Phenomenography is an empirical research tradition that was designed to answer questions about thinking and learning, especially in the context of educational research (Marton, 1986). It is concerned with the relationships that people have with the world around them. The word phenomenography has Greek etymological roots. It is derived from the words phainonmenon (appearance) and graphein (description). Thus, phenomenography is a description of appearances. (Hassselgren & Beach, 1997).
Marton (1981) suggests that there are two ways to approach questions about learning: (1) to orient ourselves toward the world and make statements about it and its reality, or (2) to orient ourselves towards peoples ideas or experiences of the world. In other words, we can either choose to study a given phenomenon, or we can choose to study how people experience a given phenomenon. Phenomenography is the latter kind of approach. Its aim is to define the different ways in which people experience, interpret, understand, perceive or conceptualize a phenomenon, or certain aspect of reality.
Different people will not experience a given phenomenon in the same way. Rather, there will be a variety of ways in which different people experience or understand that phenomenons Phenomenographers seek to identify the multiple conceptions, or meanings, that a particular group of people have for a particular phenomenon. Because the focus of a phenomenographical study is on the conceptions that a particular group of people have for a given phenomenon, the conceptions of the researcher for that phenomenon are not usually a focus of such a study. Instead, the researcher attempts, as much as possible, to act as a neutral foil for the ideas expressed by the participants of the study. As phenomenography is empirical research, the researcher (interviewer) is not studying his or her own awareness and reflection, but that of the subjects. (Marton, 1994, p. 4427).
Marton (1981, 1994) believes that there are a limited number of qualitatively different ways in which different people experience a certain phenomenon. From this theoretical stance, it is irrelevant if those conceptions are considered correct or incorrect by current standards. The aim is simply to elucidate the different possible conceptions that people have for a given phenomenon.
The main results of phenomenographic research are categories of description of the various conceptions of a phenomenon. Phenomenographic research is more than simply reporting these different conceptions, however. It involves identifying the conceptions and looking for their underlying meanings and the relationship between them (Entwistle, 1997). Marton (1981) makes the following statement about this additional goal of phenomenography:
Still, we are able to point not only to conceptions making up its constituents but also to relations between certain conceptions of one aspect of the world and certain conceptions of another aspect. What we have in mind is certainly not merely a listing of one conception after another. Some aspects are certainly more basic than others and different (and more or less fundamental) layers of the perceived world can be revealed. (p. 190)
Marton (1994) also says that the different ways of experiencing different phenomena or concepts are representative of different capabilities for dealing with those phenomena or concepts. Some ways of dealing with phenomena or concepts are more productive than others. Thus, the conceptions, or ways of experiencing, and their corresponding descriptive categories can not only be related, but also be hierarchically arranged. The ordered and related set of categories of description is called the outcome space of the concept being studied.
Unlike other theoretical research perspectives, phenomenography does not make any assumptions about the nature of reality. Phenomenographers do not claim that their research results represent truth; however, they do claim that their results are useful. Svensson (1997) makes the following point about the phenomenographical position on the nature of reality:
Phenomenography does not have an articulate metaphysical foundation. The question may be raised if it has implicit metaphysical assumptions. Individual researchers doing phenomenographic research may make such assumptions but they certainly vary between the researchers. It is possible to have any and all of the metaphysical positions within the main categories of materialism and idealism and do phenomenographic research. The tradition is not based on any of these metaphysical beliefs and it is open in this respect. (p. 165)
Although phenomenography makes no assumption about the nature of reality, it does make assumptions about the nature of conceptions. The primary assumption is that conceptions are the product of an interaction between humans and their experiences with their external world. Specifically, conceptions results from a human beings thinking about his or her external world. An assumption that is extremely important to phenomenographical research is that a persons conceptions are accessible in different forms of actions, but particularly through language (Svensson, 1997).
Phenomenographic studies strive to discover the different ways in which people understand or experience certain phenomena. Although many possible sources of information can reveal a persons understanding or conception of a particular phenomenon, the method of discovery is usually an open, deep interview (Booth, 1997). Open indicates that there is no definite structure to the interview. While researchers may have a list of questions or concerns that they wish to address during the interview, they are also prepared to follow any unexpected lines of reasoning that the interviewee might address as some of these departures may lead to fruitful new reflections that could not have been anticipated by the researcher. Deep indicates that the interview will follow a certain line of questioning until it is exhausted, until the participant has nothing else to say and until the researcher and participant have reached some kind of common understanding about the topics of discussion.
The aim of an interview is to have the participant reflect on his or her experiences and then relate those experiences to the interviewer in such a way that the two come to a mutual understanding about the meanings of the experiences (or of the account of the experiences).
The experiences and understandings are jointly constituted by interviewer and interviewee. These experiences and understandings are neither there prior to the interview, ready to be read off, nor are they only situational social constructions. They are aspects of the subjects awareness that change from being unreflected to being reflected. (Marton, 1994, p.4427)
During data analysis, the researcher will identify qualitatively distinct categories that describe the ways in which different people experience a different concept. Phenomenographers believe that a limited number of categories are possible for each concept under study and that these categories can be discovered by immersion in the data, which, in most cases, are transcriptions of the interviews (Booth, 1997).
The researcher examines the transcripts of several participants interviews, looking both for similarities and differences among them. In this process, the researcher develops initial categories that describe different peoples experiences of the given phenomenon. If the interview has covered multiple topics or multiple aspects of a given phenomenon, the researcher will attempt to develop an outcome space for each topic. The only ground rules for category development are internal consistency and parsimony, or finding an outcome space that includes the minimum number of categories which explain all the variations in the data.
With these initial categories in mind, the researcher reexamines the interview transcripts to determine if the categories are sufficiently descriptive and indicative of the data. This second review of the data results in modification, addition, or deletion of the category descriptions and a third examination of the data for internal consistency of the categories of description. This process of modification and data review continues until the modified categories seem to be consistent with the interview data.
Marton (1986) says that definitions for categories are tested against the data, adjusted, retested, and adjusted again. There is, however, a decreasing rate of change, and eventually the whole system of meanings is stabilized. (p. 43).
Once a stable outcome space has been defined, the researcher attempts to develop as deep an understanding as possible of what has been said, or rather, what has been meant (Marton, 1994, p. 4428). To do this, he needs to consider not only specific categories of description, but also how the individual categories relate to each other and how one persons conceptions compare across different topics.
One of the criticisms of phenomenography is its tendency to equate students experiences with their accounts of those experiences, as Marton (1994) evidences by stating that conceptions,the focus of phenomenographic studies, are ways of experiencing Saljo (1997) reports that, at times, there appears to be a discrepancy between what researchers observe of a participants experience with a particular phenomenon and how the participant describes his experience with the phenomenon. Richardson (1999) claims that phenomenographers do not skeptically examine the effects of the interview environment or of socially accepted linguistic practices on what is reported by the students.
In order to avoid equating experiences with accounts of experiences, Saljo (1997) suggests that we refer to studying peoples different accounting practices of phenomena, which are public and accessible to study, instead of referring to studying peoples experiences. We must keep in mind, however, that such accounting practices may be socially and environmentally influenced (ie. the student might say what he thinks the interviewer wants to hear, etc.).
It may be true peoples accounts of their experiences with a particular phenomenon are not equivalent to the ways in which they experience the phenomenon. However, the only way we can begin to understand the ways in which people experience a given phenomenon is to ask each person to describe his or her experience. There is no physical way to examine a persons brain to obtain this data. We, as researchers, can make observations of what people experience, but those observations will not tell us how they experience a given particular phenomenon, especially if we accept the idea that conceptions, or ways of experiencing, are products of an interaction between the person and the phenomenon he experiences. Phenomenographical results may not be truth, in that they may never accurately describe ways of experiencing, but they may be useful. So, then, it may not matter if accounts are equivalent to experience.
One of Webbs (1997) main critiques of researchers using phenomenography is their assumption that they can be neutral foils while analyzing research data. It is more reasonable to assume that researchers have had certain experiences and hold certain theoretical beliefs that will influence their data analysis and categorization. Webb calls for researchers to make their backgrounds and beliefs explicit, not because having these backgrounds and beliefs is bad, but rather because the readers and users of phenomenographic research need to be informed about all variables that have potentially affected the study results. My personal opinion is that such self-examination may lead to additional insights into the data and, to some extent, a more critical examination of how the researchers own beliefs have affected the research and the results of this research.
Other researchers have questioned the reliability and repeatability of phenomenographic studies. On issues of reliability, Marton (1986) says that it is possible that two different researchers would discover different categories of description while working on the same data individually. However, once the categories have been found, they must be described in such a way that all researchers can understand and use them.
Marton compares this process to botanists that discover a new plant species on an island. If the new species does not appear to fit into already existing category, the botanist must develop a new category of classification for it, and it is highly probable that a separate botanist would develop a qualitatively different category for that new species. However, once the botanist has developed and described a category, the category is now accessible and available for classifying plants that any botanist finds. Indeed, once the category is developed and described, it becomes useful to others who use the results of the study.
There are certain benefits to using the results of phenomenographic research in a higher education institution. At this level of instruction, students are generally encouraged to develop conceptual understandings (Entwistle, 1997). It is often the goal of teachers to help their students develop conceptions that are consistent with those held by recognized experts in various fields. However, students often have multiple different conceptions for a phenomenon that are not necessarily consistent with the conceptions held by experts. Marton (1986) claims that a careful account of the different ways people think about phenomena may help uncover conditions that facilitate the transition from one way of thinking to a qualitatively better perception of reality. (p. 33). Thus, phenomenographic information about the different conceptions that students hold for a particular phenomenon may be useful to teachers who are developing ways of helping their students experience or understand a phenomenon from a given perspective.
Another possible benefit of phenomenographical research is that students may become conscious of contradictions in their own reasoning and become more open to alternative ideas as they reflect on their perceptions and understandings of their world experiences (Marton, 1986).