Nonprobability Sampling

Nonprobability samplingSampling techniques for which a person’s likelihood of being selected for membership in the sample is unknown. refers to sampling techniques for which a person’s (or event’s or researcher’s focus’s) likelihood of being selected for membership in the sample is unknown. Because we don’t know the likelihood of selection, we don’t know with nonprobability samples whether a sample represents a larger population or not. But that’s OK, because representing the population is not the goal with nonprobability samples. That said, the fact that nonprobability samples do not represent a larger population does not mean that they are drawn arbitrarily or without any specific purpose in mind (once again, that would mean committing one of the errors of informal inquiry discussed in Chapter 1 "Introduction"). In the following subsection, “Types of Nonprobability Samples,” we’ll take a closer look at the process of selecting research elementsThe individual unit that is the focus of a researcher’s investigation; possible elements in social science include people, documents, organizations, groups, beliefs, or behaviors. when drawing a nonprobability sample. But first, let’s consider why a researcher might choose to use a nonprobability sample.

So when are nonprobability samples ideal? One instance might be when we’re designing a research project. For example, if we’re conducting survey research, we may want to administer our survey to a few people who seem to resemble the folks we’re interested in studying in order to help work out kinks in the survey. We might also use a nonprobability sample at the early stages of a research project, if we’re conducting a pilot study or some exploratory research. This can be a quick way to gather some initial data and help us get some idea of the lay of the land before conducting a more extensive study. From these examples, we can see that nonprobability samples can be useful for setting up, framing, or beginning research. But it isn’t just early stage research that relies on and benefits from nonprobability sampling techniques.

Researchers also use nonprobability samples in full-blown research projects. These projects are usually qualitative in nature, where the researcher’s goal is in-depth, idiographic understanding rather than more general, nomothetic understanding. Evaluation researchers whose aim is to describe some very specific small group might use nonprobability sampling techniques, for example. Researchers interested in contributing to our theoretical understanding of some phenomenon might also collect data from nonprobability samples. Maren Klawiter (1999)Klawiter, M. (1999). Racing for the cure, walking women, and toxic touring: Mapping cultures of action within the Bay Area terrain of breast cancer. Social Problems, 46, 104–126. relied on a nonprobability sample for her study of the role that culture plays in shaping social change. Klawiter conducted participant observation in three very different breast cancer organizations to understand “the bodily dimensions of cultural production and collective action.” Her intensive study of these three organizations allowed Klawiter to deeply understand each organization’s “culture of action” and, subsequently, to critique and contribute to broader theories of social change and social movement organization. Thus researchers interested in contributing to social theories, by either expanding on them, modifying them, or poking holes in their propositions, may use nonprobability sampling techniques to seek out cases that seem anomalous in order to understand how theories can be improved.

In sum, there are a number and variety of instances in which the use of nonprobability samples makes sense. We’ll examine several specific types of nonprobability samples in the next subsection.

Types of Nonprobability Samples

There are several types of nonprobability samples that researchers use. These include purposive samples, snowball samples, quota samples, and convenience samples. While the latter two strategies may be used by quantitative researchers from time to time, they are more typically employed in qualitative research, and because they are both nonprobability methods, we include them in this section of the chapter.

To draw a purposive sampleA nonprobability sample type for which a researcher seeks out particular study elements that meet specific criteria that the researcher has identified., a researcher begins with specific perspectives in mind that he or she wishes to examine and then seeks out research participants who cover that full range of perspectives. For example, if you are studying students’ satisfaction with their living quarters on campus, you’ll want to be sure to include students who stay in each of the different types or locations of on-campus housing in your study. If you only include students from 1 of 10 dorms on campus, you may miss important details about the experiences of students who live in the 9 dorms you didn’t include in your study. In my own interviews of young people about their workplace sexual harassment experiences, I and my coauthors used a purposive sampling strategy; we used participants’ prior responses on a survey to ensure that we included both men and women in the interviews and that we included participants who’d had a range of harassment experiences, from relatively minor experiences to much more severe harassment.

While purposive sampling is often used when one’s goal is to include participants who represent a broad range of perspectives, purposive sampling may also be used when a researcher wishes to include only people who meet very narrow or specific criteria. For example, in their study of Japanese women’s perceptions of intimate partner violence, Miyoko Nagae and Barbara L. Dancy (2010)Nagae, M., & Dancy, B. L. (2010). Japanese women’s perceptions of intimate partner violence (IPV). Journal of Interpersonal Violence, 25, 753–766. limited their study only to participants who had experienced intimate partner violence themselves, were at least 18 years old, had been married and living with their spouse at the time that the violence occurred, were heterosexual, and were willing to be interviewed. In this case, the researchers’ goal was to find participants who had had very specific experiences rather than finding those who had had quite diverse experiences, as in the preceding example. In both cases, the researchers involved shared the goal of understanding the topic at hand in as much depth as possible.

Qualitative researchers sometimes rely on snowball samplingA nonprobability sample type for which a researcher recruits study participants by asking prior participants to refer others. techniques to identify study participants. In this case, a researcher might know of one or two people she’d like to include in her study but then relies on those initial participants to help identify additional study participants. Thus the researcher’s sample builds and becomes larger as the study continues, much as a snowball builds and becomes larger as it rolls through the snow.

Figure 7.2

Just as a snowball grows larger as it rolls through snow, a snowball sample grows larger as a study continues.

© Thinkstock

Snowball sampling is an especially useful strategy when a researcher wishes to study some stigmatized group or behavior. For example, a researcher who wanted to study how people with genital herpes cope with their medical condition would be unlikely to find many participants by posting a call for interviewees in the newspaper or making an announcement about the study at some large social gathering. Instead, the researcher might know someone with the condition, interview that person, and then be referred by the first interviewee to another potential subject. Having a previous participant vouch for the trustworthiness of the researcher may help new potential participants feel more comfortable about being included in the study.

Snowball sampling is sometimes referred to as chain referral sampling. One research participant refers another, and that person refers another, and that person refers another—thus a chain of potential participants is identified. In addition to using this sampling strategy for potentially stigmatized populations, it is also a useful strategy to use when the researcher’s group of interest is likely to be difficult to find, not only because of some stigma associated with the group, but also because the group may be relatively rare. This was the case for Steven M. Kogan and colleagues (Kogan, Wejnert, Chen, Brody, & Slater, 2011)Kogan, S. M., Wejnert, C., Chen, Y., Brody, G. H., & Slater, L. M. (2011). Respondent-driven sampling with hard-to-reach emerging adults: An introduction and case study with rural African Americans. Journal of Adolescent Research, 26, 30–60. who wished to study the sexual behaviors of non-college-bound African American young adults who lived in high-poverty rural areas. The researchers first relied on their own networks to identify study participants, but because members of the study’s target population were not easy to find, access to the networks of initial study participants was very important for identifying additional participants. Initial participants were given coupons to pass on to others they knew who qualified for the study. Participants were given an added incentive for referring eligible study participants; they received not only $50.00 for participating in the study but also $20.00 for each person they recruited who also participated in the study. Using this strategy, Kogan and colleagues succeeded in recruiting 292 study participants.

Quota samplingA nonprobability sample type for which a researcher identifies subgroups within a population of interest and then selects some predetermined number of elements from within each subgroup. is another nonprobability sampling strategy. This type of sampling is actually employed by both qualitative and quantitative researchers, but because it is a nonprobability method, we’ll discuss it in this section. When conducting quota sampling, a researcher identifies categories that are important to the study and for which there is likely to be some variation. Subgroups are created based on each category and the researcher decides how many people (or documents or whatever element happens to be the focus of the research) to include from each subgroup and collects data from that number for each subgroup.

Let’s go back to the example we considered previously of student satisfaction with on-campus housing. Perhaps there are two types of housing on your campus: apartments that include full kitchens and dorm rooms where residents do not cook for themselves but eat in a dorm cafeteria. As a researcher, you might wish to understand how satisfaction varies across these two types of housing arrangements. Perhaps you have the time and resources to interview 20 campus residents, so you decide to interview 10 from each housing type. It is possible as well that your review of literature on the topic suggests that campus housing experiences vary by gender. If that is that case, perhaps you’ll decide on four important subgroups: men who live in apartments, women who live in apartments, men who live in dorm rooms, and women who live in dorm rooms. Your quota sample would include five people from each subgroup.

In 1936, up-and-coming pollster George Gallup made history when he successfully predicted the outcome of the presidential election using quota sampling methods. The leading polling entity at the time, The Literary Digest, predicted that Alfred Landon would beat Franklin Roosevelt in the presidential election by a landslide. When Gallup’s prediction that Roosevelt would win, turned out to be correct, “the Gallup Poll was suddenly on the map” (Van Allen, 2011).Van Allen, S. (2011). Gallup corporate history. Retrieved from http://www.gallup.com/corporate/1357/Corporate-History.aspx#2 Gallup successfully predicted subsequent elections based on quota samples, but in 1948, Gallup incorrectly predicted that Dewey would beat Truman in the US presidential election.For more information about the 1948 election and other historically significant dates related to measurement, see the PBS timeline of “The first measured century” at http://www.pbs.org/fmc/timeline/e1948election.htm. Among other problems, the fact that Gallup’s quota categories did not represent those who actually voted (Neuman, 2007)Neuman, W. L. (2007). Basics of social research: Qualitative and quantitative approaches (2nd ed.). Boston, MA: Pearson. underscores the point that one should avoid attempting to make statistical generalizations from data collected using quota sampling methods.If you are interested in the history of polling, I recommend a recent book: Fried, A. (2011). Pathways to polling: Crisis, cooperation, and the making of public opinion professions. New York, NY: Routledge. While quota sampling offers the strength of helping the researcher account for potentially relevant variation across study elements, it would be a mistake to think of this strategy as yielding statistically representative findings.

Finally, convenience samplingA nonprobability sample type for which a researcher gathers data from the elements that happen to be convenient; also referred to as haphazard sampling. is another nonprobability sampling strategy that is employed by both qualitative and quantitative researchers. To draw a convenience sample, a researcher simply collects data from those people or other relevant elements to which he or she has most convenient access. This method, also sometimes referred to as haphazard sampling, is most useful in exploratory research. It is also often used by journalists who need quick and easy access to people from their population of interest. If you’ve ever seen brief interviews of people on the street on the news, you’ve probably seen a haphazard sample being interviewed. While convenience samples offer one major benefit—convenience—we should be cautious about generalizing from research that relies on convenience samples.