Naturalistic Observation

Modified: 2007-01-31

When researchers observe an individual or group in their natural habitat while limiting their own effect on the individual or group, they are using naturalistic observation.


The goal of naturalistic observation is to discover behaviors that occur naturally in a particular environment. Goodall’s work at Gombe qualifies as naturalistic observation because she observed the chimpanzees while limiting the effects of her presence. After the chimpanzees were accustomed to Goodall’s presence, she could observe their usual behavior. Using naturalistic observation, Goodall discovered previously unknown facts about chimpanzees. Lab studies or zoo studies could not have produced data similar to hers. Of course, humans can be studied by naturalistic observation, too.


Examples of Naturalistic Observation Using Humans


Amato (1989) observed caretakers of children in public places in California and Nebraska. He found that 43% of the children he observed had male caretakers. Males were more involved with their children in recreational settings such as playgrounds, but in restaurants, females were more involved. When taking care of older children, boys, and mixed-sex pairs, however, males were more likely to be the caretakers. Amato created a naturalistic observation framework that he considered “sensitive enough to test a variety of hypotheses about male involvement with children” (p. 981). Naturalistic observation research such as Amato’s can be conducted easily in public places and can yield surprisingly large amounts of data.


Graham and Wells (2001) conducted a naturalistic observation study of bar patrons by recording late night aggressive behavior in a Canadian tavern. They used naturalistic observation and interviews in their research. In the naturalistic observation portion of their research, 117 aggressive incidents were observed during the 93 nights of the study. Most of the observation periods were weekend nights between midnight and 2:30 a.m.; the patrons were unaware that research was being conducted. The researchers documented patterns of aggressive behavior in this particular bar. For example, they found that nearly 75% of the incidents involved males only. Also, moderate or higher levels of physical aggression were observed in 67% of the incidents. About 33% of the incidents occurred outside of the bar’s premises. Graham and Wells identified several triggers for aggression in bars, including problems with bar staff, rowdy behavior, and interpersonal relationship problems. Studies such as this might be helpful to bar managers who want to reduce aggressive incidents in their establishments.


 Naturalistic observation relies on one or more observers entering a specific research environment, observing behavior, and recording it in a reliable manner. In naturalistic observation, researchers take great care not to alter or influence the behaviors they are observing. Because data collection takes place where the behaviors typically occur, naturalistic observation studies tend to be generalizable to other populations. That is, naturalistic observation studies generally have high external validity. Chimpanzees in other natural situations are likely to behave the same as chimpanzees at Gombe. Caretakers in other parts of the United States will probably behave like those that Amato observed in California and Nebraska. Bar patrons throughout the Western world  are likely similar to bar patrons in Canada. Naturalistic observation results can be generalized if observers do not alter the behaviors of those being studied. Observers must attend to issues of internal validity too. Consistency of observations is paramount as is sampling. When conducting naturalistic observation, observers must take care that their observations are unbiased. They should also address the question of whether their samples are typical of the population. We now look at the details of conducting naturalistic observation research.


Conducting Naturalistic Observation Research


Participants Either humans or animals can be the subjects of naturalistic observation. The methodological issues are availability of participants, respecting their privacy, and whether observations will be conducted from an open or concealed position. Participants can be found in a wide variety of locations other than African jungles, bars, or urban public places. However, some places are unusable. Privacy concerns limit the use of naturalistic observation to places that are generally accessible to the public. Privacy (and the law, as well) prevents researchers from peering into their neighbor’s windows. Conducting naturalistic observation in public restrooms is ill advised. Researchers also have to decide whether or not to conduct their observations openly. Both open and concealed strategies are used in animal and human naturalistic observation research. Observing from a hidden position is often preferable because it allows researchers to begin data collection immediately. Being hidden, researchers are confident that those observed are not acting differently because they are being observed. Thus, studies with hidden observers reduce or eliminate reactivity(see p. XXX), which, as you may recall from, is a threat to external validity. At times, however, researchers cannot hide. In those cases, they must collect data over a long enough period to assess whether their presence is affecting the behavior of those observed.


Apparatus The apparatus for naturalistic observation can be extremely simple or very complex. Sometimes a pad, pencil, and stopwatch will suffice. Moving to the complex end of the spectrum, observers might use video cameras, motion detectors, infrared detectors for observations in darkness, or other aids. Regardless of the apparatus used, the task is to reliably record the behavior of the participants.


Procedure  Procedures for conducting naturalistic observations vary widely. The key is to find and record behaviors in psychologically interesting situations. Suppose we wanted to naturalistically observe a second-grade classroom. Ideally, observing the students through one-way mirrors built into a wall of their classroom, or observing them via video cameras (which are standard features in many classrooms) minimizes reactivity. Using either of those methods allows data collection to begin immediately. If there are no one-way mirrors or video cameras, data can still be collected, although additional time may be needed. Observers can enter the room daily and sit in the same place. After a time, the second graders pay less and less attention to the researcher and eventually begin to act as they did before. When observers are confident that the second graders are acting as their usual selves, data collection begins.


Results Naturalistic observations may produce either quantitative or qualitative data. Often observations are coded into numerical form such as counting the number of times a behavior occurs and analyzed quantitatively. Naturalistic observation data can be qualitative as well; most qualitative methods involve observing and re–observing recorded observations and then identifying emergent commonalities and themes. Afterward, researchers summarize the commonalities and themes as results.


Ethics   As with any research method, ethics must always be considered in planning, conducting, and reporting naturalistic observation research. Section 8.05 (See Appendix B) of the Ethical Principles of Psychologists and Code of Conduct (APA, 2003) specifically excludes naturalistic observation from the requirement of informed consent. However, researchers must still protect participants’ “risk of criminal or civil liability or damage [to] their financial standing, employability, or reputation, and confidentiality” (p. xx). [get page] For animals, parts of Section 8.09 apply. Thus, researchers conducting naturalistic observations on animals should not cause animals pain or stress. An important ethical issue in animal naturalistic observation research is whether or not the researcher should intervene to prevent harm or death to that animal by other agents in the natural environment. The answer is no, researchers should not intervene.


Jury deliberations are an example of a situation in which naturalistic observation cannot be used. They are private and only members of the jury can observe and report on those deliberations. However, a who juror observes the jury’s proceedings and then reports them is an example of participant observation, our next topic.

Back to Chapter 2 Outline