Christina Maslach, Ph.D.
The Maslach Burnout Inventory
|I am a professor of psychology at the University of
California, Berkeley, and have been on the faculty here since 1971"the year I
received my Ph.D. in psychology from Stanford University. In terms of my background, I was
born in San Francisco in 1946 and went to Harvard-Radcliffe College for my undergraduate
education. I graduated in 1967, magna cum laude, having majored in Social Relations (an
interdisciplinary major in the social sciences). I then spent the next four years at
Stanford University for my graduate training in psychology.
My interest in burnout developed out of my interest in emotion and in the general question of how people "know" what they are feeling. I had been doing some experimental research on how people interpret an ambiguous arousal state"what cues do they use and what information do they seek to make sense of this uncertain feeling. As I continued to do work in this area, I became interested in a somewhat different question, which was inspired by the then-ongoing research on misattribution processes. Suppose people experienced a strong arousal state (rather than an ambiguous one): Could similar cognitive processes be used to relabel this arousal and thus change its meaning and its experiential impact? More specifically, was such cognitive relabeling a means of reducing the intensity of the arousal and thus preventing it from disrupting necessary, ongoing behavior? Such a question would be especially important for people who have to function calmly and efficiently in situations that are often characterized by crisis and chaos. For example, rescue personnel (such as police or firefighters) have to deal with such situations, as do staff of hospital emergency rooms and therapists doing crisis counseling.
|My initial review of psychological theories and
constructs found little that addressed this issue directly. However, there were two
concepts from the medical literature that seemed to be relevant. One of these was
"detached concern," which referred to the medical professions ideal of
blending compassion with emotional distance and a more detached objectivity. The second
relevant concept was "dehumanization in self-defense," which referred to the
process of protecting oneself from overwhelming emotional feelings by responding to other
people more as objects than as persons. Both of these concepts seemed to shed some
theoretical light on the issue of how people cope with strong emotional arousal.
Given that the two guiding concepts had their origins in the medical professions, my first step was to interview people who were working in health care settings. Some key themes emerged from these interviews. First, it became clear that emotional experiences played an important role in the provision of health care. Some of these experiences were enormously rewarding and uplifting, as when patients recovered because of the practitioners efforts. However, other experiences were emotionally stressful for the practitioner, such as working with difficult or unpleasant patients, having to give "bad news" to patients or their families, dealing with patient deaths, or having conflicts with co-workers or supervisors. Such emotional strains were sometimes overwhelming, and practitioners talked about being emotionally exhausted and drained of all feeling.
|A second theme was that "detached concern"
was often more an impossible ideal than an attainable reality. Although practitioners
would try to distance and detach themselves from sources of emotional strain, they found
it difficult to do so and still maintain concern. A more typical pattern was a negative
shift over time, in terms of practitioners perceptions and feelings about their
patients; in extreme cases, they began to dislike, and even despise, them. It was this
phenomenon that gradually became the topic of interest, rather than the original notion of
A third general theme had to do with the self-assessment of professional competence. All too often, the experience of emotional turmoil was interpreted as a failure to "be professional" (i.e., nonemotional, cool, objective) and led people to question their ability to work in a health career. Many practitioners felt that their formal training had not prepared them for the emotional reality of their work and its subsequent impact on their personal functioning.
Up to this point, my thinking about this phenomenon had been framed within the context of health care. However, my focus was broadened as the result of a chance event. I happened to describe the results of my health interviews to an attorney, who told me that a similar phenomenon, called "burnout," occurred among poverty lawyers working in legal services. Not only did I learn that the phenomenon I was studying had a name, but I learned that it was present in a wider range of occupations. What seemed to link poverty law and health care was the focus on providing aid and service to people in need"in other words, the core of the job was the relationship between provider and recipient. The implication was that working with other people, particularly in a caregiving relationship, was at the heart of the burnout phenomenon. Thus, I began to study providers in other types of "people-work" occupations (which included mental health, social services, counseling, education, and criminal justice). The evidence of a parallel pattern suggested that burnout was not just some idiosyncratic response to stress, but was a syndrome with some identifiable regularities.
At this stage in the research, the key issues were to develop a more precise definition of burnout and to develop a standardized measure of it. My graduate student, Susan Jackson,
|and I spent the next few years conducting an extensive
program of psychometric research. We collected systematic data from hundreds of people in
a wide range of health, social service, and teaching occupations. On the basis of this
psychometric research, we developed a measure called the Maslach Burnout Inventory (MBI).
This measure was designed to assess the three components of the burnout syndrome:
emotional exhaustion, depersonalization, and reduced personal accomplishment. There are 22
items written in the form of statements about personal feelings or attitudes, and these
are answered in terms of the frequency (on a 7-point scale) with which the respondent
Despite all of our empirical data, the initial reaction of the academic world to our new test was somewhat negative. The popular, nonscientific terminology of "burnout" led some people to immediately dismiss the concept as a fad or as pseudoscientific jargon that was all surface flash and no substance. Indeed, our psychometric article on the development of the MBI was returned by some journal editors with a short note that it had not even been read "because we do not publish pop psychology." Eventually, however, the research was published (Maslach & Jackson, 1981a, 1981b), and the availability of a standardized measure fostered much subsequent research on burnout (see Schaufeli, Maslach, & Marek, 1993).
The MBI is now recognized as the measure of choice for burnout, and it is used by researchers around the world in various translations. Three versions of the MBI are now available (Maslach, Jackson, & Leiter, 1996): the original version for use with professionals in the human services (MBI-HSS), an adaptation for use with educators (MBI-ES), and a new, 16-item general survey designed for use with workers in other occupations (MBI-GS). All forms of the MBI provide scores on the three dimensions of burnout, and these scores can be correlated with other information obtained from respondents, such as job characteristics, job performance, personality or attitude measures, health information, and demographic variables. The factors that best predict MBI scores, and the outcomes that are best predicted by MBI scores, can be assessed by multiple regression techniques and structural equation modeling.
|The multidimensional model underlying the MBI has made
it particularly appropriate for theory-driven research. New studies are pointing to the
causal significance of six areas of mismatch between the worker and the workplace, as well
as to the importance of studying the positive opposite of burnout, namely engagement with
work (Maslach & Leiter, 1997).
Looking back on the past two decades, I see both pros and cons of having been a test developer. On the plus side, I was one of the early "pioneers" whose work on defining the burnout concept and developing a measure to assess it has shaped the entire field. But this achievement was a long time in coming, and it took an enormous amount of painstaking research to develop a high-quality measure that could pass innumerable tests and replications by many international researchers. I developed the MBI because there was a need for such a research tool to study interesting questions about burnout, and not because I was particularly interested in psychometric research per se. However, the role of test developer has made me the reference point for psychometric work in this field, in addition to my roles as theorist and researcher.
Maslach, C., & Jackson, S. E. (1981a). The measurement of experienced burnout. Journal of Occupational Behaviour, 2, 99"113.
Maslach, C., & Jackson, S. E. (1981b). The Maslach Burnout Inventory (Research edition). Palo Alto, CA: Consulting Psychologists Press.
Maslach, C., Jackson, S. E., & Leiter, M. P. (1996). The Maslach Burnout Inventory (3rd ed.). Palo Alto, CA: Consulting Psychologists Press.
Maslach, C., & Leiter, M. P. (1997). The truth about burnout. (3rd ed.). San Francisco, CA: Jossey-Bass.
Schaufeli, W. B., Maslach, C., & Marek, T. (Eds.). (1993). Professional burnout: Recent developments in theory and research. Washington, DC: Taylor & Francis.
The McGraw-Hill Companies, Inc..