Field Experiment

5 Key excerpts on "Field Experiment"

• 

  eBook - ePub

  Applied Child Study

  A Developmental Approach

  • Anthony D. Pellegrini, David F. Bjorklund(Authors)
  • 1998(Publication Date)
  • Psychology Press

    (Publisher)

  In the previous section on laboratory (true) experiments, we noted that all aspects of the environment had to be randomized and controlled in order to make causal inferences about our treatments. To say the least, randomization and control are very difficult to achieve in applied settings. We offer Field Experiments as a compromise in that they are more realistically achieved in applied settings.

  Field Experiments are experiments that take place in real situations. The researcher, however, manipulates these environments. Some of the best examples of Field Experiments involve research in preschool settings. The work of Johnson and his colleagues (e.g., J.Johnson & Ershler, 1981), Lawton (e.g., Lawton & Fowler, 1989), and P.K.Smith and Connolly (1980) with preschoolers is illustrative of excellent Field Experiments where the differential effects of curriculum was studied. In other cases with primary schools, researchers have manipulated children’s exposure to different recess timing regimens to determine their effect on children’s attention to seat work (Pellegrini, 1995). In one Johnson study, for example, preschool curriculum (traditionaldidactic vs. open-exploratory) was systematically manipulated in each classroom to determine the effect of each on children’s play behaviors (Johnson & Ershler, 1981). Lawton and Fowler (1989) compared Piagetian and Ausubelian programs on teachers’ and children’s language. In the P.K.Smith and Connolly (1980) study, classrooms that differed on aspects of social and spatial density as well as types of toys were designed. Pellegrini and colleagues compared the effects of long versus short periods during which children were deprived of recess.

  A major advantage of Field Experiments over those in a laboratory is that the children are being observed in a realistic and familiar setting. As a result, the conclusions made about children’s observed behavior also are likely to be supported by observations in similar, nonmanipulated, settings.

  The Field Experiment also has its disadvantages. Experimenters typically exercise control on an external institution, like the classroom, as they can in an experimental situation. For example, teachers in a Field Experiment, because they were often hired for reasons other than their ability to act as experimenters, may not be consistent in their use of a particular experimental technique. As a result, the actual experimental treatment may not be consistent with the ideal treatment. Experimenters who are trained and retrained to implement a specific procedure may be more reliable.

  Sign up to read

  Learn more about book
• 

  eBook - ePub

  Population-Based Survey Experiments

  • Diana Mutz(Author)
  • 2011(Publication Date)
  • Princeton University Press

    (Publisher)

  not the location in which the research takes place or the people who participate. Nonetheless, these two factors—the research setting (laboratory versus field) and the research design (carefully manipulated versus passively observed independent variables)—are often confounded. Just because non-experimental studies are limited in their ability to attain internal validity does not mean that experiments therefore must be limited in their ability to attain external validity.

  Field Experiments provide an obvious and widely recognized exception to this purported tension: to the extent that Field Experiments make it possible to randomly assign people successfully to experimental treatments, they are considered roughly equal to laboratory experiments in the strength of causal inferences that may be drawn. In other words, greater internal validity does not necessarily come at the obvious expense of external validity. In cases where the treatments cannot be assured of reaching the correct subjects successfully, Field Experiments sometimes devolve into quasi-experimental designs, but when they are executed as intended, they are true experiments by virtue of using random assignment to conditions.

  But because Field Experiments are, by definition, executed in more naturalistic settings, it is often erroneously claimed that they, therefore, must have greater external validity than laboratory studies. Results of Field Experiments are routinely claimed to be more generalizable simply because they are Field Experiments.

  The logic behind this claim is based solely on the similarity of settings. As the conventional wisdom has it, “To generalize you need similarity between the laboratory and real life—that is, field settings; lab and field settings are very dissimilar; therefore generalization is not justified (or highly dubious).”6

  Sign up to read

  Learn more about book
• 

  eBook - ePub

  Ethical Issues in Psychology

  • Philip Banyard, Cara Flanagan(Authors)
  • 2013(Publication Date)
  • Routledge

    (Publisher)

  The argument is that merely placing a study in a laboratory does not make it unethical. Laboratories simply increase our ability to control variables, which is a necessary part of good research. Equally, a study using an experimental technique is not unethical because it is an experiment – what we want to focus on are the particular features in any study that raise ethical concerns.

  Field studies

  Some

  field studies

  are experiments (such as Bickman's Field Experiment about obedience to authority, described on p. 37) but there are also field studies that are not experimental.

  There are two key features of field studies that make them ‘desirable’ for understanding everyday human behavior. First, a field study is conducted in a relatively ‘natural’ environment so people are more likely to exhibit everyday behavior. But probably, most importantly, in most field studies the participants do not know their behavior is being recorded and this is the main reason why they exhibit everyday behavior. It is this aspect of field research that gives rise to ethical problems. It can be argued that this aspect of field research disregards their right to privacy and right to autonomy and respect.

  A well-known example of a field study is the ‘Good Samaritan’ experiment by Piliavin et al. (1969). Psychology students boarded subway trains in New York city with the aim of seeing how passengers would respond to an emergency situation. A confederate faked some kind of collapse (either holding a bottle and smelling of alcohol or using a cane). The student observers took note of how long it took for help to be forthcoming.

  What are the ethical issues here? There was no possibility of gaining prior consent, nor was it possible after the experiment to inform people about the study (i.e. debrief them). This challenges the participants’ right to autonomy. In addition, participants might have experienced some psychological distress either from what they witnessed or their own feelings of self-doubt because they did not help. There would have been no opportunity to reassure them afterwards.

  There is also the issue of privacy. A general rule (as we will see later) is that people should not be observed in a study unless it is in a public place, in other words one where they might expect to be observed by others. The subway setting would then not count as an invasion of privacy. According to this rule, the Good Samaritan study would be acceptable but the Tearoom Trade study (p. 28) would not.

  Sign up to read

  Learn more about book
• 

  eBook - ePub

  Relationships, Well-Being and Behaviour

  Selected works of Harry Reis

  • Harry Reis(Author)
  • 2018(Publication Date)
  • Routledge

    (Publisher)

  flow (a mental state in which people are fully and energetically immersed in whatever they are doing) tends to be low among adolescents in many school activities and while watching television. Third, new technologies allow researchers to ask context-sensitive questions (Intille, 2007; Intille, 2012). For example, accelerometers (which identify motion patterns) let researchers prompt participants to record their thoughts or feelings upon awakening or completing exercise. Similarly, questions tailored to the participant’s location can be administered on the basis of readings from global positioning devices (e.g., on a crowded city street, at home, or in nature).

  Laboratory experimentation sometimes does not consider the extent to which the laboratory setting itself may contribute to the outcomes of research. This seems ironic; if settings had no influence on behavior, why would they need to be controlled? Every laboratory has unique physical features, but beyond this, the laboratory setting itself may engender certain expectations and scripts (e.g., scientific legitimacy, serious purpose, suspicion about possible deception, concerns about being observed, the need for attentiveness), all of which may affect the participant’s thoughts and behavior (Shulman & Berman, 1975). One example of this, demand characteristics (cues that suggest to research participants the behaviors that researchers expect of them), are a well-known source of bias in research (Wilson et al., 2010). To be sure, as described earlier, research findings obtained outside the laboratory are often influenced by context. However, those contexts tend to be characteristic of the participant’s life and experience, which, far from being a confound to be eradicated, contributes to the ecological validity of daily life studies. Moreover, natural contexts tend to offer more distractions and alternatives (e.g., participants have some choice over what they do, when, where, and with whom), affording self-direction and spontaneous selection. In field research, the setting thus becomes fundamental to theoretical accounts of behavior (Weick, 1985). In a laboratory cubicle, participants can do little else but complete the tasks assigned to them by researchers as quickly as possible.

  Contexts differ along many dimensions, some of which seem likely to have minimal impact on research. For example, administering a standardized survey in a classroom versus a laboratory cubicle may make little difference, whereas conducting a Field Experiment on the impact of affectionate smiles on attraction at a singles bar versus a laboratory room may matter more. Snyder and Ickes (1985) differentiated situations in terms of the strength of their cues about behavior. So-called strong situations are relatively structured, providing salient, unambiguous, and compelling cues about appropriate behavior. Weak situations, in contrast, are unstructured, offer little or no incentives, and have few or ambiguous cues to guide behavior. Snyder and Ickes propose that strong situations are likely to support normative theories—that is, most people behaving the same way— whereas weak situations are more likely to reveal individual differences (a sensible proposal that has yet to be tested empirically; Cooper & Withey, 2009). Either is amenable to daily life research.

  Sign up to read

  Learn more about book
• 

  eBook - ePub

  The SAGE Handbook of Contemporary Cross-Cultural Management

  • Betina Szkudlarek, Laurence Romani, Dan V. Caprar, Joyce S. Osland(Authors)
  • 2020(Publication Date)
  • SAGE Publications Ltd

    (Publisher)

  Given the complexity of experiments, you may wonder why we should bother with experiments. There are multiple answers that vary in degree, but they all come back to the same answer: because they provide useful information about how the world is likely to work (Falk & Heckman, 2009). We often have competing intuitions about how the world works – should I reward individuals regardless of their input (so as not to upset group harmony) or should I reward the most efficient single employee (to incentivize individual performance)? Is it better to consult employees about forthcoming changes (I risk appearing weak) or should I make decisions because I have all the information (while employees do not have access to all the information) and increase my status as a determined and effective leader? Experiments can help us to sort out which of the two options might be more beneficial in what context, how much of a difference it may make and also when to change course of action. You may now counter that experiments are often highly artificial and do not represent real-world contexts. Survey research or observations might provide much richer context and can tell us when to use which strategy (because we may see that teams perform better if we use one incentive strategy but not the other). The problem is that both survey and observational studies cannot establish cause–effect relationships (they are providing correlational evidence). Furthermore, in real-world contexts we often face situations where we cannot be sure which of the many interrelated variables is actually the most important for the changes in our variable of interest. Experiments allow us to isolate variables and test whether they can account individually for changes in the dependent variable.

  Of course, experiments should be just one method in the larger toolset of a researcher. We need to triangulate our methods and test whether we get similar results when using different methodological tools. Similarly, a single experiment is not going to invalidate results obtained with different methods. But a series of experiments showing a consistent set of results will be informative and can provide important insights for more effective management.

  Operationalization as the Key to Experiments

  Experiments are conducted to test some specific theoretical hypotheses. For example, are multicultural teams more or less creative than monocultural teams? In this example, we have theoretical variables and we specify how an independent variable is thought to be causally relevant for changing a dependent variable. Yet, to test this hypothesis we need to translate the theoretical variables into measurable factors, such as responses to questionnaires or employee evaluations. This process is called operationalization and it is key for experiments.

  Sign up to read

  Learn more about book