Praxis

1. Introduction

[1.1] When I began my work in fan studies in the summer of 1988, I faced discouragement from all quarters. The advisors at my university did not think looking at the influence of aspects of popular culture on development constituted a serious study. Some of my advisors made it clear to me that the psychology of fan behavior was a trivial thing to study (the word "trivial" was actually written on an application for graduate student funding that I submitted). If it had not been for my thesis chair, Alan Brown, I would not have been able to do the work I did at all. I was at the very beginning of my academic career, a master's student working on a thesis, and I was brand new to all aspects of research and statistics. In addition, when I surveyed the literature in psychology to explore current work, I found next to nothing, suggesting that perhaps my advisors were not the only ones in the field of psychology who felt this way.

[1.2] This paper can be conceptualized as an autoethnography (Ellis, Adams, and Bochner 2011). In explaining this concept, the authors observe, "As a method, autoethnography combines characteristics of autobiography and ethnography. When writing an autobiography, an author retroactively and selectively writes about past experiences. Usually, the author does not live through these experiences solely to make them part of a published document; rather, these experiences are assembled using hindsight (¶5)". This statement describes exactly the process that was used in writing this paper. I have searched through my past publications, data, and other records of my past work in order to assemble a narrative that explains thirty years of methods that I have used in my fan studies. Additionally, again as explained by Ellis, Adams, and Bochner, I have framed my reflections and recollections in the context of the data I have collected and the analyses, both quantitative and qualitative, that I have performed.

2. Acafan or metafan?

[2.1] When considering methodologies in fan studies, I have realized for some time that the way I approached it when I began my work back in 1988 was not typical. First, I had a question I wanted to answer rather than a specific fandom I wanted to study. As a psychologist, I wanted to understand the roots of superstar mania as exemplified by the Beatles, Elvis, Frank Sinatra, and others I had been exposed to when growing up (I was born in 1954). Therefore, I surveyed the culture of that time and picked the best current exemplar of this phenomenon that I could find, and that was the fandom of Michael Jackson. When scholars study fans, they most often choose a fan group of which they are already a member; they are referred to as acafans (Duffett 2013). In my case, I knew next to nothing about Michael Jackson. I had seen a television special (of the fall 1987 Bad tour in Japan) where superstar mania was evident, and I decided to study his fan base.

[2.2] During a presentation I gave at the Eastern Communication Association in 2010, where I was explaining my research interests, the acafan/metafan issue became clearer to me. I had just given a presentation on the Michael Jackson fandom I had studied as a case of preinternet fandom, compared to the Josh Groban fandom as a case of postinternet fandom. I was trying to explain that I was much more interested in going to fan events to see the fans than to see the particular artist performing at that event. The session chair had an "aha" reaction, an epiphany of sorts, and said, "Oh! You're a metafan!" It was the perfect way to describe the fact that I was, indeed, a fan of fans, so to speak.

[2.3] Each group I have studied represents something that I have developed a passion for in order to engage fully with my participants, many of whom have become dear friends as a result of the length of the studies and the traveling we often did together. But I have always known in my heart of hearts that I was not a true fan because when it was time to move on to the next case study, I never had any trouble leaving behind the one with which I had finished. I talked recently to a woman I had met during the Michael Jackson fandom study and with whom I have not spoken for some years (perhaps as many as twenty?). She had a difficult time believing that I was not a fan in the same way that she is a fan (she still follows him after all this time). I had to point out to her that after moving on from the study, I learned next to nothing about Jackson and his career from the time of about 1992 to the present. When Jackson died in 2009, I had a brief time of interest again as many fans who had known me during the study contacted me to share their grief. I now have about a dozen Michael Jackson fans on my Facebook page with whom I had lost contact from 1992 to 2009. However, interest in a specific fandom has not ever been sustainable for me. My fascination has always been with the fans.

[2.4] I could tell the same story for the other fan case studies I have done, on the fandoms of Paul McCartney, Madonna, Star Trek, Lord of the Rings, Josh Groban. It could be argued that because I was not a true fan, I could not have understood the fan experience. Again, I point to the conversation I had recently with the Michael Jackson fan. After twenty years, she had forgotten that I was doing research, forgotten the questionnaires and the interviews and all that she knew I was doing at the time, as I have never been covert about my research. She just remembered me as a fan just like her. This is an indication I had succeeded in my goal of full participation.

[2.5] One way that I have connected with various fandoms and become a full and engaged participant has been through charity work. During the time I spent in Star Trek: Deep Space Nine (1993–1999) fandom, I became a coordinator of charity events for the various fan clubs and actors. In the years from 1994 until 2003, the groups raised over $350,000 for the actors' charities, and I was in charge of and coordinated most of that. In the Josh Groban fandom, I became very involved in the Josh Groban Foundation, Grobanites for Charity, and later the Find Your Light Foundation. I coordinated a number of gatherings before various concerts on the different tours and helped raise a significant amount of money for those causes. Indeed, I organized the charity event in 2009 before the Hollywood Bowl concert where Groban was honored as part of the Hollywood Bowl Hall of Fame. We raised $22,000 that night for his foundation.

[2.6] I believe that in order to do authentic participant-observer research, one has to find a way truly to engage with the fandom at a grassroots level with full participation. It means becoming a fan for that time and doing all of the things that fans do. I have done this for the past thirty years. In order to get at the heart of participant meanings in such a social setting, one has to be all in. In one case, the Paul McCartney fandom, I was not interested in or committed to his solo career but was relying on a childhood interest in the Beatles. One of the group members confronted me and said, "You're not one of us," and she was right. She was kind about it, but her comment made clear to me the importance of fully embracing the mission, once it was undertaken. That never happened to me again as it was a mistake from which I learned and did not repeat.

[2.7] One of the benefits of this approach to case studies on fandom was that after I had done a number of them, patterns began to emerge, and overall observations began to guide my writing on the subject. While it is true that a case study is not totally generalizable to a broader set of fan groups, when one is involved in the ninth or tenth group and recognizes that this is yet another example of something observed from the very beginning, the insights gained are very valuable. An example will illustrate this point.

[2.8] Numerous fan studies writers have talked about the ways that fans can meet celebrities. As Hills (2016) and Ferris and Harris (2011) described fan-celebrity interactions, one major type seems to be left out of their classification schemas. They enumerated the prestaged encounter, usually commercial, the fan-staged encounter, which sounds very much like stalking, and the unstaged or random encounter. The category left out is the informal meeting that is arranged by the celebrity in cooperation with fans. These meetings are not commercial in that no money changes hands. They might be associated with a commercial event but are not part of the official event. When I heard about such meetings during the Michael Jackson fan study, I thought they were an idiosyncrasy of his particular fan group. But thirty years later, having seen these meetings in every single fan group studied, I know now that they are a part of the everyday life of a fandom (Stever 2016). Additionally, Matt Hills stated that "fans are required to respect the celebrity's privacy" (466) and stick to a limited range of conversations. Other scholars have made similar statements; for example, Marwick and boyd (2011) agreed that autograph signings and other opportunities to meet celebrities are "highly managed and limited in scope" (149). I frequently observed in both Star Trek and Josh Groban fandoms that opportunities to have genuine conversations with favorite celebrities are quite commonplace. Important to the current discussion is the fact that I would never have known about this kind of encounter had I not been a participating member in the various fandoms.

[2.9] Ethical considerations were always in the forefront of choices I was making. I was committed to carefully preserving the anonymity of my participants. I also took all chances to debrief fans after having them fill out surveys or give me interviews. If they filled out a Myers-Briggs Type Indicator (MBTI), I was careful to interpret the results with participants and to answer any questions they had. I went through the procedures necessary to become a qualified administrator of the MBTI and familiarized myself with any issues that might arise with helping participants understand their results. Today, self-administered personality inventories are easily found on the internet, but in 1988, the only way to take such an instrument was from a qualified person. I took that responsibility very seriously. My studies were all reviewed and approved by institutional human subjects review committees. I have never solicited participants on the internet. Participants through the mail and at events were given informed consent statements. They were included on all surveys and questionnaires, for example: "This survey is part of research being conducted by Gayle Stever at Institution. Your participation is completely voluntary and if at any point you want to stop participating, you may do so. Your answers are completely confidential and will be kept without identifiers. All data will be reported only in groups of data so that no one individual's replies will be distinct. If you have any questions, please feel free to ask."

3. Disciplinary approaches to fan studies

[3.1] In order to build a literature review for my thesis in 1988, I had to go outside of psychology. The work I summarized came principally from cultural studies, media studies, anthropology, sociology, and communication. I surveyed the academic literature in my field and in the related social sciences and discovered that psychologists had written very little by that time. Featuring prominently in that first literature review were multiple articles by popular culture scholars Browne and Fishwick (1983), as well as studies by sociologists like Kortaba (1989) and Snow (1987). Anthropologist John Caughey (1978, 1984) featured prominently in the earliest works that I cited.

[3.2] When Haspel (2006) did her doctoral dissertation summarizing work in psychology in the area of fan studies, she recognized that my Celebrity Appeal Questionnaire was one of the first publications in the field in a psychology journal: "Stever expands the concept of spectator/viewer to encompass fans…Significantly, Stever is one of the first researchers to suggest that psychologists should study para-social relationships" (16).

[3.3] One of the things I have encountered through the years has been the differing attitudes toward fans from different disciplines. One of the foundational theories in fan studies is parasocial theory (Stever 2017). While parasocial interaction (PSI) (Horton and Wohl 1956) was first proposed in Psychiatry, the work did not really take off in a major way until the 1970s, and then it was done by communication scholars and not by psychologists (Levy 1979; Rosengren and Windahl 1971). The emphasis in the early literature on PSI and parasocial relationships (PSR) was on consumer behavior with an attempt to identify predictors of PSI and PSR (Giles 2002). Some of the factors looked at were loneliness, gender, and age, although the research expanded to include other things as well (Canary and Spitzberg 1993; Rubin, Perse, and Powell 1985; Rubin and McHugh 1987; Wang, Fink, and Cai 2008).

[3.4] In the early years of the twenty-first century, psychologists reentered the discussion in a major way with the celebrity worship literature (Maltby et al. 2005; Maltby, Houran, and McCutcheon 2003; McCutcheon, Lange, and Houran 2002). Celebrity worship is a very specific kind of PSR, but in the early days of this literature, the terms celebrity worshipper and fan were used as if they were synonyms. This resulted in some justifiable indignation on the part of numerous researchers in fandom who took exception to what they perceived to be the pathologizing of fandom (Jenkins 2014; Larsen and Zubernis 2011). What resulted was the literature breaking apart into two separate pathways: one discussing normal fan practice by both communication and also cultural studies scholars (e.g. Duffett 2013; Sandvoss 2005) and one looking at the fringe of fandom who showed signs of mental illness, this by sociologists and psychologists.

[3.5] As a psychologist who was interested in normal fan behavior and PSRs, I again found myself out of step with my discipline and its emphasis on pathology. In addition to the celebrity worship literature, there were studies being done on erotomania (Houran, Navik, and Zerrusen 2005; McCutcheon et al. 2003) and stalking (Ferris 2005, 2001). While these are each real and legitimate topics, the overgeneralization of these things to all fans created a divide in the world of fan studies.

[3.6] In order to provide support for the hypothesis that these theories did not represent the mainstream of fan cultures, I collected data in 2008–2009 (Stever 2011b) where I took samples of behaviorally identified fans in both Star Trek and Josh Groban fandoms. In each case, the participants were a purposive sample of people who had a strong commitment to their fandoms. By showing that the percentage of people in each category of celebrity worship for the participants from fan groups was the same as from the general community samples used by the celebrity worship scholars, it was quite easy to show that being a fan and being a celebrity worshipper were two distinct and different things.

[3.7] While communication scholars treat the term parasocial as a descriptor of mainstream behavior, some cultural studies scholars (Duffett 2013) have taken exception to the term and interpreted it as meaning something pathological. I believe that this is a result of that early celebrity worship literature and the problems with it. The term parasocial was never meant to infer pathology and more recent usage (e.g. Klimmt, Hartmann, and Schramm 2006) bears that out. It is important for the various disciplines to come to a common understanding of these concepts in order to be able to build on each other's work. This has been a longtime professional goal of mine, to attempt to mend the breach and bring these various disciplines together under one umbrella of fan studies.

[3.8] Parasocial theory was originally developed to explain the social relationships and interactions we have with distant others not met face-to-face. Caughey (1984) used the term imaginary social relationships to describe the same phenomenon, an interaction that is carried on vicariously within the thoughts and introspections of the individual without reciprocation. The term secondary attachment was used in developmental psychology to have an equivalent meaning (Adams-Price and Greene 1990). The viewer has an internal representation of the media figure and interacts with that imagined other. Here are three disciplines—Anthropology, developmental psychology, and communication—each with a term for the same phenomenon (Stever 2017). As research has progressed, parasocial appears to have become the term of choice, with PSI, PSR, and parasocial attachment being three levels of parasocial engagement (Tukachinsky and Stever 2018). Since anyone who consumes media engages in parasocial interaction, the idea that it in any way indicates pathology is an impossible construct to support.

4. Procedures

[4.1] There were steps I followed in each of my fandom case studies. First, I chose a fan study case that was potentially interesting to me. Knowing the amount of time I would spend in the fandom, I recognized that trying to study something completely antithetical to my own personality and interests probably was not going to be very productive. In my original round of data collection, using surveys, field notes, interviews, and observations, I focused on first Michael Jackson but then also Bruce Springsteen, Paul McCartney, Madonna, Janet Jackson, Prince, and George Michael. All of these artists were active at that time in either touring or producing new material, or making public appearances, or all of these. In 1991, I began to shift my focus to Star Trek and other science fiction and fantasy fandoms. In 2005, I took on the case of Josh Groban fans, while continuing my work in Star Trek fandom.

[4.2] The second step was to come up to speed on the relevant content of the specific fandom. This meant learning about the music, films, programs, or other content produced by the artist, as well as learning about the kind of events where fans were likely to be found. By the mid-1990s, included in this stage of research was a monitoring of fan web sites and discussion boards on the internet.

[4.3] The third step was to figure out what data I wanted to collect. This resulted in my construction of the Celebrity Appeal Questionnaire (Stever 1991a, 2008), a survey I administered at fifteen Michael Jackson concerts, six Madonna concerts, two Janet Jackson concerts, and one Paul McCartney stadium show, as well as a number of samples taken through the mail. I later used it in 2007 on the Josh Groban Awake tour. In addition, I planned to do open-ended interviews, later using grounded theory coding on those interviews (Strauss 1987). I also had fan-written accounts submitted to me through the mail. This narrative data formed the data set for my qualitative doctoral dissertation analysis (Stever 1994). I also did quite a bit of interviewing over the telephone, investing a substantial sum in long distance charges, as this was another good way to reach out to fans across the country. Local fans were interviewed mostly in person.

[4.4] The fourth aspect of the work I did, possibly the most important and potentially unique aspect, involved traveling…a lot of traveling! I felt that it was important to see fans in a naturalistic setting among other fans and at events where they would be seeing, in person, the object of their fan interest. See table 1 for a list of the fan events attended from 1988 to 2018.

[4.5] After making these trips I would pull together the data I had collected from that trip, whether surveys, interviews, or observational notes. This process ended up yielding a book (Stever 2019) and numerous published papers, plus my master's thesis and doctoral dissertation in the years from 1990 to the present.

[4.6] The other thing I did over these years was to join fan groups and fully participate in whatever it was they were doing. My research was not covert in that the people I engaged as participants knew what I was doing. In about 90 percent of the cases, the reception was welcoming, and fans were eager to cooperate because they were motivated to try to understand their own interest and wanted to talk to me about this as a form of self-exploration. My training as a counselor doubtless helped me, as I knew how to ask probing questions that would get below the surface of what I was investigating, and I was fully versed in the ethics of this kind of situation, having taken a graduate-level course in ethics and also being very careful to guard the anonymity of my participants. Again, at both Arizona State and years later at my current SUNY College (Empire State), I passed human subjects review for my research.

5. Instruments: Celebrity Appeal Questionnaire, Celebrity Attitude Scale, and MBTI

[5.1] As a graduate student heading out to large venue concerts in order to collect data on fans, I looked for instruments I could use to explore the questions that were of interest to me. I was most interested in discovering why fans were so devoted to their celebrity of interest. After an exhaustive search of the academic literature yielded no help, I set about constructing my own questionnaire with the assistance of my graduate dissertation committee member, statistics professor David Krus. Under his guidance, I developed an instrument composed of Likert scales that factor analyzed into four discrete categories, each describing an aspect of attraction to celebrities. These included sex appeal, hero/role model, mystique, and entertainer groupings of items. These groupings each successfully predicted a superordinate Likert scale that asked "How big a fan are you of X?" I used the Celebrity Appeal Questionnaire in my early work on pop stars and again when I engaged the Josh Groban fandom (Stever 1991a, 2008).

[5.2] Lynn McCutcheon, John Maltby, and others have developed the already mentioned theory known as celebrity worship and have been doing studies in this area beginning in the early 2000s. For this purpose, they developed the Celebrity Attitude Scale (CAS) (McCutcheon, Lange, and Houran 2002). This particular theory focuses on the less healthy aspects of fan behavior, defining borderline pathological celebrity worship with items like "If my favorite celebrity were to ask me to do something illegal, I would do it." The Intense Personal scale on their instrument contained items like "My favorite celebrity is my soul mate" and "If my favorite celebrity were to die, I wouldn't want to live." Both of these categories have correlated in the research literature with various kinds of problem behaviors, for example eating disorders (Maltby et al. 2005) or addiction (Maltby, Houran, and McCutcheon 2003). A third category, entertainment-social, was more reflective of mainstream fan behaviors.

[5.3] In early work in this area, authors neglected to make a clear distinction between celebrity worship and ordinary fandom. To fill in this gap, I gave the CAS to 212 behaviorally identified fans of either Star Trek or Josh Groban, showing that the percentage of celebrity worshippers among identified fans was no higher than among general population samples or samples of convenience that had been used in the celebrity worship studies (Stever 2011b). My study contrasted the two fan populations I surveyed with various other samples, showing that in fan and nonfan populations alike, borderline pathological celebrity worshippers comprised 3 to 5 percent of all groups (fan or nonfan). Intense personal celebrity worshippers comprised 15 to 20 percent of surveyed groups. Clearly, fans and nonfans alike share the problem behaviors and conditions of celebrity worship.

[5.4] The hypothesis of my master's thesis (Stever 1990) was that fans were attracted to celebrities that they perceived to be congruent with their own personality types. I asked participants to fill out a Myers-Briggs Type Indicator (MBTI) for themselves and then fill out a second one the way the celebrity might answer. MBTI measures personality dichotomies including extrovert/introvert, sensing/intuition, and thinking/feeling. Using this procedure in a number of fan groups, I found that several findings emerged, available in two articles published in the Journal of Psychological Type (Stever 1991b, 1995). Overall, the studies suggested a congruence between the fan's personality type and the perceived type of the favorite celebrity (note 1).

6. Qualitative methods: Grounded theory and ethnographic content analysis

[6.1] Grounded theory is the qualitative methodology of choice in an area where there is not much theory or literature in existing academic writings (Charmaz 2008). For this reason, it was a good way to approach fan studies in psychology, an area not much explored, as has already been discussed. Charmaz identifies key principles of this approach that include "minimizing preconceived ideas about the research problem and the data," and "using simultaneous data collection and analysis to inform each other" (155) (note 2).

[6.2] Ethnographic content analysis (ECA) was a technique developed to incorporate constant comparison in order to discover emerging patterns and themes in qualitative data. The technique makes use of a database program such as Excel in order to organize and sort data. For more detail on using this technique, see Altheide (1987). His discussion specifically contrasts ECA with the then more commonly used qualitative content analysis (QCA), with the observation that ECA is more appropriate for discovery and validity while QCA is more often used for verification and reliability. ECA was the form of analysis I used in my doctoral dissertation and subsequent work that built on the dissertation.

7. Conclusions

[7.1] In thirty years of work, I have concluded that participant-observer ethnography is a highly effective way to explore fan groups of which one is not already a part. Multiperspectivist analysis (Johnson and Onwuegbuzie 2004) that incorporates both quantitative (mostly survey) and qualitative data analysis is most likely to lead to conclusions that stand up to scrutiny and repeat studies.

[7.2] To summarize, my quantitative research involved using research instruments including the Celebrity Appeal Questionnaire, The Celebrity Attitude Scale, and the Myers-Briggs Type Indicator. Qualitative data collected included observational field notes, written documents (mostly letters from fans as much of the research was preinternet), transcribed interviews, and analysis of social media web sites (mostly Twitter). From my observations of various works in fan studies, what made my work unique was the blending of these various methodologies and the movement from one fan group to another.

[7.3] One cannot diminish the longitudinal perspective gained from having done this work over the span of half a lifetime. There are fans I met and interviewed in 1988 at my first Michael Jackson concerts that I still know today. Indeed, in 2009 when Jackson passed away, at least a dozen of the original participants from my dissertation contacted me on Facebook to share their grief. There are Star Trek fans I met in 1993 when I began that work that I still know today, twenty-five years later. There are many Josh Groban fans I met back in 2006, when I began that work in earnest (having done preliminary background work in 2005), that I still know today. Knowing people over many years and seeing the impact their fan interest has had on them over a long period of time has added significantly to my thinking about how fan participation affects development across the human lifespan.

[7.4] Mixing quantitative with qualitative methods allowed me to look at not only larger data sets with the surveys but also individual cases with focused interviews. For fan scholars considering these approaches, it is helpful to consider each fan group with both breadth and depth. By doing both kinds of data collection and analysis, I could not only take an interpretivist approach, considering the meaningfulness of fan activity to the participant, but also look at larger issues like fan demographics. The Celebrity Appeal Questionnaire looked in a quantitative way at motivations to be a fan, and then the ethnographic content analysis of interview and other narrative documents provided a different window into those same kinds of motivations. An analysis of the areas where findings converged supported the validity of those findings.

[7.5] Ultimately all of this allowed me to write theoretical articles based on my experiences and observations, for example the developmental model of the way parasocial relationships progress in fandoms (Tukachinsky and Stever 2018) or a model of parasocial attachment that emerged from findings (Stever 2013). A third example was Stever (2011a) where I was able to use qualitative methodologies based on the work of George and Bennett (2005). They proposed that when comparative case studies are done, a single researcher best conducts them, and three things need to be defined: the phenomenon of interest from which the two cases are taken, the overall objective of the description, and the standardized general questions that should be asked of each case. In general, it is helpful to find models of analysis for qualitative studies in order to have firm protocols to follow. Several have been described here but others are available, for example, Creswell (2007).

[7.6] A conversation I had on the telephone in October 1989 with Park Dietz, a noted forensic psychologist and recognized expert in fan behavior, shaped the nature of my ongoing research in a major way. At that time, he was well known for having been an expert witness in the John Hinkley trial, a trial that resulted in a redefining of erotomanic schizophrenia (the delusional belief that someone of high status is in love with you). During that call, I asked him how many fans he had met in person and he replied, "I haven't met any fans." His studies in fan psychology principally were an analysis of fan mail (Dietz et al. 1991). It was at that point, motivated by that conversation, that I decided that if one was going to understand fans, one had to meet and talk to fans. That was the beginning of what was to become a thirty-year journey into participant-observer fan studies (Atkinson and Hammersley 1994).