Quantitative Data Collection Methods.