Читать книгу The Research Experience - Ann Sloan Devlin - Страница 14

In one of Kahneman and Tversky’s classic examples, participants were presented with the following: “All families of six children in a city were surveyed. In 72 families the exact order of births of boys and girls was GBGBBG. What is your estimate of the number of families surveyed in which the exact order of births was BGBBBB?” (Kahneman & Tversky, 1972, p. 432). Not surprisingly, a significant number of the respondents (75 of 92) said the second option was less likely to occur because, as Kahneman and Tversky argued, it seems less representative of the population. When the question is posed in terms of the frequency with which two birth sequences occur (BBBGGG vs. GBBGBG), the same participants pick the second sequence. The first looks “fixed” or nonrandom to us (and them). How representative something looks is one heuristic or bias that may influence the research process. A researcher might select a stimulus (e.g., photograph) as representative of a population of interest (e.g., urban parks) without knowing the full range of existing sites (compare Figure 1.1 and Figure 1.2). In this instance, both of these photos come from the same park (Brooklyn Bridge Park in Brooklyn, New York) but would communicate very different aspects of the park depending on which photograph was used. If you try to generalize from a single picture or even a limited range of pictures to say something definitive about people’s evaluations of such settings, the results might be overstated.

Figure 1.1 Pier 2 at Brooklyn Bridge Park

©Ann Sloan Devlin

Figure 1.2 View of the Brooklyn Bridge From the Brooklyn Bridge Park Greenway

©Ann Sloan Devlin

Although the work of Kahneman and Tversky focuses on cognitive decision-making processes (e.g., the decisions we make about stimuli), the idea of representativeness emerges in other ways in research. You may be familiar with such phrases as “a representative sample” or “a randomly selected sample” (the example from The Big Short earlier in this chapter raised the issue of sampling; see Chapter 11 for a fuller discussion of sampling).

One central question in every research project is who the participants are and to what extent the results of the study are therefore “representative” of the population of interest. If research uses a participant pool that consists of students enrolled in an introductory psychology course, there are several questions to ask about who participates, starting with the degree to which people who take an introductory course in psychology are representative of that student body as a whole (by gender, race, income, and many other qualities). Every decision about securing participants (e.g., the time of day the study is offered) is likely to influence the representativeness of the sample and, in turn, of the results.