Stratified Sample
An alternative to random and systematic sampling is stratified sampling which requires dividing the population into sub-groups before sampling. These sub-groups are based on certain characteristics relevant to your questionnaire (e.g., age, income, education, or gender) and each is referred to as a stratum. The collective group is the strata.
Stratifying continues by dividing each stratum into more sub-groups until the strata accurately reflect the population. Random samples are then drawn from each stratum. Each sample size should reflect the stratum's size in the overall population
Qualitative Research
Sampling
Qualitative research does not generate statistical information and most of the concepts of statistically valid data do not apply; therefore, quantitative sampling issues do not apply to qualitative research. However, as the results of qualitative research are typically generalized to a population of customers or employees, a customer/employee profile is typically developed to help ensure the validity of these generalizations.
The first step is to categorize the population of people about which you want to be able to generalize the research results. For example, customers who purchase specific types of products, live in specific locations, and purchase a minimum dollar amount of products on an annual basis. This profile will be used to ensure that the data collected is as representative as possible. As with any sampling discussion, sample size increases with the complexity of the population being studied.
The next step in sampling is to develop a screening routine that will identify appropriate people from each category. This screening routine can be used to control the mix of participants in a focus group or to fill interviewing quotas.
Group Discussions
Focus Groups
A focus group is a highly structured meeting with three main elements: it is led by a trained and impartial moderator; discussion is based on specific issues pre-determined by the client; and, optimally, the group is comprised of eight to twelve participants. In such a setting participants develop and expand on each others' comments and ideas.
Benefits
- Some quantitative data collection possible.
- Props and examples can be used to stimulate discussion.
- You have the ability to gather uniform data across many different groups in a variety of locations.
- Regular polling of participants generates easily analysable data and facilitates the comparisons of information gathered across different groups. It also guarantees participation from all people.
Brainstorming
Brainstorming, which is often a component of focus groups, is a rapid-fire generation of ideas from a group of people. This technique is most useful for exploratory data collection, and should be of short duration on a specific topic. For example, a group of customers could be asked why company x receives low scores on satisfaction with technical support.
Benefits
- All ideas are valued and no idea is criticized.
- High energy environment.
- Generation of a wide variety of ideas.
- Can be done in a short period of time.
Individual Discussions
Telephone Interview
A qualitative telephone interview is a very structured discussion on a specific topic. These interviews are used to ensure that data collected from diverse respondents is closely aligned. A good example is a survey pretest--a structured conversation with a potential respondent about a proposed survey form. The pretest discussion can focus on clarity, ease of use, and potential response rates.
Benefits
- A combination of qualitative and quantitative information can be discussed. Props and examples can be faxed or shipped to respondent for discussion.
- A wide variety of people from many different locations can be interviewed economically.
- The interviewer can use many probing questions to clarify customer responses.
- Fast turn around of data and reporting.
- Generally, more economical than focus groups.
Face-to-face Interview
Program stakeholders (such as company executives) are often interviewed in person as part of a Needs Analysis. This structured, face to face conversation is used to develop the goals and objectives of a survey program and to identify specific content areas to include in a survey.
Benefits
- Typically held in a respondents home or office.
- Ideal for reaction to a form or product.
- Extensive probing can be used to collect detailed information.
- Very convenient for the respondent.
- Forms or products can be tested in the environment in which they will be used.
- Respondents body language can guide the interviewer and recorded to help interpret comments.
Other common qualitative techniques include open-end questions on a written survey and mall intercept interviews. Qualitative data can be a rich source of information and should be considered for both survey development and for exploring survey findings.