An agenda for the second generation of qualitative studies

Abstract

An agenda is proposed that calls for investigations of and instruction in the skills and dispositions needed by practitioners of qualitative studies. Qualitative studies re-emerged in the social science and health disciplines in the early years of the 1970s. In the subsequent 30-plus years of the first generation of qualitative studies, there has been a dramatic growth in their use and in attention to their theories and methods. In addition, concerns about the quality of qualitative studies have been expressed. In qualitative studies, researchers themselves serve as the data gathering and analytic instruments. Attention to the development of researcher cognitive and conative skills and researcher virtues is recommended.

• Practice
• qualitative studies
• instruction
• virtues
• agenda

The occasion of the initial issue of this new journal, The International Journal of Qualitative Studies on Health and Well-being—QHW, provides me with an opportunity to suggest future developments for studies whose aim is to describe and clarify experiences that are fundamental to being human.1 A few examples of these experiences in the area of health and well-being are given by Hallberg and Dahlberg (2005) in the guidelines for QHW. Their list of experiences includes “suffering, hope, trust, anxiety, love.” A deepened understanding of these experiences allows greater awareness about and sensitivity to those served by practitioners of care as well as greater self-awareness of those who serve.

The characteristics of what is to be studied underlie the decision of how to study it. When what is to be studied are human phenomena2 that appear and are felt in people's experience, an approach needs to be employed that is able to capture and elucidate such phenomena. As other people's experiences are not directly available to researchers, studies of experiential phenomena depend on people's expressions about those phenomena. Studies that generate or collect people's expressions about an experience and analyze or synthesize those expressions in order to understand and clarify the experience have been classified as qualitative studies. A study on health and well-being is not worthy because it uses a qualitative approach; rather, its worth is manifested if it provides insight and understanding of the phenomenon of interest. The use of a qualitative approach can (but does not necessarily) produce meaningful results. Strict adherence to a particular qualitative method and its set of techniques and sequences does not lead to a deepened understanding in itself.

The development and use of qualitative approaches to the study human experiential phenomena has taken place in the past 35 years across the social science disciplines. Researchers in the area of health and well-being have made use of developments in the approach initiated by other disciplines and have, themselves, significantly contributed to its advances. Qualitative study is not a discipline-specific approach, but one made use of, in varying degrees, by all the human and social sciences. In this article, I will address concerns about the general practice of qualitative approaches and, thereby, hope to contribute to its more effective use in the study of health and well-being.

The QHW guidelines (Hallberg & Dahlberg, 2005) refer to “two serious weaknesses with qualitative studies in general”; one weakness is that there are too few international journals for the publication of its studies; and a second is that “the quality is still, sometimes too poor.” The publication of this journal is a beginning response to the first weakness. In this article, I will suggest a possible response to the second weakness. I will propose that the quality of qualitative studies is a function of the instructional focus used to teach this approach. We have focused on teaching various procedural methods to the neglect of training in the cognitive and conative skills needed for generating rich data and meaningful analysis. I will recommend that the center of attention in our teaching be the proficient and expert practice of qualitative study.

The first generation of qualitative studies

Interviews and participant observation had been used as research data in the early decades of the 1900s; for example, sociological studies using symbolic interaction theory (Meltzer, Petras & Reynolds, 1975) and anthropological based on field studies, its importance and use had waned by the middle decades of the 1900s. By that time, the academy's social sciences had come under sway of logical positivism and had adopted a singular emphasis on the use of numeric data and statistically based analysis as its approach to research. It was into this environment that qualitative studies began to re-emerge in the 1970s. Schwandt (2000) locates the re-emergence of qualitative studies in the academy as the early 1970s. It is a mere three decades—a single generation—since that time, a short period for the refinement of a research approach.

The emergence of qualitative studies

The groundwork for the call for an alternative research approach in the social sciences was prepared in the late 1950s and 1960s by several developments. Voices were raised against the assumptions about people that are implied in statistically based research designs. Critics questioned the deterministic suppositions about human actions that informed behaviorist informed research designs. Anthropologist Clifford Geertz (1955) questioned the notion that researcher observations represented an unbiased, objective point of view. These were joined together in an existential and humanistic movement in psychology which spread to education and sociology (see Wertz, 1994). The critics called for reform in the social sciences that incorporated their ideas and values. The common theme among these voices was that human and social phenomena had unique characteristics that could not be captured adequately by positivistic informed statistical designs. These designs, which dominated the social sciences, had been developed to study the physical realm and could only produce third-person explanations of law governed human activity.

Among these voices calling for reform were those who addressed the need to develop and implement a research approach that reflected their view of the person as a responsible and caring agent. The approach would need to attend to the fine-grained attributes of human experience and activity, to the lifeworlds in which people engage in significant actions, and to the felt-meanings of being-with others. Such an approach would have to focus on human expressions rather than numeric counts. It would be a human science, that is, a science whose methods were particularly designed to understand human beings, rather than one whose methods were designed to study physical objects (for example, Giorgi, 1970a; Polkinghorne, 1983).

Early calls for such methods proposed a renewal of the earlier research approaches developed in anthropology and sociology (for example Filstead, 1970; Willems & Raush, 1969). However, the earlier sociological and anthropological qualitative programs had handed down, in written form, few instructions in how to conduct this kind of research. How to do and teach this kind of research had to be re-created from scratch. Two newly developed research programs—Glaser and Strauss's grounded theory (1967) and Giorgi's phenomenological research (1970b)—had not only produced studies that reflected the concerns of the reform movement, but they had also had taken care to explain how to conduct such research. The texts in which they explained their methods and gave descriptions of how to carry them out came to serve as founding documents in the emergence of qualitative studies. In the first generation, their two methods have retained a prominent place throughout the continued development and practice of qualitative studies.

Since its beginnings in the 1970s, after a period of so-called “paradigm wars,” the reform movement and its qualitative approach to knowledge generation has gained strength in the academies (especially among students). Its accomplishments during its three decades are remarkable. It has had an international and cross-disciplinary impact. Courses in qualitative methods are offered in most university social science departments. Thousands of studies using its qualitative approach have been conducted (although, so far, it appears that the majority of the studies were carried out as master's theses or doctoral dissertations). These studies have served to give insight into and deepen the understanding of the experiences of people served by the practices of care. However, the use and acceptance of the qualitative approach to study of persons has varied in the different disciplines. Perhaps its greatest impact has been in the health care disciplines of nursing and occupational therapy. The discipline of nursing has provided leadership in the advancement and refinement of qualitative study. It has provided qualitative texts (for example, Benner, 1994; Dahlberg, Drew & Nyström, 2001; Speziale & Carpenter, 2003) and a volumes of collected studies (Diekelmann, 2002–2005). The discipline of psychology, of all the social sciences, has, perhaps, been least accepting of the qualitative approach; however, one of the American Psychological Association's journals has recently published a special issue on qualitative research (Haverkamp, Ponterotto & Morrow, 2005).

In the other social science disciplines (education and sociology), the qualitative approach has gained a foothold, but as a separate, if not equal, research approach. Recent emphasis on evidence-based practices has served to widen the distance between qualitative and quantitative approaches to research. Two recent US government reports are examples of the mainstream view of qualitative studies. The National Research Council, which advises the United States federal government on funding research, recently issued the report Scientific Research in Education (Shavelson & Towne, 2002). It placed value on the kind of research that issues claims about cause-and-effect relationships that are generalizable to populations. It recommended that funding should focus on producing the kind of knowledge claims that answer questions about “what works” (p. 108). The report recognized a limited role for claims based on qualitative interviews and document analysis. It stated, “They [qualitative descriptions] cannot (unless combined with other methods) provide estimates of the likelihood that an educational approach might work under other conditions or that they have identified the right underlying causes” (p. 108). The report stated:

We assume that it is possible to describe the physical and social world scientifically so that, for example, multiple observers can agree on what they see. Consequently, we reject the postmodernist school of thought when it posits that social science can never generate objective or trustworthy knowledge. (p. 25)

Grover Whitehurst (2003), Assistant Secretary of Education, proposed that research for the No Child Left Behind legislation should be limited primarily to randomized trial (true experimental) research that produces knowledge claims about what interventions work in education. He proposed that not all evidence is created equal. At the bottom of his list of credible evidence are case studies and anecdotes. He assumed that knowledge which is of value to education is the kind that claims that certain interventions will cause desired effects and that can be generalized across settings.

In general terms, qualitative approaches advocated in the reform movement have settled into a co-existence, if not integration, with the adherents of mainstream statically based research. This separation is manifested in separate journals, conferences, and textbooks devoted to qualitative studies. In spite of its growth, in the US, the reform movement has retained a minority and marginal status in American universities and professional societies.

Focus on method and instruction

The first generation of qualitative researchers focused on two tasks, the construction of an inquiry method and the teaching of that method. The method needed to be consistent with reformist values and had to be able to produce understandings of special characteristics of human activity and experience. Courses, texts, and procedures for teaching this new approach to research needed to be developed. Among others, three factors influenced the manner in which the first generation went about accomplishing the two tasks: (a) the refinement and teaching of qualitative study occurred within an environment in which the quantitative approach was dominant; (b) students needed rules and procedures that, if correctly followed, would assure that their qualitative efforts would successfully pass the judgment of their instructors; and (c) keeping pace with the continuing developments in the philosophy of social science required ongoing revisions and alternations in attempts to solidify the method.

The first generation's development of a qualitative approach to the study of human experience occurred within a context in which positivist ideas about how to create knowledge predominated. Justification and acceptance of a qualitative approach had to attend to and accommodate the concerns of quantitative colleagues and authorities. In addition, most of us attempting to develop a qualitative approach had been first trained as quantitative researchers. Sediments of this background and our understanding about how to generate valid knowledge had an influence on our concerns and written products. In quantitative research, method is primary. Method protects findings from a researcher's personal biases and assures objectivity. The method produces valid results, not the researcher, and any study that has correctly applied the method will have fashioned compelling (if not important) finding, regardless of the person who employed the method. Perhaps it was the carry over from this maxim of quantitative research that efforts to construct the right method for conducting qualitative studies assumed primary importance. In retrospect, one can question whether the efforts given to constructing “the correct method” for conducting a qualitative study was prudent.

A second factor was that, from the beginning years of the reformist movement, many students have been attracted to its ideas and values. They have been persistent to meet their research requirements using an approach that was consistent with these values (and, for some, did not require statistical skills). Thus, it was necessary to begin teaching students how to conduct a qualitative study at the same time that the method itself was under construction. We began living in the house while it was being built. Without sufficient experience in carrying our qualitative studies to refine and consolidate its approach, it was being taught to students. Students, necessarily, have an interest in what steps and procedures they need to follow to be assured of a passable grade. There is often resistance to the notion that, unlike quantitative research, simply correctly performing the steps in an algorithm does not necessarily lead to a worthwhile result. I believe that, in being responsive to the student need for rules, the construction and teaching of qualitative methods has tended to emphasize what steps to follow in place of the creativity and cognitive skills required to transform qualitative data into meaningful findings.

A third factor influencing the manner in which the first generation carried out its two tasks was that it occurred during a period of turmoil in the philosophy of science. As waves of changes appeared, many members of the first generation attempted to re-construct what had been advanced to that point in the method of qualitative study. The re-constructions incorporated these changes either as re-interpretations of earlier methods or as new methodological approaches. Denzin and Lincoln (2000) note that there were five stages in the philosophy of science since 1900: the traditional period (1900s–1940s, the modernist phase (1950s–1970s), the period of blurred genres (1970–mid-1980s), the time of the crisis of representation (mid-1980s–mid-1990s), the postmodern and post-experimental stages (mid-1990s to the present). They propose that further changes will occur in a future stage. The effect of adjustments to these changes was to prevent a congealing of a singular method for qualitative study and to promote a proliferation of various theories and methods.

The goal of constructing a method for doing qualitative studies that served the same role as it did in quantitative research was not achieved by the first generation. Instead of a consensus around a method, the effort ended in an excess of diverse methods. Nor did the effort produce a progressive refinement and improvement over earlier constructions. Newer constructions have not replaced earlier ones; newer methods, such as Foucauldian Discourse Analysis and Critical Theory research, have been added to the repertoire and they have simply taken a place alongside the older ones, such as grounded theory and Giorgi's phenomenological method. Textbooks on qualitative methods differ substantially from those about quantitative methods. On the one hand, quantitative textbooks present a somewhat unified picture of the underlying statistical procedures in which controversies (such as Fisher or Neyman-Pierson statistical models) are ironed over. They portray research as a practice of implementing techniques and rules, which have been worked out by the experts and are no longer in flux. On the other hand, textbooks on qualitative research (for example, Bogdan & Biklen, 2002; Merriam & Associates, 2002; Patton, 2002) and the Handbook of Qualitative Research (Denzin & Lincoln, 2000) describe a wide variety of qualitative research methods and theories without taking a position that one approach is more advanced than another.

Thus, as we stand at the end of our first generation, qualitative studies has become an umbrella term under which are situated diverse research methods influenced by diverse philosophies of sciences. Some studies even display a combination of contradictory perspectives; a practice Lincoln and Guba (2000) term “interbreeding.” “The various paradigms are beginning to ‘interbreed’ such that two theorists previously thought to be in irreconcilable conflict may now appear, under a different theoretic rubric, to be informing one another's arguments” (p. 164). The common notion that brings the diverse methods under a single umbrella is the opposition to the hegemony of mainstream statistical research over the study of the human and social realms and their commitment to the use of human expressions as the primary data source. The current state of qualitative research offers a confusing array of methods using different vocabularies, developed by different disciplines and traditions, and based on different ideas of science. Teaching qualitative studies at this time usually consists of selecting a few methods from under the umbrella and clarifying the meaning of terms used by the different methods.

The two tasks of the first generation were the construction of a method for qualitative studies and the teaching of this method. We were developing the subject matter—the method—that was to be taught as we taught it. As the first task, the development of a method has ended in a somewhat confusing complex of methods; we are left to teaching a rather disorganized array of methods and procedures. That the results of our teaching do not always lead to quality results in the studies undertaken by those we have taught might be expected.

The second generation

The first generation has successfully established space in the social sciences for studies whose purpose is to disclose and deepen the understanding of the personal and social realms that make up the experiential life of human beings. As qualitative study moves into its next three decades, its second generation will have its own tasks to accomplish. I am proposing that it shift its focus from construction of a method or methods that are concerned with epistemologically justification to the study of the cognitive and conative processes researchers must actually engage in to produce meaningful and informative findings.

The first generation, perhaps necessarily, attempted to emulate the mainstream model for knowledge production in which following the method is primary for producing valid or trustworthy results. However, the resulting proliferation of qualitative methods suggests that this mainstream model is not appropriate for generating the kind of knowledge and understanding that is the goal of qualitative studies. The purpose of qualitatively generated knowledge is to disclose and make manifest the shared and personal characteristics of the experiential lives of human beings. This kind of knowledge is similar to the kind of cognitive contributions that Gadamer (1960/1975) attributes to aesthetic works. The validity of aesthetic contributions are not dependent correctly following a method or set of rules, but they demonstrate their validity and worth by their capacity to uncover and bring to attention attributes and dimensions of a phenomenon in a newly appreciated respect.

Generating knowledge that can function in this way is not a mechanical process; rather it is the result of the practices of diligence, creativity, and wise judgment. The parallel between conducting qualitative studies and pushing aesthetic works can be pushed too far (although Stake (1995) writes of the art of the case study, and Lawrence-Lightfoot and Davis (1997) talk of qualitative analysis as similar to the creation of portraits). However, the production of quality qualitative knowledge, like the production of quality aesthetic products, is not a “paint by the numbers” enterprise. Knowing the techniques and procedures in the literature about qualitative methods is helpful, and perhaps necessary, but is not sufficient. The practice of qualitative studies requires more than simply applying techniques.

After a review of eighteen varying theoretical traditions in qualitative inquiry, Patton (2002), in his excellent third edition, reached the following conclusion about the practice of qualitative research.

While these intellectual, philosophical, and theoretical traditions have greatly influenced the debate about the value and legitimacy of qualitative inquiry, it is not necessary, in my opinion to swear allegiance to any single epistemological perspective to use qualitative methods. Indeed, I would go farther (at the risk of being heretical) and suggest that one need not even be concerned about theory. While students writing dissertation and academic scholars will necessarily be concerned with theoretical frameworks and theory generation, there is a very practical side to qualitative methods that simply involve asking open-ended questions of people and observing matters of interest in real-world settings in order to solve problems, improve programs, or develop policies. (p. 136)

Patton's conclusion, with which I agree, is that the quality of the results of a qualitative study does not depend on the strict adherence to a sequence of steps or the application of techniques advocated by a particular method. Rather than method centered, the conduct of a qualitative study is problem centered. That is, producing solutions to show data can be generated and collected that are rich enough to exhibit the detail and depth of the experience under study, and to show meaning can be drawn from these data through an analysis that intensifies its understanding.

Thus, a possible direction for the work of the second generation is to turn its attention to an intensive investigation of the practices involved in a qualitative study. The goal would be to achieve a clarification and extension of our understanding of the cognitive and conative performances and processes involved in the actual practice of conducting a qualitative study. The results of this work will have direct application for teaching and may change the focus of instruction from the presentation of various methods to the development of the requisite skills.

The practice of qualitative study

Practice can be thought of as activities which people perform in order to produce an intended result (Polkinghorne, 2004). For example, the purpose or reason for the practice of engaging in exercise is to achieve or maintain good health. Practices are goal directed and, thus, have a means-end or instrumental structure. The value of a practice activity is a function of how well it contributes to achieving an intended outcome.

Qualitative study is an activity whose intended goal is the production and communication of an insightful and disclosing understanding of a human phenomenon. Qualitative researchers engage in those activities that they believe will bring about the accomplishment of this goal as it relates to the particular phenomenon they are studying. Once a goal is set, the practice questions focus on which activities will lead to achieving it; that is, what needs to be done (and in what order) to bring about the desired result.

Practical knowledge is knowledge concerning what to do to attain a particular desired outcome. People attain practical knowledge from various sources. For example, they may develop practical knowledge through trial and error in which learning what to do is derived from retaining those past actions which were successful in bringing about a particular desired outcome and discarding those which were unsuccessful. Others may come to practical knowledge by applying their background knowledge of what actions were successful in accomplishing similar goals. They may consult acquaintances to learn what worked for them in bringing about a goal. At a more formal level, they may consult texts or receive instruction in knowledge about what to do developed by experts and authorities is passed on to them.

For qualitative researchers, the practical knowledge about what to do to produce a discerning understanding of a human phenomenon is, in the main, initially transmitted to students through textbooks and formal instruction. (They may add to and refine their initial practical knowledge through their own experiences of conducting successful studies.) The information about what to do to achieve the goal of a qualitative study is drawn from collection of methods devised in the first generation. A method consists of a set of activities that are thought to guide a practitioner to the achievement of a desired outcome. In order to produce an insightful and trustworthy qualitative study outcome, students are instructed to engage in activities such as: selecting appropriate participants for interviewing, interview in such a way that rich and deep descriptions are produced; break the data into meaning units and code it into meaningful concepts; combine these first level concepts into higher order concepts; locate a central concept that structures or relates the higher order concepts into a unifying understanding; do member checks; and have an expert audit the analysis.

I have two concerns about the practice knowledge we pass on about what to do to achieve a quality outcome in a qualitative study. First, some of the goal producing activities are under-explained in the methods literature and involve the use of highly developed skills. For example, advice to engage the activity that produces rich data needs to be supplemented with training and skill development in order to achieve the desired rich data, and the analytic processes of categorization and identifying concepts is an under understood practice. Recent exploration in cognitive science and semantics has enhanced the understanding of the formation, function, and nature of concepts (see, Fauconnier, 2002, No. 3533; Margolis, 1999, No. 3550; Murphy, 2002, No. 3534). Second, the achievement of desired outcomes in the human realm differs from achieving results in the physical realm. The remainder of this section will address this second concern.

The stability and lawfulness of the physical realm allows a greater confidence in the prediction that if certain activities are undertaken, then the intended result will necessarily occur. For example, if my goal is to have the water boil in my pot, then I can have assurance that the goal will be achieved if I heat the water to 212° Fahrenheit. If my goal is to italicize a word in this text, if I mark the word and press the control and “I” keys, then the goal will be accomplished. However, if my goal is to make my friend happy, or to understand how he feels, I have to take into account the context. What I need to do to achieve this result will depend on my reading of his mood and needs at this particular time and situation. Accomplishment of such goals often involves more than a single action on my part, typically a course of actions. Thus, deliberation on what to do next will involve sensitivity to his response to a prior action. Simply mechanically following through on a set of previously determined series of actions would not bring about the intended consequence.

As qualitative studies take place in and are about the human realm, the methods describing what to do in these studies need to be understood as general, often helpful, guidelines to be used in deliberations about the actions to take in the course of a study. Taking them as equivalent to the kind of practical knowledge useful for achieving results in the physical realm is to misunderstand them. Qualitative studies involve interactions with people and with their expressions. Each qualitative study takes place in its own situation and with its own participants. Reaching the goal of understanding the phenomenon under investigation requires researcher flexibility and wise judgment. What to do next in an interview or in an analysis needs to be judged based on whether at this time and in these circumstances it is leading the accomplishment of the goal.

Levels of practitioner proficiency

Understanding the inquiry activities described in a qualitative method as a set of rules about what to do to achieve a viable and contributing study is a hindrance to the actual accomplishment of that achievement. Dreyfus and Dreyfus (1986) have distinguished different levels of the attitude of practitioners to the role of rules for accomplish their tasks. They located five levels of practitioners’ obedience to rules and five corresponding levels proficiency in accomplishing their goals. Benner (1984) examined the performances of nurses and found that Dreyfus and Dreyfus’ five levels served to differentiate the levels of attitude toward and implementation of rules and levels of performance. The first four levels represent progressive stages in sophistication in the use of the application of established rules and procedures. Practitioners at the fifth level, experts, make use of situated judgment in their performances. I am proposing that, like Benner's findings that a correlation may hold between the quality of findings in a qualitative study and the researcher's obedience to the rules of a method.

The five levels differentiated by Dreyfus and Dreyfus are as follows:

1. Novice

The first level is the novice practitioner, whose practical knowledge is limited to a set of context-free rules. The rules specify what is to be done in a given set of conditions; that is, a certain rule is to be applied no matter what else is happening. The rules are treated as universal statements and are applicable regardless of the time and place in which they are carried out. Dreyfus and Dreyfus cite as an example of the novice practitioner a beginning automobile driver. The novice driver knows at what speed to shift gears and, at any given speed, what distance they are to follow a preceding car. However, in deciding what to do, the novice ignores such contextual conditions as traffic density and anticipated stops. Novice qualitative researchers are inflexible in their practical actions, seeking out and treating methodological rules as decontextualized orders to direct their actions.

2. Advanced beginner

At this level, practitioners have gained some experience in actual situations. They have learned to perceive similarities between new and prior situations and they are able to recognize overall characteristics of potential events. Advanced beginners, however, cannot yet reliably sort out the most important elements of complex situations. Rather than depending exclusively on rules, they respond directly based on what they did previously in similar situations. Their understanding of what they should do is based primarily on what one ought to do in a specific kind of instance.

3. Competent performer

Competent practitioners move further away from the automatic application of rules that call for a particular action when confronted with a specific kind of situation. Their practical knowledge about deciding what to do has come to include the awareness of the consequences of various possible responses. Whereas novices and advanced beginners deem their actions correct if they have applied the appropriate rule, competent practitioners assess the consequences of their actions. They understand that the importance of facts is dependent on the presence of other facts and that the meaning of a set of givens depends on circumstances. Nevertheless, their practical knowledge is organized into sequences that require the step-by-step process of assessment, goal setting, and consideration of possible alternative actions.

4. Proficient performer

At this level of performance, practical knowledge has matured to the point that it consists of whole patterns rather than simply a set of general rules or a sequence of problem-solving steps. They perceive the meaning of a situation in terms of long-term goals. When certain cues activate this level's performers, they recognize a situation has a pattern that is similar to another situation. Their response is no longer a sequence of separate actions but a pattern of integrated actions. Access to their patterned and holistic practical knowledge comes, not from a deliberative search for the appropriate pattern, but from direct awareness. Proficient performers are able to pick out salient aspects of a situation that identify it as requiring different actions that would normally be the case.

5. Expert practitioners

Expert practitioners apply knowledge in the form of patterns. These patterns, like perceptual gestalts, cannot be separated into its parts, for the meanings of particular actions vary according to the role they play within the whole performance. They understand that practical knowledge does not consist of a compilation of independent universal rules. Rather, actions achieve meaning and move a situation toward its goal in a total configuration. In the practicing knowledge of experts, mental designs are revised and adjusted in the light of professional experiences and reflective thought. An expert practitioner usually gains access to his or her knowledge in a non-reflective manner, but when some aspect of the situation is inconsistent with the assumed pattern, he or she deliberates before acting (Schön, 1983). Such reflection, however, is not the same as the calculative practical reasoning of the competent performer, for it requires the formation or the recognition of a new pattern that gives meaning to their actions in a new situation.

Dreyfus and Dreyfus (1977) described the move expert practice:

As long as the beginner pilot, language learner, chess player, or driver is following rules, his performance is halting, rigid, and mediocre. But with mastery of the activity comes the transformation of the skill which is like the transformation that occurs when a blind person learns to use a cane. The beginner feels pressure in the palm of the hand which can be used to detect the presence of distant objects such as curbs. But with mastery the blind person no longer feels pressure in the palm of the hand, but simply feels the curb. The can has become an extension of the body. (p. 12)

The production of qualitative studies of high quality necessitates its practitioners function not as rule followers, but as masters. If Dreyfus and Dreyfus’ findings are transferable to practitioners of qualitative study, then it can be expected that higher quality studies will require less obedience to rules and more sensitivity to a study's gestalt.

In Dreyfus and Dreyfus’ model, advancement to higher levels of practice proficiency is linked to years of experience. They suggest that it takes five years of engagement in a type of task, before the expert level can be reached. However, experience alone is not sufficient to attain mastery. Reflection on the consequences of a practitioner's actions, engagement in diverse experiences, attention to the uniqueness as well as similarity among experiences, and a willingness to be open to experiment with alternative actions.

A relationship between level of proficiency and adherence to rules presents a problem for university instruction in qualitative study. Often students take only a single course in qualitative research before conducting a study for a master's thesis or doctoral dissertation. On the one hand, because of the limited time available for instruction, we can attend to turning out first-class novices who have several sets of methodological rules that they can follow to get through the production of a study. However, this tack will likely result in mediocre studies. On the other hand, we could focus on the beginning development of interviewing and analytic skills and the practical problems involved in conducting a study; however, students may be left without an understanding of the methodological purpose for the skills and the variety of theoretical traditions under the qualitative umbrella. Of course, a combination of these tasks, given sufficient time, is preferable.

In my experience, our textbooks and courses have tended to emphasize the first approach. I suggest that emphasizing the second approach will raise the level of practice of students beyond the novice level. Students learn what tasks to do from our current methods, but not how to accomplish these tasks. For example, they are told to do interviews that produce detailed and thick descriptions. However, accomplishing this task beyond a novice level requires practice and an understanding the issues in transforming experiences into language. They are told to gather individual units of text under concepts, but this is a complex task to master. Knowledge about the kinds, characteristics, and structure of concepts and conceptualization is often lacking. Pulling off these tasks with sophistication and insight requires judgments approaching expert levels. I propose that we incorporate into our training insights and advances from other disciplines (semantics, philosophy of language, literary criticism, philosophy of mind, and cognitive science) that have focused on the characteristics and analysis of human expressions.

It may be that we have underestimated how difficult it is to produce quality results in a qualitative study. Doctoral students in my department are required to complete at least five courses in quantitative procedures. Those who choose to conduct a quantitative study do additional specialized study in the statistic they will use in their dissertation. Yet, students who choose to conduct a qualitative study, in addition to the five quantitative courses, take the one qualitative course we offer. In some other university programs, students undertake qualitative studies with even less preparation and under supervision of faculty who are not trained in qualitative research.

The epistemic virtues and qualitative study

One of the reasons that qualitative studies are difficult is that their processes are centered on the skills and character of the individual conducting the study. The data creation is an accomplishment of the person doing the study, not a number-generating questionnaire. The analysis is a result of personal (or group) judgments, not a statistical computer program. The strength of a qualitative study is that its findings depend on the diligence and judgments of a researcher, not on adherence to a method. The person is the research tool.

Recent efforts in philosophical studies of epistemology have turned attention to the attributes or virtues practiced by researchers (see, Axtell, 2000; DePaul & Zagzebski, 2002; Fairweather & Zagzebski, 2001; Steup, 2000). In place of concentrating on formal and procedural rules and truth conditions that would produce reliable knowledge, they are studying what are called knowledge-conducive virtues. They draw on Aristotle's discussion of the virtues, which are involved in making sound judgments in moral deliberations. They propose there are analogous epistemic virtues that “guide the use we make of our skills and capacities in carrying out enquiries effectively” (Hookway, 2003, p. 91).

In his presentation of virtue-based epistemology, Norris (2005) says that it holds that “there are certain distinctive virtues—among them honesty, integrity, caution, openness to criticism, willingness to give up cherished beliefs in the face of conflicting evidence—that are knowledge-conducive in so far as they characterize competent, responsible, and well-disposed epistemic agents” (p. 129). These virtues are basic and prerequisite in the process of knowledge acquisition. It is not enough simply to make judgments in carrying out a qualitative study; the judgments need to be made responsibly. Cultivation of dispositions containing the epistemic virtues is particularly important for qualitative researchers because their findings are a consequence of those judgments.

A responsible qualitative practitioner is guided by the epistemic virtues. They are responsible “for their claims, that is to say, put them forward only on condition that (1) they have been arrived at through a duly reliable process of belief formation; and (2) they have been subject to an adequate degree of intellectual and critical self-scrutiny” (Norris, 2005, p. 163). This means that they give to their readers an adequately argued account of the process they used to arrive at that particular judgment. Qualitative researchers need to not only present their findings, but they have “an obligation to monitor and report their own analytical procedures and processes as fully and truthfully as possible” (Patton, 2002, p. 434). The trustworthiness of the findings of a qualitative study is not judged by its conformity to a method or set of procedures. Instead, trustworthiness is a status given by a reader who is convinced that the researcher made responsible judgments and exercised care in the production of the study.

Much has been accomplished in the first generation of the re-emergence of qualitative studies. It has become clear that in the thirty-plus years since its re-emergence qualitative studies have made a major contribution in increasing our understanding of the lives of people as they live them. It is being employed by scholars across the world in all the disciplines that are concerned with the human and social realms; but it is an unfinished product. I have proposed an agenda whose purpose is to continue its development and growth as the primary vehicle for understanding the human world.

We need to know more about the intricacies and judgments that make up the skills and dispositions used in conducting qualitative studies. We also need to translate this knowledge into instructional practices so that they can conduct studies that result in quality findings. I have proposed that the second generation focus its attention on the development on the personal skillfulness and proficiencies of the researchers who engage in qualitative studies. These personal characteristics are primary in an approach to studies in which researchers themselves are the instruments producing increased understanding of a human phenomenon. The first generation of the re-emergence has succeeded beyond what could have been anticipated. However, we are only at the beginning stage of the refinement of and sophistication in the subtleties entailed in conducting this kind of inquiry. Such is the work to be accomplished in the second generation of qualitative study.

Notes

1. otes¹ A significant contribution of the journal is its international perspective. Although I have been a contributing member to qualitative study since its re-emergence in the 1970s, I am aware that my perspective is parochial. The content in this paper is informed by my own by particular experiences. I am a psychologist and for the past fifteen years have taught in a School of Education in Southern California. My knowledge of what has occurred internationally during the re-emergence of qualitative studies, is restricted; my readings, discussions with colleagues, and visits to several European countries. I expect that the description of the development of qualitative studies by someone from another country and academic discipline would differ from mine

2. The subject matter investigated by qualitative studies does not appear as publicly observable objects. I have chosen to use the term phenomenon to refer to the instances of the subject matter that serves as the focus of qualitative studies. I am using phenomenon in an ontologically neutral sense and am not taking a stance of the nature of their reality. I am also using the term phenomenon to include social and cultural items where access to them is through their experienced presence in people's experience. I am also including descriptions of patterns of behavior as well as structures and instances of experience as they appear across individuals and as they appear uniquely in a particular individual's life. Other terms, such as experiential objects, beings, entities, seem to me to be even more problematic