Observation is a very “natural” way of gathering data and information – probably everybody can identify situations where humans are scrutinizing their surroundings. Selltiz et al. note: “We are constantly observing – noticing what is going on around us . . . ; as long as we are awake, we are almost constantly engaged in observation. It is our basic method of getting information about the world around us” (Selltiz et al. 1967, 200). However, these everyday observations are not sufficient for social research, since their informational basis stems from unclear viewpoints, undocumented (re)constructions of the observer, and random events. So in order to become a scientific technique, an observation study has to follow some rules – most notably, it has to be systematic in both its planning and recording phases. This does not mean that it has to be standardized; however, even if an observation study collects information in an open way, it has to follow a clear research strategy.
In comparison to the other two central ways of data gathering in communication studies – content analysis and interview – observation plays a smaller role in research nowadays. In many handbooks of communication studies, it is just noted as a means of getting some ideas “from the field” in order to develop items for survey studies or categories for content analysis – and sometimes it is missing altogether. This is not necessarily the case in neighboring disciplines, like sociology or psychology: here, observation studies play a central role in research. In sociology, field studies are a core instrument for getting access to social phenomena, and in psychology, observations play an important part in experimental study designs.
Roots And Development Of Observation As A Scientific Metho
While observation as a mode of perception is simply a part of human existence and therefore cannot be traced back to some historical roots, its scientific origins can be identified more clearly. Anthropologists of the nineteenth and early twentieth centuries were among the first who used observations systematically as a way of gathering information during their travels around the world, and their ethnographic studies of cultures foreign to them can be seen as one central developmental branch of sociological studies. Still, these observations were very open in their focus and methods, so the resulting travel reports were not scientific when judged by modern standards.
With the dramatic changes of societies during industrialization, sociologists started to study the life of the working class and new forms of existence in metropolitan areas in contrast to traditional rural life. While there were still many smaller, open studies among the observations applied in this context, some researchers followed a more coherent and methodologically strict perspective on social phenomena. One of the most prominent examples of this development is the so-called Chicago School, with sociologists like Nels Anderson concentrating on urban life, using observation as a central tool for getting access to social realities.
However, observation studies of human behavior are not limited to anthropology and sociology, and accordingly, one will find some roots of the method in other fields as well, most notably in psychology. Behaviorists (e.g., B. F. Skinner and J. B. Watson) started to analyze animal (and later human) behavior in the first half of the twentieth century, using observations in controlled experimental settings. Later on, some spectacular psychological experiments like the Milgram study or the Stanford Prison experiment led to a certain image of psychological observation studies, which is still influential in popular beliefs of what psychology is about.
In media and communication studies, (field) observation studies were a central research tool right from the beginning of the discipline. Both the behavior of audience members in natural home environments and the work routines of journalists in newsrooms were the targets of many observation studies. As within sociology, the early studies did not stick to methodological standards and the documentation of the data gathering was often very weak. For example, Walter Gieber notes about the method of his early study of 16 telegraph editors: “Field visits to the newsrooms were in two waves to permit an extensive interview with the deskman and observation of the editing process” (1956, 424). Following the development in other social sciences, later studies became more rigorous in their study methodology. However, large-scale observation studies with standardized instruments are still rare in communication studies, at least when compared to the other types of data gathering.
Just lately, the influence of cultural studies in communication and media studies has led to a growing number of observation studies, mostly based in the “qualitative” paradigm – so they favor open instruments and exploration studies, based on a few handselected cases. This has been criticized by some researchers as well, who regard these as being merely “chance observations,” so there is an ongoing debate on the status of the method in communication studies.
When mentioning the term “observation,” many researchers in communication studies think of their first steps when doing a study, just as Donohew and Palmgreen (2003, 119) describe it: “Most theory building begins with rather crude observations of events or other phenomena that seem to be related. At first, the scientist does not know exactly what to look for. Observation at this point tends to be rather unstructured and informal.” This type of information gathering could be identified as open, (probably) participant observation without a structured instrument. While this is a typical part of the research process, it is usually not mentioned as a special method. There are other variations of observations, however, that deserve this status.
When trying to differentiate between several types of observations, one can discern several dimensions for discriminating these types. Three dimensions are especially useful for such a differentiation, since other variations could be related to them:
1 level of standardization;
2 involvement of the observer;
3 visibility of the observer.
First, the level of standardization refers to the use of strict observation rules, which may be defined in code-books and formal observation guidelines, in contrast to open documentation or diary methods. Very often, researchers use a combination of both open and closed elements, in order to control the observation process while retaining the openness for surprises from the field.
Second, in many cases, access to the field is only possible if the observer becomes part of the observed reality: “The participant observer gathers data by participating in the daily life of the group or organization he studies” (Becker 1958, 652). So the observer is directly involved in the field – for example if he or she is working as a journalist when doing an observation study of newsroom selection processes.
Third, the involvement of the observer is often confused with the visibility of the observation process. However, active participation of an observer in the field does not necessarily mean that the observed persons are aware of being studied – for example, if the observer is working undercover. On the other hand, a completely passive observer can be very visible, for example sitting next to and observing somebody, thus being obtrusive and probably causing reactivity.
Typical Steps Of An Observation Study
No matter what type of observation study a researcher chooses as adequate for a given research interest, it will follow a few typical steps, which are somewhat different from those of interview studies or content analyses. Of course, observation studies start with a research interest or research questions, like other types of empirical research. One should extract and clarify the underlying dimensions of meaning and central concepts touched by these questions – even when doing a “qualitative” exploration study. The identification of dimensions in the research interest/questions is not only aimed at the (sometimes unknown) social phenomena, but also at one’s own perspective – thus, it is also necessary if the researcher does not know a lot about the field itself. Based on this first clarification, the researcher has to decide where, when, what, and whom he or she wants to observe – which means: identifying the research field (place, time, social context), subjects/objects (individuals, groups, inanimate objects), and cases (actions, events, time units). Usually, this leads to refined research questions, and, in some cases, also to a further clarification of research questions, up to testable hypotheses and variables in the case of highly standardized studies. Following this step, the researcher has to develop observation procedures, which include a research plan and observation tools, like code-books or diaries, and might involve the use of technical devices for recording aspects of the social phenomena under observation (microphones, digital audio recorders, cameras etc.).
When working with several observers, the researcher has to train them. Very often, this means a simulation of the observation situation or just an explanation of the code-book, since advance access to the field is not possible or is problematic. The same applies to pretests. Sometimes, one could chose to do the training and pre-tests in a “test” field similar to the actual observation field, with comparable observation objects/subjects. However, even with advance training, going into the field can be problematic, for example if an observer starts to openly observe an editor in a newsroom. Both the observer and the observed subject(s) have to become accustomed to the situation, so the field phase needs some kind of introductory phase as well. Material from these first observations should be left out of the final data analysis and interpretation. In some cases, the data analysis and interpretation can also lead to further research questions – and to a second “loop” of the whole research process, including the development of refined research instruments and more field time.
Applications In Communication Studies
As described above, observations are frequently used as a first “explorative” access to social phenomena, for example if there is not sufficient in-depth knowledge to design “closed” items for questionnaires. The method can be applied in a much broader way. For example, it is often the method of choice when it comes to newsroom studies. Observations are also used to gain access to the everyday life of media audiences, especially if one is interested in habitualized use patterns that cannot be reflected upon by the audience members themselves. There are many reasons why subjects might not be able to verbalize certain aspects of their social reality, both in open interviews or closed surveys. They might not be aware of them, or they might not have the language capabilities to describe them (for example, children); here, observations are sometimes the only solution.
Observations of newsrooms and media audience members in everyday settings are usually direct observations by human observers. However, one should not forget other types of observation studies. There are also (sometimes automatic) observations with the help of technical equipment. These machine observations are used heavily in market research (TV quota measurement) and in experimental studies. It has to be noted, though, that the analysis of automatically acquired behavior traces (for example, computer log files) and experimental studies are very different from the classic “role model” of field observations, and many researchers actually do not even think of these types of research as being observation studies.
- Becker, H. S. (1958). Problems of inference and proof in participant observation. American Sociological Review, 23, 652 – 660.
- Berger, A. A. (2000). Media and communication research methods: An introduction to qualitative and quantitative approaches. Thousand Oaks, CA: Sage.
- Donohew, L., & Palmgreen, P. (2003). Constructing theory. In G. H. Stempel, III, D. H. Weaver, & G. C. Wilhoit (eds.), Mass communication research and theory. Boston, MA: Allyn and Bacon, pp. 111–128.
- Gieber, W. (1956). Across the desk: A study of 16 telegraph editors. Journalism Quarterly, 33, 423 – 432.
- Quandt, T. (2006). Methods of journalism research: Observation. In M. Löffelholz & D. Weaver (eds.), Journalism research in an era of globalization. New York: Blackwell.
- Selltiz, C., Jahoda, M., Deutsch, M., & Cook, S. W. (1967). Research methods in social relations. New York: Holt, Rinehart and Winston. (Original work published 1951).