Brief Communication

User studies are important for both library and information professionals, educators and researchers, and are also one of the main themes in library and information science (LIS). User studies in the area of health information, presented at the EAHIL conference, suffer from a lack of methodological variety. The objective of this article is to discuss the role of qualitative research as a complement to the tradition of quantitative research in user studies in the health care context. Qualitative research is today undertaken across all disciplines in social science. This has not always been the case and the debate between quantitative and qualitative researchers has in many disciplines been both intensive and polarized. The quantitative research tradition has also been strong in LIS, a fact that a number of commentators have pointed out.1-5 Since the beginning of the 1980s, however, there has been an increasing reorientation towards pluralism and methodological development, not least in the field of user studies. The qualitative tradition in social science has a long history that has its roots mainly in sociology and anthropology.6 Qualitative research is often confused with particular methods for data gathering such as interviewing and observation. These methods dominate in qualitative research but they are not exclusively used in this tradition. Instead, a structured interview or observation can also be used for quantification and statistics. Thus, the choice of research approach is rather a question of how we pose our research questions. In other words: What do we want to know? Does the answer to the research question need to be expressed in percentages and figures, or is the question better answered by interpretation of rich descriptions of how people understand their world and give meaning to their experiences? A good definition of qualitative research in relation to quantitative research is given by Denzin & Lincoln.6 The primary aim of quantitative research in user studies is to measure, often to generalize and thereby to predict users’ behaviour and/or attitudes in relation to information through methods such as questionnaires; analyses of log files or structured interviews. Qualitative research, on the other hand, aims to interpret and thereby understand the underlying motives of individual users’ behaviour by taking into consideration the whole picture in which information seeking and use occurs. Qualitative research includes such data-gathering methods as participating observations, interviews and text analysis used in an unstructured and non-quantitative way. The qualitative tradition emphasizes the importance of going beneath the visual and measurable surface of the users’ information behaviour and asks how that behaviour can be understood. Furthermore, qualitative research questions the view of the researcher as a neutral observer and argues that the research result is partly constructed by the research process itself. This can be exemplified by the way in which an informant’s reading habits might change as a consequence of knowing that they will be interviewed. In this case, the researcher has abandoned a neutral position and become instead a part of the informant’s life. In user studies we try to answer questions such as: Who are the users? How do users seek information? What are they seeking? Are they satisfied with the available information service? Why do they choose to seek (or not seek)? What are the results of their information seeking and use? A broad definition of user studies is used here, which includes all aspects of empirical studies where the purpose is to gain knowledge of or opinions from seekers and users of information, for example database use, reading habits, library usage and publication patterns. This definition includes everything from more traditional user studies to assessments of particular services such as user education. User studies in the health care context constitute an important subfield where the users’ relation to information sources and channels are studied. In the first six conference proceedings of the European Association for Health Information and Libraries (EAHIL),7-12 approximately 98 user studies are reported.† † The task to identify user studies at the EAHIL conference has in some cases been difficult. Many papers do no explicitly present their method and sometimes this has implied a need for interpretation on the researcher’s part. Of these there are a few examples where a combination of qualitative and quantitative approaches are used and only four studies that depended solely upon a qualitative approach.13-16 This would seem to indicate that the qualitative approach is seldom used, although this sample should obviously not be taken as representative of LIS literature as a whole. The EAHIL conference is, of course, not the only forum for presenting user studies in health information, but it is a central arena for the dissemination of knowledge and professional development in Europe and therefore of particular interest. Another striking feature is that users are almost exclusively studied from the viewpoint of librarians and the library systems. This means that it is information seeking and use seen from the library and information professionals’ perspective, not from the users’. This suggests that user studies are undertaken mainly to evaluate a certain project or service and often constitute a part of an action-orientated approach aimed at improving the particular project or service. In a situation where budget constraints are the normal situation for many medical libraries and other information services it is of course important to attract notice to the value of the service, and statistics are an effective way of doing this. In the EAHIL conference proceedings, however, there has been a lack of papers that aim to understand the users from their own perspective; to understand the role of information in the users’ working life, irrespective of particular information systems. In LIS the library point of view in research is often called system-orientated in contrast to user-orientated. The desirability of adopting a more user-orientated approach was stressed as early as 1986 by Dervin & Nilan.17 Today, this change towards user-orientated research is clearly visible in contemporary studies on information needs, seeking and use.18 The combination of an almost absolute dominance of a quantitative research and a system orientated perspective in EAHIL conference papers needs challenging. Three common qualitative approaches are: the descriptive approach, the interpretative approach and the discursive approach. The approaches have moved beyond the view of qualitative data as representative of a stable objective reality and peoples’ actual behaviour in this reality. This division can be found in a number of texts on qualitative research. For the presentation below Kvale19 and Ashworth20 are used to present three approaches that focus on different aspects of qualitative data. To some extent these also contain contradictory claims. The descriptive approach emphasizes experienced reality, the so-called life-world, as it is seen and expressed by the individual human being. Phenomenology is the most well-known methodological orientation in the descriptive approach. Open and unstructured interviews are the preferred method of collecting data because they take into account the informants’ own experience of what is important regarding the phenomena under investigation. It is, accordingly, subjective experiences that interests the descriptive researcher and the question of the origin of the experiences is not of interest. External validity is without interest in a phenomenological study; the external world exists only in so far as it is experienced by individual human beings. In the descriptive approach it is also emphasized that the researcher should suppress his or her pre-understanding of the phenomenon under investigation when entering research. This entails suppression of the researcher’s awareness of previous research, scientific theories and other previously held knowledge. Only when this is achieved can a direct description of the essence of the phenomena appear for the researcher. A central part of the descriptive approach is that the search for common themes of how different individuals describe their life-world should be left to the later stages of the analysis. With its focus on individuals’ own experience the descriptive approach has dominated in qualitative research. According to an interpretative tradition, interpretation is required when our direct understanding of the phenomenon is not sufficient. Hermeneutics is an interpretative approach with a long tradition, especially in the humanities, and today different types of data are used, such as published texts, interviews and observations. The will to make sense out of the world is a fundamental trait common to every individual and interpretation is, thus, essential in human life. The researcher does not and cannot explore the world from a neutral, outside position. In the meeting between researcher and the object under investigation a ‘fusion of horizons’ is negotiated where understanding is reached depending on the viewpoint of the interpreter and that which is interpreted. In modern hermeneutics, it is emphasized that since no objective reality exists, objectivity is therefore an impossible goal in research. The descriptive approach avoids interpretation, at least until the end of the analysis. According to a hermeneutical approach, human experience is always interpreted and the researcher enters the research field as an interpreter. Thus, to suppress the researcher’s pre-understanding is, in a hermeneutic tradition, impossible. Instead the researcher should be aware of his or her prior knowledge and assumptions. Furthermore, the interpretative approach emphasizes that data gathering should as far as possible be done in the informants’ local context and with the informants’ own words. The meaning of separate parts must be understood in the light of the whole. This understanding is encapsulated in the concept of the hermeneutic circle. The interpretative approach still focuses on the unique individual human being even if the context plays an more important role than in the descriptive approach. In the discursive approach, emphasis moves from the unique individual to the discourses they utter. Discourse is determined largely by the social institutions that exist. For example, to analyse data gathered from interviews is an analysis of the discourses that exist in a particular culture rather than an investigation of the experiences of independent human beings. Discourse analysis questions the role of language used by informants as a mirror ‘reality’, as in the descriptive and interpretative approach and also questions the role of language used by informants as a mirror of an individual’s unique and independent experiences. Language categories and discourses to some extent construct the way individuals experience reality in a particular culture. A discourse is seen as collective resources of a particular culture, or setting, that individuals use to experience and act in the world. Thus interview data may be seen as being constructed by the interview context and, indeed, could be said to reflect only the interview situation. For this reason researchers within the discursive approach often work with ‘naturally’ created data, i.e. data created beyond the influence of the researcher, such as conversations and media texts. A researcher can be more or less ‘orthodox’ in the particular approaches presented above. From my point of view, though, the approaches have no clearly defined boundaries but instead they represent a continuum—from a focus on the individual’s subjective experiences towards a focus on the culture and its collective discourses through which the individual speaks. The three approaches can be used at different levels in the analysis of data in a study. Studies of this type start by asking what the informants actually say from their own point of view, then proceed to interpret the underlying meaning of the interviews seen as a ‘fusion of horizons’ between researcher and informant, and finally try to capture the collective discourses that are used in the interviews. None of the three approaches aims to mirror an external reality by investigating questions such as what users read, seek for, or how much time they actually spend while doing this. The study of information users has a long tradition within LIS. I have tried to emphasize throughout this article that there is a need for a more diverse understanding of users in the health care context as represented at the presentations at EAHIL conferences. The need to continuously study the use of particular information systems and services and to quantify users’ information behaviour is obvious, but there is also a need to gain a better understanding of what underlies the use, or non-use, of information systems and how individuals create meaning out of their information seeking. There is a need to understand the user from his or her perspective, and to understand users’ information seeking and use from a broader picture than just the users’ interaction with the information system. These are not controversial or radical claims. On the contrary, they are claims that are evident in the broad field of social science today and they are also claims that have attracted attention in contemporary LIS. Thus, qualitative and quantitative approaches to study users ought to complement each other if we are to learn more about users’ information needs, seeking and use.