Subjectivity in Research: Why Not ? But…
Martin Drapeau

The Qualitative Report, Volume 7, Number 3 September, 2002


This article addresses the question of subjectivity in research. In order to facilitate the use of subjectivity in a research context, the author reminds readers of possible procedures as suggested in the literature. Particular attention is given to the idea of peer debriefing. Inspired by psychoanalysis, the author expands on the concept of discussant or debriefer and suggests that by doing so, subjectivity can be better understood. It is suggested that this may actually be fully integrated into a study in order to both better understand the subject under examination as well as the influence of the research mentor and student dyad. The author shares an example of this approach taken from a previously completed study on pedophile sex abusers.

Key Words: Debriefer, Discussant, Tiers, Subjectivity, Self-Deception, Self-Reflection, Defense Mechanisms, Psychoanalysis, Qualitative Research

1. Introduction

Subjectivity in research is a topic that has led more than once to much discussion and to many debates. For quantitative researchers, it is -and rightfully so- a variable needing to be controlled (see Gilbert, 1998, for review). For qualitative researchers, the answer does not appear to be as simple. Some suggest maintaining subjectivity at a near zero level by sticking to the text which is analyzed (e.g. Mucchieli, 1979), thus establishing a clear connection between qualitative and quantitative research through a search for objectivity and a near-positivist perspective (see Rennie, 2000a, for discussion). On the other hand, many researchers suggest making use of subjectivity and drawing on one's inner experience in order to better understand the subject of a study (Rennie, 1994; Schneider, 1999). For them, distancing themselves from the subject through the use of standardized or semi-standardized methods only keeps the subject… at a distance (Patton, 1990).

Unfortunately, such attempts also present certain risks such as projection on behalf of the researcher (Kahn, 1996), limitations due to the researcher's own blind spots (Drapeau & Letendre, 2001), and a sometimes unclear demarcation between what belongs to subjectivity and what belongs to delusions (Brillon, 1992). The question then is how to make use of subjectivity while avoiding these difficulties. In this article, I attempt to remind readers of a few elements which could help the researcher make better use of his subjectivity and to expand on one specific concept often referred to as the peer debriefer. I will then illustrate the possible use of this by giving an example taken from a previously completed study on pedophilic sexual abusers. As it will become clear for the reader, this effort aims at a certain reconciliation of modern and postmodern epistemology (see Rennie, 2000b, for discussion of the raging debate between the two positions). I will explicitly suggest that subjectivity should be explored as much as possible thus making a strong stand for modernism. As such, subjective impressions may somewhat be considered as a representation of the object of study, hence suggesting that qualitative research is mostly constructionistic or hermeneutic (Smith & Heshusius, 1986). I also believe, however, that subjectivity can stand -per se- in the way of "truth", or at least throw findings off track. As such, it must also be penetrated as much as possible. The study of subjectivity can then give the researcher better leverage in order to understand the object of his study. This last point, combined with efforts to triangulate the findings, can only be considered as a postmodern perspective where rhetoric is replaced by demonstration.

2. The Problems and a Few Possible Solutions

Before a research project is completely elaborated, it is already influenced by many external factors and constraints (such as the funding institution's policies). But it is also oriented according to important internal factors such as the researcher's desires, interests, and preoccupations. Even at an earlier time, the researcher's choice of profession and, later on, of research topics are in great part the result of this subjectivity. Such an idea often appears to be so obvious that it is easy for us to forget. For example, during a regular meeting with the other research teams of our faculty, a colleague presented the results of her study on couples and on the factors leading them to divorce. As the meeting was brought to a close, a few participants remained in the room and discussed more personal topics. When another researcher overheard this colleague who had that day presented her study complain about her chaotic relationships with men, he couldn't help but reply: "Well, with a bit more research, maybe you'll find your answer!".

This example illustrates the importance of understanding and "owning" our subjectivity. It not only demonstrates that it can influence our work but it also suggests that what we find may be nothing more than what we were specifically looking for, sometimes without even knowing it. As such, avoiding the question of subjectivity altogether will only invite it to have a more subtle but yet very important impact on what we do or find (Brillon, 1992). Obviously, this makes doing such research quite more complicated, as this difficulty will also be present when the time comes to analyze anxiety-provoking data. Devereux (1980) clearly demonstrates this when he suggests that the scientist tries by all means to defend himself against anxiety by overlooking certain data or meanings, overanalyzing others, forgetting certain major or minor concepts, or giving unclear or vague descriptions of his findings.

These difficulties, often referred to as self-deceptions (Salner, 1999), are the result of the researcher's use of defense mechanisms. Devereux's comments have already introduced us to a few of them:

  1. Denial, through which an individual "deals with emotional conflicts, or internal or external stressors, by refusing to acknowledge some aspect of external reality or of his experience" (Perry, 1990, p.17);
  2. Repression by which an "individual deals with emotional conflicts, or internal or external stressors, by being unable to remember or be cognitively aware of disturbing wishes, feelings, thoughts or experiences" (p. 31);
  3. Displacement used in order to deal with conflicts by generalizing or redirecting a feeling or thought onto another less threatening object and;
  4. Intellectualisation through which an individual deals with conflicts, thoughts or feelings by the excessive use of abstract or generalized thinking.

Another defense mechanism used by individuals -researchers included- is projection. Projection can be defined as dealing with conflicts by falsely attributing feelings, impulses, or thoughts to others. An individual often makes use of projection when "confronted to an object by whom he feels threatened or to whom he feels some affinity" (my italics; Perry, 1990, p. 19). A good example and examination of the effects of projection can be found in Neck, Godwin, and Spencer's study of decision making processes (1996) and in the later replies by Kahn (1996) and Godwin and Neck (1996). Finally, although many other -if not all- defense mechanisms can possibly be influential in a research project, two more seem to be often overlooked: reaction formation and omnipotence.

Perry (1990) defines reaction formation as dealing with conflicts by "substituting behaviors, thoughts, or feelings that are diametrically opposed to the unacceptable thoughts or feelings" (p. 35), whereas omnipotence refers to the individual's response to emotional conflict by acting superior to others, as if he possessed special abilities. For example, anxiety and a feeling of not having complete control over the object of a study or over specific elements involved in the study may be replaced by a feeling of pleasure and of complete mastery. According to Goleman (1985), self-deceptions tend precisely to exaggerate the researcher's impression of control as well as his self-esteem. These risks would be more important when (a) the researcher is already "devoted" to a certain understanding or position, (b) when certain meanings rising from the data may make this position uncertain and (c) when the researcher feels anxious when confronted to these difficulties. A review led by Rosenthal (1978) of 345 studies seems to confirm these factors.

Because of these difficulties and if one wants to make positive use of subjectivity, many precautions must be taken. For these reasons and as suggested by Ben Slama (1986), Brillon (1992), Caspar (1995), Goldberg (1994), and Lepage and Letendre (1998), it seems important that the researcher wanting to "put to use" his subjectivity undertake introspective work through a personal psychoanalysis or psychotherapy. Although this may be a first step, it cannot it any way be the last. As such, many other precautions are often suggested in the literature. For example, a researcher may keep note of all thoughts and difficulties he may have had in working on his research as well as of all modifications he might have brought to his research design (Salner, 1999). Unfortunately, although often recommended in research method manuals, few researchers clearly explain how to put these personal remarks to use or how to integrate them in the data analysis and presentation. Methodologists also recommend many other steps in order to insure a valid use of subjectivity. Amongst these, we can find:

  1. Submitting the research results to peers and to other experts in the field or comparing the results with what other studies have given;
  2. Doing the data analysis in groups in order to obtain consensus;
  3. Triangulation and other validity and reliability precautions (for a review, see Van der Maren, 1997);
  4. Presenting the results of a more objective, that is text-based, analysis before proceeding with subjective analysis (Mucchieli, 1979) and;
  5. Making use of a "discussant" during the research process (Lincoln & Guba, 1985).

This last procedure can take place through different means. For some authors, this discussant plays the role of a research mentor. As such, he must insure self-reflection and lead the researcher towards new possibilities, whether it is in terms of hypotheses or methods. Other authors prefer avoiding possible authority issues (Guba & Lincoln, 1989) and suggest that the discussant be what they call a peer debriefer. This person is chosen according to his knowledge of the research topic and/or of methodology. It is often suggested that this debriefer be in all ways equal to the researcher (not senior nor junior). His role is once again to facilitate a more personal discussion and to explore the possible ways a researcher may bias his results. According to Spall (1998), researchers often choose as debriefer a friend, a colleague or their spouse. In all cases, the debriefer is selected according to three main axis:

  1. trust,
  2. confidentiality, and
  3. expertise.

Although our research team has always tried to implement most of these suggestions, our interest for psychoanalytic theory (e.g., Bergeret, 1974; Kernberg, 1994; LeGuen, 1974) and for psychoanalytic qualitative research (see Drapeau et al., 2001; Hamann & Letendre, 1996; Lepage et al., 1998) has led us to expand on this idea of the peer debriefer. This was motivated by two main ideas:

  1. As we are convinced that a proper use of subjectivity may help in understanding the subject under study, it seemed important to insure a setting and procedures facilitating its use and study in a research context and;
  2. We also were convinced that subjectivity played a major role in the mentor or research director-student dyad and that this had a major impact not only on the orientation of the project but also on how one might understand the subjects.

With this in mind, it was suggested that the researchers find a debriefer in order to examine some of the issues possibly involved. It was recommended that this debriefer, which was in this context referred to as "le tiers" (the third party1) , have psychoanalytic training and have had prior experience of a psychoanalytic cure. He was also to be found outside the research team and have no relation to any member of the team. The ideal "third party" would then be one's psychoanalyst if one were at that time in analysis.

The idea of making use of a "third party" is to insure other points of reference outside the research team or the supervisor-student dyad. This should facilitate the researcher's creativity and exploration through the use of free association and help determine how he or she subjectively and mentally sees the object of study, while at the same time helping to have an appropriate distance from both the subject of study and the implications of the relation to a supervisor. As such, this "third party" may help the analyzand distinguish what may in fact be delusions or fantasies from the "reality" of the situation. Obviously, this shares much resemblance with an analysis session.

3. An Example of the Possible Use of a "Third Party" in Research

In the lines that follow, I will first present a synopsis of a previously completed research project on the motivation for treatment of pedophile sex abusers. Some of the problems associated to the study of sex abusers will then be presented. I will finally attempt to demonstrate with an example taken from the study how one may put to use his subjectivity in a research setting while making use of a "third party". In order to make this clearer, I have divided the example into two parts:

  1. The question of subjectivity in the study of the subjects per se and;
  2. The question of subjectivity in the research mentor-student dyad and its influence on the study.

3.1. Methodology of the Research Project

In 1998, our research team undertook a qualitative study on 24 pedophiles in treatment at a specialized clinic for sex offenders. This study was to be my doctoral thesis. The initial goals of this study were to understand (1) What motivates pedophile sex offenders to enter and remain in a specialized treatment program, and (2) How they experience the treatment on a daily basis. Non-directive semi-structured interviews (Kandel, 1972; Legras, 1971) were used to gather the subjects' thoughts about their experience of the treatment as well as what they considered to have motivated them to enter and remain in the treatment program. The interviews were analyzed using diverse qualitative methods, including text-based comparative analysis (Maykut & Morehouse, 1994), Multiple-case depth analysis (Schneider, 1999), and psychoanalytically-based qualitative analysis (Brillon, 1992; Drapeau et al., 2001; Lepage et al., 1998).

In order to examine specific questions of interest, Plan Analysis (Caspar, 1995, 1997) was also used on a selected number of subjects. Plan Analysis was done by independent judges based on qualitative theme analysis (Deslaurier, 1987; Mucchieli, 1986). It involves gathering clinically relevant information about an individual's behavior and experience through careful observation and synthesizing it into a meaningful whole. As Caspar writes (1997, p.26), "the fundamental question that guides Plan Analysis is as follows: which purpose, conscious or unconscious, underlies an individual's behaviors and experiences?" The different components of a Plan Analysis are then interrelated and hierarchically organized. They include a patient's goals or desires but also the means by which these goals are attained. Some obvious advantages of Plan Analysis are that it takes into consideration many different theories, it is idiosyncratic and robust and it is pragmatic and relies on concrete methodology.

Finally, after using the qualitative methods mentioned above, some specific pre-established hypotheses derived from the literature on sex abusers were tested using quantitative methods on selected cases. The first method used was Perry's Wish and Fear List (1996). The Wish & Fear List is based on the notion that behavior is motivated. These motives can be both biologically and psychologically determined in varying degrees and consist, in their simplest form, of wishes and fears. According to Perry (1996, p.15), "a dynamic Wish differs from a simple desire by playing a causal role in a variety of behaviors, fantasies and experiences". A Fear can be defined as "a negative belief, expectation, or aversive experience which the subject wishes to avoid" (p.16). The method developed by Perry encompasses notions of Erikson's developmental stages (Erikson, 1985) and of self-psychology and reintroduces an object-relations perspective via an interest for motives. Such a perspective, although avowedly psychodynamic, has the advantage of being free of strict psychoanalytic jargon and of not violating any basic propositions from learning theory.

Finally, the last instrument used was the Core Conflictual Relationship Theme (CCRT) (Luborsky & Crits-Christoph, 1990). Amongst others things, the CCRT identifies three components: a subject's wishes (W), the responses from others (RO), and the responses of self (RS). There is much evidence that these components describe core repetitive relationship patterns and share many characteristics with Freud's transference theory (Luborsky, 1990). The CCRT was used by one independent judge in order to examine our subjects' relationship patterns.

3.2. An Introduction to the Study of Pedophilic Behavior

A research topic like sex abuse -even more so when it concerns children and pedophilic behavior- presents many major difficulties. The first one a researcher is confronted to is the question of definition of sex abuse and, more specifically, of sexual abuse of children. Despite the spectacular increase of research in this field, no consensus has yet been reached (Friedman, 1990). Between the writings of Krafft-Ebing (1912) and the more recent definition of pedophilia in the DSM-IV (APA, 1994) can be found a multitude of definitions and concepts often revealing strong legal and moral implications. Some of these definitions, still in use today, include an individual masturbating while thinking or fantasizing about children, a child being victim of an exhibitionist or a more clearly defined sexual contact between a child and an adult (for a review, see Li, West & Woodhouse, 1993; Okami, 1990). In that sense, some definitions are still the result of strong prejudice or personal conflicts. It is then not surprising to see that sexology has often been infiltrated by demonology and that many examples mentioned in the literature clearly are limited to the most provocative cases. This has often led to certain exaggerations such as considering all sexual activity of or between children as abusive (Finkelhor, 1979). The problem of studying sexual abuse of children and pedophilic behavior is even greater when one considers the reactions a research project may bring up. Since many studies are often used in court in order to obtain either a guilty or a non-guilty verdict, some researchers have been thought of in the public opinion as being pro-pedophile. The least that can be said is that the study of sexual abuse is an emotionally charged one. As my study was on the motivation for treatment of pedophilic sexual abusers, I could not help but confront these questions as well as what brought me to give attention to this topic. But my main concern (and what I wish to emphasize here) was what could be understood from interviewing pedophilic subjects. With all of this in mind and as I was in a psychoanalytic cure at the time, I proceeded to bring up these questions with my analyst, who later on acted as "third party".

3.3. Subjectivity in Understanding the Subjects: A Case Illustration

My first subject was a man in his sixties2. Prior his arrest, he had held an important and very well considered professional position. He was serving a long-term federal sentence after being condemned for sexually abusing young boys over a period of many years. He showed up at the interview very well dressed and looking confident. He answered all questions without hesitation, giving much detail in a way a professor would lecture his students. After 70 minutes of interviewing, the meeting was interrupted by a guard asking that the subject return in his cell for inspection. I then proceeded to note down my subjective impressions in a notebook I reserved for this use. These close to two pages of notes included the following:

"(Subject) seems very educated and definitely very articulate. His strong accent gives him a sort of aristocratic twist. It felt as if he was actually lecturing me or teaching me something that I should know. (I) felt like a kid who's begging for information and desperately waiting for his next word. Seemed like I was learning something !!! (…) The contact was good, maybe even a little too good? Something fascinating about him?, almost father or grand-father like (…) What is it? (…) At the same time, it felt intrusive, as if I should better watch out and not get skrewed. Don't know what to make of it (…) Very confusing, ambivalent."

My complete notes were later on brought in an analytic session and a total of close to four sessions were actually used to try to understand what I had felt during this interview. The most striking aspect was how I had felt like I was twenty or twenty-five years younger, almost like a little boy. I had felt charmed and fascinated by this person as if I could learn much from him. Through analysis, I managed to establish clear relations between this and my own history, more precisely with my father. In that sense, it seemed as if I was longing for something that was missing and that this subject could actually give me, that is: a feeling of being an "adult, mature man". As such, during the interview, I had had fantasies of establishing contact with him once his imprisonment would be over, hoping that what I had missed during the interview could be obtained later through some kind of contact. While I had these feelings, I also felt as if I should better make sure I don't get "skrewed". This last term brought up much anxiety, as it could easily be related to sexual abuse. And somehow, it did feel like rape. It then became clear that I felt ambivalent towards the subject, both fearing and wanting him to give me the feeling of being an "adult man". Something the subject lived somehow resonated in me, while at the same time not being completely mine. This led to the idea that, although the subject and I did share certain common fantasies, differences could be found. It so appeared that through different processes, namely projective identification and identification to the aggressor, the subject had led me on a playground where his rules prevailed and where he could actually bring me to feel like he might have felt while maintaining himself in a superior -fatherly- position. Spontaneously, this leads to many hypotheses regarding the subject's functioning:

  1. He might have missed having a strong -if any- father figure capable of leading him to adult life (see Balier, 1996);
  2. He may have struggled with an authoritarian maternal figure (see Balier, 1993); and
  3. He may feel like a young boy and not entirely like an adult man and these feelings are intolerable for him.

As I later realized, many psychoanalytic theories have been developed around these ideas and identification to the aggressor is often a core component of them (Bril, 1993; Centerwall, 1992; Groth, Hobson & Gary, 1982). According to Laplanche and Pontalis (1967), identification to the aggressor involves the subject identifying to his aggressor by making himself the perpetuator or making his certain characteristics of the aggressor. As such, the pedophilic act involves mastery by leaving the victim position to become the aggressor. More recent theories (Balier, 1993, 1996) also suggest that the pedophile is in many ways "his mother's narcissistic doll". Indirectly, this suggestion raises the question as to what place -if any- the father had in the child's life. It also suggests that the pedophile struggles with symbiotic issues with the mother and lacks positive identification with the father (Szwec, 1993). Through analysis with a "third party", these hypotheses seemed applicable. But the question still remained as to how true or significant they were.

3.3.1. Verifying these hypotheses As quantitative measures had been used in order to examine other aspects of the research question, quantitative results were readily available for the subject presented above. Hence, it was interesting to triangulate hypotheses resulting from the examination of my subjectivity with other methods. Using Caspar's Plan Analysis (1997), it appeared that some of the most central themes brought up by this subject were "a need to have a sense of mastery", "a fear of being dominated", and "a wish to put the therapists to test by confronting them in order to make sure they are strong, authoritarian and reliable "father" figures" (Drapeau, Körner, Granger, Brunet & Caspar, 2002). The Wish and Fear List (Perry, 1996) was used by independent judges and suggested that the subject's most prevalent motives were: a wish to be autonomous, a wish to control and dominate others, a wish to have and be a role model or a mentor, a fear of being dominated, and a fear of physical harm. Overall, most motives mentioned by the subject were related to what Erikson (1985) called autonomy versus shame and doubt issues. Finally, the CCRT (Luborsky & Crits-Christoph, 1990) indicated that the subject was most often motivated to have control over others, to be independent, and to be his own person. Others most often reacted to these wishes by being controlling (in the case of women) or being distant - feeling controlled (in the case of men). The subject then was left feeling helpless, incompetent and disappointed.

What is interesting about these methods is that they all seem to underline and empirically support different aspects of what had been subjectively felt during the interview. Hence, examining my subjectivity provided -in itself- a good indication of the subject's functioning. But as I will describe next, examining one's subjectivity can also be important when the time comes to analyze the data at hand as other factors also have an influence on the research process.

3.4. Subjectivity in the Research Mentor-Student Dyad

As I had had intensive training in quantitative research and still worked in that field, I chose to do a Ph.D. where I could learn new and different research methods. This led me to a psychoanalysis professor who had been doing qualitative work for many years. Very early in the research project, a deep abyss could already be seen between the way this professor saw therapeutic work and the way patients were treated at the penitentiary. All pedophilic patients there had the opportunity to start a treatment program in a specialized clinic associated to the penitentiary. This treatment, as most programs for sexual abusers, was cognitive-behavioral oriented and included the use of aversive conditioning using ammonia or, occasionally, mild self-inflicted electroshocks. Although such treatments have been proven most effective with sexual abusers (Marshall, Eccles, & Barbaree, 1993; Marshall, Jones, Ward, Johnston, & Barbaree, 1991) and while psychoanalytically orientated therapies had largely failed in this matter (Li et al., 1993; McGrath, 1991), my professor was somewhat shocked and offended when I first described the treatment to him.

In the months that followed, I ended up being a moderator for both parties, that is the cognitive-behavioral and the old fashioned psychoanalytic models. Clearly enough, as my professor was the one who not only rated my work but had the final word on the thesis project, I made clear efforts when analyzing the data in order to demonstrate his impressions were right: "people just cannot be satisfied with such a rigid treatment that gives them so little room!". For example, I would over emphasize the fact that the subjects had little chance to talk (which is a central aspect of psychoanalytic work) despite their desire -and capacity- to do so. In that sense, when I first analyzed the interviews, I took special care to show that the subjects had a strong wish to talk, while looking for clues that they were not satisfied with the present treatment. For these reasons, I would over emphazise a subject's strengths and capacities in order to demonstrate that he could be amenable to psychoanalytic therapy but was deprived of it.

As work with my research mentor eventually became difficult and even counter-productive, I became more and more frustrated and then switched positions. In analyzing data, I would give more attention to clues indicating that the subject was not amenable to psychoanalytic therapy. At the same time, I would over emphasize the "cruelty" (as he put it) of the cognitive-behavioral treatment. It became clear that my efforts, although not entirely conscious at the time, aimed at contradicting and offending him. This confrontation was motivated by technical as well as emotional motives. For instance, although it quickly became clear, through analysis, that I deliberately confronted my research mentor in order to contradict him as a result of being frustrated for having been placed in a moderator position, there also appeared to be other reasons for doing so. One of these reasons seems of particular importance and was the result of certain "similarities" between the relation I had with my supervisor and the one my first subject had with his father (as seen in theme analysis): neither one would "put out", that is give what they were supposed to give (e.g., support and input). Because of this, I identified easily with the subject and used his (most shocking) words to confront my supervisor, as the subject had done himself with his father. The only way to overcome these difficulties was to analyze them with the help of the "third party".

4. Discussion

The question then is for the researcher to determine just how much -if at all- he or she wishes to validate his findings with other methods. Although a study relying entirely and exclusively on one's subjectivity may not respect the criteria established by science, it nonetheless can be a central part of the research project. Its advantage is clearly that, unlike in quantitative methods, it can at least make us cautious when analyzing data. For example, through the use of a "third party", it quickly became clear that the subject I interviewed had had a major impact on me. Because of this, it later was clear that I had a tendency, when analyzing the data, to over emphasize positive aspects of the subject's personality or to search for possible explanations for his behaviors (excuses) while overlooking comments which may have negatively influenced how I saw him. This general tendency could be explained by two possible reasons. The first one was my motivation or desire to preserve the image or representation I had of this subject. The second one was the result of the relationship that was established with my thesis supervisor and the (temporary) conflicts in which it resulted. As I look back at that time, I can only be convinced that without making use of a "third party", many of the conclusions I had suggested would have been clearly biased and not representative of the subjects' dynamic functioning. In other words, I believe that the use of a "third party" helped in being more objective about my own subjectivity and the impact it had on the data analysis.

Once again, such a stand reveals an effort to reconcile modern and postmodern epistemology. By seeking to explore my subjectivity and to accept that it may be richly informative about the subject's functioning, it suggests that the essence of meaningful understanding is in a subjective representation of the subject or, as psychoanalysts would put it (Widlöcher, 1995; Drapeau, 2002), in a shared and approximate metaphor of his functioning. At the same time, the use of a "third party" in order to better distinguish subjectivity from delusion implies that subjectivity can stand in the way of a better understanding of the subject. Hence, it should then be penetrated as much as possible in order to better differentiate what belongs to the researcher from what belongs to the subject. As such, it reveals a definite postmodern epistemology where demonstration, as can also be seen in any effort to triangulate findings, is necessary. As my efforts were aimed at making use of both epistemological stand points, they can certainly be seen as an attempt, although certainly incomplete, to reconcile the best of both perspectives.

5. Conclusion

As I have attempted to demonstrate, subjectivity analysis can be included in a research project as it can serve two fundamental purposes:

  1. It can help understand the object of investigation and
  2. It can help understand how significant personal relationships such as the mentor-student dyad may influence data analysis and understanding.

Not all researchers need to fully make subjectivity analysis a central part of their investigation. For quantitative researchers, it seems best to control this variable as much as possible. This may be the same for some qualitative researchers. Everything obviously depends on what the aims of a study are. But for those who wish to make use of subjectivity in a research setting, many precautions such as the ones suggested in the literature must be taken: triangulation, consensus, text-based analysis, introspective work of the researcher, and so on.

As for the question of using a debriefer, it is definitely a valuable one. Although this debriefer may play many roles and serve many purposes, we believe that another form of debriefing, one with what we have called the "third party", can play a crucial part in a research project. As the tendency or trend is now to have bigger research teams with many researchers, this form of debriefing can certainly help the researcher have a moment of reflection outside the institution on what is involved in terms of constraints and dynamics. For those who choose such an orientation, the question then is to determine just how much they are willing to say about themselves when the time comes to publish results. Furthermore, one must keep in mind that such procedures are very time consuming and require a strong desire and will to explore his -often anxiety provoking- personal conflicts and biases (for a good review, see Hamann et al., 1996).

As it becomes clear that the researcher himself -and not only the subjects- should interest science, one must come to a balance between what he makes public and what he keeps to himself. After the challenges of analyzing subjectivity lies that call for humility and there, after all, is the true challenge which a researcher must face.


      American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: author.

      Balier, C. (1993). Pédophilie et violence. L'éclairage apporté par une approche criminologique. Revue Française de Psychanalyse, 2, 573-589.

      Balier, C. (1996). Psychanalyse des comportements sexuels violents. Paris: PUF.

      Ben Slama, F. (1986). Le contre-transfert dans la recherche: De la notion au paradigme. Bulletin de Psychologie, 39(16-18), 791-796.

      Bergeret, J. (1974). La personnalité normale et pathologique: Les structures mentales, le caractère, les sympt“mes. Paris: Dunod.

      Bril, J. (1993). Préhistoire et pédophilie. Revue Française de Psychanalyse, 2, 527-534.

      Brillon, M. (1992). Recherche clinique d'inspiration psychanalytique: Essai méthodologique. Revue de l'Association pour la Recherche Qualitative, 7, 7-20.

      Caspar, F. (1995). Plan analysis: Toward optimizing psychotherapy. Seattle: Hogrefe & Huber Publishers.

      Caspar, F. (1997). Plan analysis. In T. Eells (Ed.), Handbook of psychotherapy case formulation (pp. 119-151). New York: Guilford Press.

      Centerwall, B. S. (1992). Vladimir Nabokov: A case study in pedophilia. Psychoanalysis and Contemporary Thought, 15(2), 199-238.

      Deslauriers, J. P. (1987). Les méthodes de la recherche qualitative. Sillery: Presses de l'Université du Québec.

      Devereux, G. (1980). De l'angoisse à la méthode dans les sciences du comportement. Paris: Flammarion.

      Drapeau, M. (2002). Lorsque certains analystes résistent encore aux chiffres. Psychoscope: the Official Journal of Swiss Psychologists, 23(3), 28.

      Drapeau, M., Körner, A., Granger, L., Brunet, L., & Caspar, F. (2002). Therapy: Yes or no ? A plan analysis study of pedophile sexual abusers' motivations for treatment. Manuscript submitted for publication

      Drapeau, M., & Letendre, R. (2001). La recherche qualitative d'inspiration psychanalytique: Quelques propositions pour en augmenter la rigueur. Revue de l'Association pour la Recherche Qualitative, 22, 73-92.

      Erikson, E. H. (1985). The life circle completed: A review. New York: W.W. Norton.

      Finkelhor, D. (1979). What's wrong with sex between adults and children? Ethics and the problem of sexual abuse. American Journal of Orthopsychiatry, 49(4), 692-697.

      Friedman, S. R. (1990). What is child abuse? Journal of Clinical Psychology, 46(3), 372-375.

      Gilbert, S. (1998). L'idéal de l'objectivité et objectivité idéale. Revue de l'Association pour la Recherche Qualitative, 18, 43-170.

      Godwin, J. L., & Neck, C. P. (1996). Researcher "projection" revisited: A response to Kahn. Journal of Applied Behavioral Science, 32(3), 323-331.

      Goldberg, A. (1994). Farewell to the objective analyst. International Journal of Psycho-analysis, 75(21), 23-30.

      Goleman, D. (1985). Vital lies, simple truths: The psychology of self-deception. New York: Simon & Schuster.

      Groth, A. N., Hobson, W. F., & Gary, T. S. (1982). The child molester: Clinical observations. Journal of Social Work and Human Sexuality, 1(1), 129-144.

      Guba, E. G., & Lincoln, Y. S. (1989). Fourth generation evaluation. Newbury Park, CA: Sage.

      Hamann, S., & Letendre, R. (1996). La recherche qualitative comme moyen de situer les sujets au cœur des interventions dans le domaine de la santé. Revue de l'Association pour la Recherche Qualitative, 15, 9-21.

      Kandel, L. (1972). Réflexions sur l'usage de l'entretien, notamment non directif, et sur les études d'opinion. Épistémologie Sociologique, 13, 25-46.

      Kahn, W. A. (1996). Comment on "Understanding researcher "projection" in interpreting case study data: The South Canyon fire tragedy. Journal of Applied Behavioral Science, 32(1), 62-69.

      Kernberg, O. (1994). Validation in the clinical process. International Journal of Psycho-Analysis, 75, 1193-1200.

      Krafft-Ebing, R. (1912). Psychopathia sexualis. Paris: Ed. Pocket.

      Laplanche, J., & Pontalis, J. P. (1967). Vocabulaire de la psychanalyse. Paris: Presses Universitaires de France.

      Legras, D. (1971). Quelques contributions à la méthodologie de l'entretien non-directif d'enquête. Bulletin du C.E.R.P., 20(2), 131-141.

      LeGuen, C. (1974). L'Œdipe originaire. Paris: Payot.

      Lepage, L., & Letendre, R. (1998). L'intervention de manifestations contre-transférentielles dans le déroulement de la recherche: Réflexions sur une pratique et exemples. Revue de l'Association pour la Recherche Qualitative, 18, 51-76.

      Li, C. K., West, D. J., & Woodhouse, T. P. (1993). Children's sexual encounters with adults: A scientific study. New York: Prometheus Books.

      Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Beverly Hills, CA: Sage.

      Luborsky, L. (1990). The everyday clinical uses of the CCRT. In L. Luborsky & P. Crist-Christoph (Eds.), Understanding transference - The core conflictual relationship theme method (pp. 211-221). Washington, DC: American Psychological Association.

      Luborsky, L., & Crits-Christoph, P. (1990). Understanding transference - The core conflictual relationship theme method. Washington, DC: American Psychological Association.

      Marshall, W. L., Eccles, A., & Barbaree, H. E. (1993). A three-tiered approach to the rehabilitation of incarcerated sex offenders. Behavioral Sciences and the Law, 11, 441-455.

      Marshall, W. L., Jones, R., Ward, T., Johnston, P., & Barbaree, H. E. (1991). Treatment outcome with sex offenders. Clinical Psychology Review, 4, 465-485.

      Maykut, P., & Morehouse, R. (1994). Beginning qualitative research: A philosophical and practical guide. Washington, DC: The Falmer Press.

      McGrath, R. J. (1991). Sex offender risk assessment and disposition planning: A review of empirical and clinical findings. International Journal of Offender Therapy and Comparative Criminology, 35(4), 328-349.

      Mucchieli, R. (1979). L'analyse de contenu des documents et des communications. Paris: Éditions ESF.

      Mucchieli, R. (1986). L'analyse de contenu des documents et des communications. Paris: Connaissance du problème.

      Neck, C. P., Godwin, J. L., & Spencer, E. S. (1996). Understanding researcher "projection" in interpreting case study data: The South Canyon fire tragedy. Journal of Applied Behavioral Science, 32(1), 48-61.

      Okami, P. (1990). Sociopolitical biases in the contemporary scientific literature on adult human sexual behavior with children and adolescents. In J. R. Feierman (Ed.), Pedophilia: Biosocial dimensions (pp. 91-121). New York: Springer-Verlag.

      Patton, M. Q. (1990). Qualitative evaluation and research methods. Beverly Hills, CA: Sage Publications.

      Perry, J. C. (1990). Defense mechanism rating scale. Unpublished manuscript. Harvard School of Medicine, Boston, USA.

      Perry, J. C. (1996). Wishes & fears: A standardized list for assessing dynamic motives (3rd ed.). Unpublished manual. Harvard School of Medicine, Boston, USA.

      Rennie, D. L (1994). Human science and counselling psychology: Closing the gap between research and practice. Counselling Psychology Quarterly, 7, 235-250.

      Rennie, D. L. (2000a). Anglo-American counselling and psychotherapy qualitative research. Paper presented at the meeting of the Society for Psychotherapy Research, Chicago, USA.

      Rennie, D. (2000b). Grounded theory methodology as methodological hermeneutics. Theory & Psychology, 10(4), 481-502.

      Rosenthal, R. (1978). Interpersonal expectancy effects: The first 345 studies. The Behavioral and Brain Sciences, 3, 377-415.

      Salner, M. (1999). Self-deception in qualitative research: Validity issues. Paper presented at the meeting of the Association for Qualitative Research, Melbourne, Australia.

      Schneider, K. J. (1999). Multiple-case depth research. Journal of Clinical Psychology, 55(12), 1531-1540.

      Smith, J. K., & Heshusius, L. (1986). Closing down the conversation: The end of the qualitative-quantitative debate amongst educational inquirers. Educational Researcher, 15, 4-12.

      Spall, S. (1998). Peer debriefing in qualitative research: Emerging operational models. Qualitative Inquiry, 4(2), 280-292.

      Szwec, G. (1993). Faudra mieux surveiller les petits! Revue Française de Psychanalyse, 2, 591-603.

      Van der Maren, J. M. (1997). La recherche qualitative peut-elle être rigoureuse?. Revue de l'Association pour la Recherche Qualitative, 17, 80-89.

      Widlöcher, D. (1995). Un cas n'est pas un fait. L'inactuel, 3, 87-103.


1. For lack of a better translation of the French term "le tiers" into English, I have chosen "the third party". "Le tiers" is a term used in psychoanalytic literature (e.g. Bergeret, 1974; LeGuen, 1974) which refers to the third component involved in the oedipal triangle, namely the father, and whose role is to help the child distance himself from the mother in order to gain access to neurotic functioning.

2. For confidentiality reasons, all information regarding the subject has been modified.

Author Note

+ Martin Drapeau, M.Ps., Ph.D. Candidate is currently waiting for his Ph.D. thesis defense in psychology at the Université du Québec à Montréal, Québec, Canada. He has recently started postdoctoral studies at the Lausanne University Adult Psychiatry Department in Switzerland. As a licensed psychologist and researcher, he is interested in clinical research. His current areas of research include psychotherapy process, motivation for treatment, change in therapy and psychopathology.

He can be contacted at CEPP-DUPA, Rue du Tunnel 1, CH-1005 Lausanne, Switzerland; telephone: 011 41 21 316 7983; and email: Correspondence regarding this article can be sent to Martin Drapeau, M.Ps., C/O Carolyn Dovick (Direction of Research), Institute of Community and Family Psychiatry, Jewish General Hospital, 4333 Chemin de la Côte Ste-Catherine, Montreal, Quebec, Canada, H3T 1E4

Acknowledgements: This project was supported by a Quebec Government Fonds pour la Formation de Chercheurs et l'Aide à la Recherche (FCAR) Doctoral Grant and by a FCAR Foreign Exchange Research Grant to the author.

The author is indebted to Pr. David Rennie of York University for his comments on epistemology.

Article Citation

      Drapeau, M. (2002, September). Subjectivity in research: why not ? But… The Qualitative Report, 7(3). Retrieved [Insert date here], from

Martin Drapeau
2002 copyright

Return to the top of the paper.

Return to the Table of Contents.