Therapeutic implications of preferred view more

METHODS: A JOURNAL FOR HUMAN SCIENCE Annual Edition 2001 Articles A Note on "Fundamental Structures," etc., Thirty Years Later by Pau F. Colaizzi 7 The Epistemology-Practice Gap in Experimental Psychology: A Review and a Proposed Sulution by Russell A. Walsh 11 The Wisdom of the Empty Places by Klaus Nielsen 29 Therapeutic Implications of Preferred View by Donald J. Domenici and Jonathan D. Raskin 42 The Knowing Self in Personal Projects Analysis by Elizabeth Engelke 60 Literacy and Literary Theory in East Palo Alto by Bernadette Waterman Ward 76 42 methods THERAPEUTIC IMPLICATIONS OF PREFERRED VIEW DONALD J. DOMENICI AND JONATHAN D. RASKIN State University of New York at New Paltz Abstract: The importance of individuals' preferred views of self (see Eron & Lund, 1996) is discussed and explored em- pirically. The present study addresses the hypotheses that indi- viduals experience negative affect and have difficulty dealing with others when they are viewed in ways that are discrepant with the preferred views they have of themselves. Twenty-five participants were randomly assigned to have their self-identi- fied preferred views either confirmed, discontinued, or nei- ther, through results of a personality assessment. A One-way Multivariate Analysis of Variance (MANOVA) revealed that participants whose preferred views were disconfirmed rated various aspects of the personality assessment as more nega- tive than either of the other two groups, but there were no dif- ferences in self-reported psychological discomfort. Implica- tions of preferred view for therapy and diagnosis are discussed. Preferred View and Therapy 43 Therapeutic Implications of Preferred View Throughout the history of the field of psychology, consid- erable emphasis has been placed on the study of various facets of the self. More specifically, several theorists have stressed the importance of ways in which one views the self (Higgins, Bond, Klein, & Strauman, 1986). According to some theorists (e.g., Lund & Eron, 1998), it is these self-images or views of the self that influence how individuals deal with life events. Eron and Lund (1993, 1996) propose a model of how problems arise in everyday life in relation to one's preferred view of self. Accord- ing to Eron and Lund (1993, 1996), preferred views are strong preferences that individuals have about themselves which are integral in resolving problematic issues. This way of conceptu- alizing problems is strongly rooted in the work of humanistic psychotherapists Carl Rogers (1951,1961) andR.D. Laing (e.g., 1969). Although Eron and Lund (1996) have documented sev- eral case examples which demonstrate their model, as of yet there has not been any empirical research conducted to support their theory. Eron and Lund (1993, 1996) place a firm emphasis on the role of preferred views in how problems arise and are resolved. "People have strong preferences with regard to how they would like to behave, how they would like to see themselves, and how they would like to be seen by others" (Eron & Lund, 1996, p. 44). These preferences are referred to as a person's "preferred view." Eron and Lund (1996) describe preferred views as fluid and evolving, consisting of past, present and future dimensions. Similarly, they describe preferred view not as a "thing" that one has, but as a variety of possible views or preferences that people find suitable or "that fit with who they wish to be" (Eron & Lund, 1996, p. 45). The concept of preferred view grew out of the approach of strategic family therapy, which emphasizes the therapeutic im- portance of understanding how people want to be seen by others (Eron & Lund, 1993). Eron and Lund (1993) link the strategic 44 methods approach with the framework of a broad life narrative, focusing on how the stories people use to structure their experience can lead to problems. The idea of individuals having preferred ways of viewing themselves can also be traced back to Rogers' (1951) notion of the self-ideal, which represents the person one would like to be. Rogers (1951) contrasts this ideal view of self with the self as perceived, which is a more realistic view of one's self as he or she actually perceives it. Rogers (1961) studied out- comes of client-centered therapy, finding that clients who com- pleted therapy perceived their selves as being more congruent with their ideal selves. Other researchers have also explored self- views in ways that relate to the concept of preferred view. Ex- amples include Higgins' (1987) work on domains of the self, research on self-verification theory done by Swann and col- leagues (e.g., Swann, De la Ronde & Hixon, 1994) and Markus and Nurius' (1986) discussion of possible selves. A basic assumption of Eron and Lund's (1993,1996) theory is that the way a person acts in a given situation is based on his or her construal of the situation. This theory has its foundations in the work of Kelly (1955) and the idea of constructive alternativism, "the notion that there are many workable alterna- tive ways for one to construe his world" (p. 560). Eron and Lund (1996) add that the construction of events and situations depends largely on a person's view of other people's intentions, as well as his or her view of how others view him or her (see also Laing et al., 1966). It is the relationship between one's perception of how others view him or her and the preferred view of self that he or she holds that is an important factor in the development of potentially problematic interpersonal relationships (Eron & Lund, 1996). This relationship between views of the self and others echoes the work of Laing (1969), who wrote extensively about these vantage points. An important key to the way problems develop in relation- ships involves the idea that individuals experience anxiety, frus- tration and other unsettling emotions when they "behave in ways that are discrepant with preferred views of self," or when "they Preferred View and Therapy 45 see themselves in ways that are discrepant with preferred views" (Eron & Lund, 1996, p. 46). Furthermore, individuals also ex- perience negative emotions when they perceive others viewing them in ways that are discrepant with their preferred views (Eron & Lund, 1993,1996). Eron and Lund (1993) describe how indi- viduals "begin to see others seeing them in ways they object to, find distasteful, or in general find discrepant with their preferred views of self5 (p. 297). Laing (1969) uses the term "disjunction" to refer to this discrepancy; "being seen by the other not as one sees oneself (p. 92). Eron and Lund (1996) discuss this dis- crepancy as a gap which, when widened, causes people to see events in ways that contradict their preferred views, with these new constructions or stories becoming fixed and remaining as problems. Therefore, an important aspect of the clinical work of Eron and'Lund (1993, 1996) involves treating the client in a way that is consistent with his or her preferred view. When this is done, the "problem" can be refrained in a way that reaffirms the client's preferred view, and the gap experienced between one's preferred view and how one sees others viewing him or her can be narrowed. This way of doing therapy is consistent with Rog- ers' (1951) client-centered approach, in which the client's per- ception of how the therapist views him or her is important in carrying out effective therapy. In the process of therapy, Eron and Lund (1996) utilize aspects of Rogers' approach in a way that is synthesized with Laing's more interpersonally based theory. Rationale and Hypotheses of the Current Research The purpose of the present study is to find support for the effects that confirming or disconfirming individuals' preferred views can have, as described by Eron and Lund (1993, 1996). Specific effects that will be explored include how individuals react to having their preferred views either confirmed or disconfirmed through results on a personality assessment. It is hypothesized that participants whose preferred views are 46 methods disconfirmed will report experiencing a higher level of psycho- logical discomfort (based on a psychological discomfort scale) than those whose preferred views are confirmed. It is also hy- pothesized that participants whose preferred views are disconfirmed will report feeling more negatively about various aspects of the personality assessment in general (as measured by a questionnaire about various aspects of the assessment) than those who have their preferred views confirmed. These hypoth- eses are based on the intrapsychic and interpersonal implica- tions of preferred views discussed in the work of Esron and Lund (1993,1996). Method Participants A total of 25 undergraduates (13 female and 12 male) at- tending a mid-sized state university participated in this study. Participants were volunteers taken from undergraduate psychol- ogy courses at the university. The age range of participants was between 20 and 34 years of age and participants reflected a va- riety of ethnic backgrounds. All participants were treated in ac- cordance with the ethical standards proposed by the American Psychological Association. Materials Q-Sort Deck. The assessment of participants' preferred views was accomplished through the use of the Q-sort tech- nique, a method that has been widely used in the assessment of self-concept (Rogers, 1951, 1961; Stephenson, 1953). The Q- sort technique involves using a deck of Q-sort items (self-state- ments about personality traits), which participants usually rate along a series of dimensions (e.g., with categories ranging from "most desirable" to "least desirable"; Denzine, 1998). Using the Q-sort method, participants were instructed to sort various adjectives that described themselves (i.e., personality traits) into five categories, such that they give an account of themselves as Preferred View and Therapy 47 they would ideally be at their best. Categories ranged from "least like how you ideally are at your best" (Category 1) to "most like how you ideally are at your best" (Category 5). To have participants respond in a way that provided a wide range of traits that characterized their preferred and non-preferred views, they were asked to place 10 adjectives each in categories one and five, 15 adjectives each in categories two and four, and 30 ad- jectives in category three (neutral statements). Thus, the Q-sort deck contained a total of 75 cards, each one with a different personality trait typed on it (e.g., sentimental, obedient, mod- est, clever, ambitious). Items were selected from traits that have been previously rated as being at least somewhat desirable, but not overly so, to avoid confounding preferred views with posi- tive and negative traits. The traits were selected from a list of 555 personality traits rated on the likableness and meaningful- ness by college students in a study performed by Anderson (1968). In addition to selecting traits that are neither excessively desirable nor excessively undesirable, only traits that were rated as having a high level of meaningfulness were used, that is, participants in Anderson's (1968) study rated the words as ones that the meanings of were understood fairly well. This criterion is used in the present study to exclude words that are somewhat ambiguous in meaning. Personality assessment and results. An unpublished per- sonality assessment called the Simple Rating Scale for Personal- ity-Revised (Fossum, Weyant, Etter & Barrett, 1998) was used in this study. It was identified to participants as the Core Self Per- sonality Assessment-Version A. The assessment consists of 35 questions about one's personality, which are to be rated on a 5- point Likert scale. However, the assessment was not scored or used to attain valid personality assessment results in this study. The assessment was administered under the cover story that it would be used to assess one's personality, but the results were actually generated based on participants' preferred views, which were assessed using the Q-sort technique. The assessment involved 48 methods participants filling out a computer bubble-scoring sheet to give the impression that it could be quickly scored by a computer. Hie "results" for this assessment were generated using a stan- dard 16 sentence report into which one of three types of feed- back were inserted. Thus, each participant received the same results report except for the adjectives that were inserted. The first type of feedback included information confirming partici- pants' preferred views (using adjectives from categories four and five of the Q-sort). The second type included information that neither confirmed nor disconfirmed the participant's preferred views (using adjectives from category three of the Q-sort). Fi- nally, the third type of feedback included information that was discrepant from the participant's preferred views (using adjec- tives from categories one and two of the Q-sort). Each partici- pant received one of these three types of feedback, which were randomly assigned. Post-assessment questionnaire. The dependent measure that was utilized in this experiment was a brief, two-part Post-As- sessment Questionnaire. The first part was a nine question sur- vey about what the participant thought of three different aspects (the assessment questions, the assessment environment, and the assessment results) of the personality assessment process. All questions involved rating statements from "Strongly Disagree" to "Strongly Agree" on a seven-point Likert scale. The second part of the Post-Assessment Questionnaire was an unpublished Psychological Discomfort Scale (Elliot, personal communica- tion), which was used as a measure of participants' levels of psy- chological discomfort immediately following their receiving "re- sults" of the personality assessment. This scale consists of 24 mood statements (e.g., uncomfortable, content, anxious, dis- tressed) to which the participants rated as "Does not describe me at all" to "Describes me very well" on a five point Likert scale. Procedure The present study was conducted in a psychology labora- tory in the basement of an academic building, where all parts of Preferred View and Therapy 49 the study were administered individually. Each participant was introduced to the primary researcher and an assistant, and a cover story was presented, explaining that the research assistant was a student researcher who had helped the primary researcher de- sign a new personality assessment. Participants were told that this assessment, the Core Self Personality Assessment (CSPA; Version A), had been found to accurately measure one's "core self or underlying personality. Participants were also told that the student researcher would be serving as the experimenter in this study by administering the assessment in addition to other measures designed to help construct items for an alternate form of the assessment. After the primary researcher left the room, participants were asked to perform the Q-sort task, for which they were given written and verbal instructions (as noted above). Following the completion of this task, participants were given a copy of the CSPA (Version A) to complete, and were again told that it has been found to accurately measure one's core self. As the partici- pant filled out the assessment, the research assistant brought the results of the Q-sort to the primary researcher, who used them to prepare the participant's personality "results." At this point the participant was randomly assigned to one of three conditions. In the first condition, the results were generated such as to confirm the participant's preferred views based on nine of the traits rated highest on the Q-sort (six selected randomly from category five and three from category four). In Condition 2, the control group, results were generated as to neither confirm nor disconfirm the participant's preferred views, based on nine of the traits rated neutrally on the Q-sort (statements were selected randomly from category three). Finally, in the third condition, the results were generated to disconfirm the participant's preferred views based on nine of the traits rated lowest on the Q-sort (six selected ran- domly from category one and three from category two). After the participant completed the assessment, the research assistant (who had now re-entered the room) explained that she 50 methods now had to score the personality assessment and print out the results using a computer program. In actuality, the assistant went back to the primary researcher, waited for two minutes, and re- turned with two copies of the results that had already been gen- erated. The participant was then given one copy of the results, as the research assistant read the other copy out loud. This pro- cedure was performed in order to increase the effect of the ma- nipulation; to simulate the participant being viewed in a certain way by the experimenter. After receiving the results, the partici- pant was asked to fill out the Post-Assessment Questionnaire regarding the personality test, and was then debriefed. Debrief- ing was done both in writing and verbally by the primary re- searcher and the research assistant, and focused on explaining the false nature of the personality assessment, and all deception involved. The entire procedure was repeated for each partici- pant, with the only difference being the type of "results" re- ceived for the personality assessment. Results To test the previously stated hypotheses, first the multidi- mensionality and the reliability of the two quantitative depen- dent measures were assessed. Next, multivariate and univariate statistics were employed to test for significant effects of the manipulation for participants' responses on both parts of the Post-Assessment Questionnaire. Follow-up tests were performed when necessary. Scale Multidimensionality Analyses to examine multidimensionality were performed for the dependent measures used in this study, as neither mea- sures were published previously. The first part of the Post-As- sessment Questionnaire was designed specifically to assess par- ticipants' attitudes toward the Core Self Personality Assessment (CSPA). To assess multidimensionality for this attitude scale, a principal components factor analysis was conducted for the nine items on the scale. The findings revealed three components which Preferred View and Therapy 51 were interpretable. The first component was interpreted as a gen- eral factor incorporating attitudes toward the assessment, and was comprised of all nine items on the scale. The second com- ponent was comprised of the three items from the portion of the scale related to the environment of the CSPA, and the third was comprised of the three items from the portion of the scale spe- cifically related to the results of the assessment. Thus, the three interpreted factors consisted of the CSPA General factor, the CSPA Environment factor, and the CSPA Results factor. The 24-item Psychological Discomfort Scale (Elliot, personal communication) that was used for the second part of the Post- Assessment Questionnaire was also analyzed using a principal component analysis. The findings for this analysis revealed three interpretable components. The first was again a general compo- nent, comprised of 22 of the 24 items on the scale. The second component consisted of nine items, and was interpreted as a fac- tor involving items related to negative self-relevant affect (For example, items such as "angry at myself," "shame," and "disap- pointed with myself). The third component was comprised of the six items on the scale that were positively framed (i.e., re- verse-coded items; e.g., " content," "optimistic," and "happy"). Therefore, the three factors interpreted from the Psychological Discomfort Scale consisted of the General Psychological Dis- comfort (GPD) factor, the Negative Self-Relevant Affect (NSRA) factor, and the Positively Framed Items (PFI) factor. Scale Reliability To assess the inter-item reliability of both scales and the subscales created for each, a Cronbach's alpha coefficient was computed. After deleting one of the nine items on the scale to increase reliability, the general CSPA Attitude Scale demonstrated a reliability of a = .79. For the two CSPA subscales, reliability was also determined to be sufficient, a(CSPA Environment^ .88, a(CSPA Results)= .90. For the Psychological Discomfort Scale, it was determined that reliability was increased when six of the items were deleted, a(GPD)= .91, leaving the scale with 52 methods 16 remaining items. It was also determined that the reliability of the NSRA subscale was increased after deleting one of the items, a(NSRA)= .79, leaving the total number of items in this subscale at eight. Finally, reliability was determined to be increased in the PFI subscale after deleting one item, a(PFI)= .90, leaving five items in this subscale. Multivariate Analyses A one-way between groups Multiple Analysis of Variance (MANOVA) was conducted to examine specific differences in the six scales and subscales mentioned above (CSPA General, CSPA Environment, CSPA Results, GPD, NSRA and PFI). Mul- tivariate significance was demonstrated across this set of six dependent variables, Wilk's L(7,16) = .003, p. < .01. Univariate Analyses and Follow-up Tests Univariate analyses were performed to further examine the effects of the type of results participants were given. Signifi- cance was revealed for CSPA General, F(2,22) = 5.696, p_ <.01, and for CSPA Results, F(2,22) = 10.088, p. <.01. Differences in CSPA Environment were found to be non-significant, F(2,22) = .718, p. >.05. Differences in the scores for all subscales of the Psychological Discomfort Scale were found to be non-signifi- cant, GPD, F(2,22) = .977 ., p >.05; NSRA, F(2, 22) = 2.335, p >.05; and PFI, F(2, 22) = .176, p >.05. Post-hoc tests were performed to examine the significant differences between groups for CSPA General and CSPA Re- sults. A Tukey Honestly Significant Difference (HSD) proce- dure for CSPA General indicated that participants who received disconfirming results (M= 39.63, SD=8.43) exhibited a signifi- cantly lower rating score than those who received confirming results (M= 48.88, SD=5.14) or neutral results (M= 47.33, SD=3.24). For the CSPA Results subscale, a Tukey HSD proce- dure also revealed that participants who received disconfirming results (M= 10.25, SD=4.71) displayed a significantly lower rat- ing score than those who received disconfirming results (M= 18.38, SD=3.25) or neutral results (M=16.11, SD=3.14). Preferred View and Therapy 53 Discussion Based on the results of the present study, there was some support demonstrated for the predicted hypotheses. Hypothesis 1 was that participants whose preferred views were disconfirmed would report experiencing a higher level of psychological dis- comfort than those who had their preferred views confirmed. This hypothesis was not supported by the results of this study. On average, participants whose preferred views were disconfirmed reported a higher level of psychological discomfort than those whose preferred views were confirmed, particularly among the questions in the Negative Self-Relevant Affect subscale, although this difference was not found to be significant. The second hypothesis predicted was that participants whose preferred views were disconfirmed would report feeling more negatively toward the personality assessment in general, as com- pared to those whose preferred views were confirmed. This hy- pothesis was supported by the results of this study, based on how participants rated various aspects of the Core Self Personality Assessment (CSPA) after receiving their results. Overall ratings of the CSPA were significantly lower for participants whose pre- ferred views were disconfirmed than for participants whose pre- ferred views were confirmed or for participants in the control group, whose preferred views were neither confirmed nor disconfirmed. An examination of specific aspects of the CSPA that were rated was done through generating subscales of the CSPA attitude scale. The two subscales that were examined involved questions relating to the environment of the assessment and ques- tions relating to the results of the assessment. Significant differ- ences were found in the Results subscale, but not the Environ- ment subscale. Thus, the participants whose preferred views were disconfirmed rated the CSPA significantly lower than other par- ticipants, particularly regarding the results of the assessment. Implications of Preferred View This study was an attempt to test some of the notions re- garding the concept of preferred view (Eron&Lund, 1993,1996). 54 methods According to Eron and Lund (1996), these preferences that in- dividuals have about themselves are relevant to how people re- late to each other interpersonally. When individuals begin to see others viewing thern in ways that are discrepant with their pre- ferred views, they often experience negative emotions, which can lead to problem cycles between people (Eron & Lund, 1996). Thus, the effects that this kind of interaction can have on people have both intrapsychic and interpersonal components. The present study was an attempt to tap both of these components by confirming and disconfirming individuals' preferred views through the results of personality assessment. Based on the present study, the intrapsychic effects of disconfirming one's preferred view in this way appear to incon- clusive. As stated above, on average psychological discomfort was higher in participants whose preferred views were discontinued, particularly regarding negative self-relevant af- fect, such as shame, anxiety, and feeling disappointed or dis- gusted with oneself. These differences were not significant, how- ever. The CSPA attitude scale did have one item relating to af- fect ("The results of this assessment frustrated me"), which was also included in the Results subscale. Although significant dif- ferences were found for the Results subscale, this one question does not appear to be enough to accurately assess an intrapsy- chic component of preferred view. The results of the CSPA attitude scale confirmed that par- ticipants' ratings of the CSPA were lower if their preferred views were disconfirmed, particularly regarding the results of the as- sessment. This finding could indicate that participants who re- ceived results that disconfirmed their preferred views of them- selves attributed the results to a faulty assessment. Therefore, participants chose not to acknowledge the validity of results that were discrepant with how they prefer to see themselves. Consistent with Eron and Lund's (1996) description of how pre- ferred views function, this could indicate participants' discom- fort with seeing themselves in a way that is discrepant with their preferred views, or perceiving others as seeing them in this way. Preferred View and Therapy 55 Perhaps expressing disagreement with the assessment results was a way for participants to relieve stress about the results they received. Although the results of this study are far from conclusive, some important implications can be addressed. First, the find- ings of the study show that significant differences in attitudes toward the assessment were found when comparing participants who received disconfirming results to participants in the other two groups. Since no differences were found between partici- pants in the confirming results and the neutral results groups, it can be deduced that attitudes toward a given assessment are only affected when results are given that are discrepant with one's preferred view. This idea that negative attitudes toward an as- sessment-can arise from results that disconfirm one's preferred view leads to the question of how assessment results in general affect individuals. More importantly, can an assessment that disconfirms one's preferred view ultimately help the person bet- ter understand him or herself? It would seem that an assessment that disconfirms one's preferred view of self might not be help- ful to the person, or possibly even be harmful. Another important implication of the findings of this study relates to how they translate to the use of clinical diagnosis. Of- ten, diagnoses are given based on a clinician's interpretation of the results of a battery of personality tests and other types of assessment. Once again, the question of whether the use of diag- nosis is a helpful way of labeling clients presents itself. The present findings suggest that a diagnosis given to a client that is discrep- ant with his or her preferred view, it may not be found useful by the client. Furthermore, how many people would actually prefer to see themselves as depressed, delusional or paranoid? There- fore, clinical diagnosis, although often a necessary framework in today's society, should be used with care in actual practice. These implications are also important to be aware of when conceptualizing the relationship between client and therapist. Eron and Lund (1996) stress the importance of developing rap- 56 methods port with a client by treating him or her in a way that is consis- tent with his or her preferred view. Building on the relationship established with the client, the therapist can then work on clos- ing the gap between how the client prefers to be seen and how the client is actually behaving- The implications of the present study appear to be consistent with this way of working with a client. In fact, one might argue that the phenomenon of resis- tance occurs in therapy when the client perceives the therapist viewing him or her in a way that is inconsistent with his or her preferred view. Likewise, the participants in the present study whose preferred views were disconfirmed were being "resistant" toward the results of the personality assessment they were given. Thus, part of the art of performing successful therapy involves being aware of all that we are communicating to our clients through our perceptions of them. Study Limitations and Conclusions Although the current study did contain a good deal of limi- tations, only a few will be touched on here. It is important to note that this study was conducted as a starting point for future research on the idea of preferred view and its implications. Ob- vious problems with the study, such as its small sample size (N=25), use of non-standardized rating scales and lack of generalizibility are evident, and can be addressed with further research in this area. Carrying out a similar study over a longer period of time is recommended for finding more support for some of the issues addressed in this study. Other limitations of the present study involved using the Q-sort technique to measure individuals' preferred views and the use of a personality assess- ment to confirm and disconfirm preferred views. Because pre- ferred views consist of specific ways that people prefer to be seen, they are by nature very personal convictions. The items used on a Q-sort technique are often taken from common lists of traits, statements or generalizations made by people. In the present study, Anderson's (1968) list of personality trait words was used to generate items on the Q-sort. Rather than having Preferred View and Therapy 57 participants rate statements based on their preferred views, a more useful way of measuring preferred views might be to have par- ticipants generate preferred views in their own words. Isaacson and Landfield (1965) performed a study comparing Personal Construct descriptions (elicited using Kelly's Role Construct Repertory Test) with Butler-Haigh Q-sort statements. They found that the more meaningful language of the Personal Construct descriptions were more acceptable to participants in describing the self, especially for negative aspects of the self (Isaacson & Landfield, 1965). Therefore, using participants' own language to examine preferred views in this way appears to be a better way of studying this topic. Further, since the influence of pre- ferred view is most clearly demonstrated as a phenomenon that affects interactions between people, using personality assess- ment results to confirm and disconfirm preferred views may not be the best way to create a manipulation of this sort. A more useful way of confirming and disconfirming participants' pre- ferred views would be to do so using actual interactions between people, as this would make for a more "real" kind of interper- sonal situation in which preferred views seem to play a role. The results of the current study indicate that disconfirming one's preferred view through personality assessment results can negatively influence how the person feels about the assessment itself, particularly about the accuracy and usefulness of the re- sults. This finding has potential implications for how disconfirming individuals' preferred views can affect their atti- tudes toward assessment, diagnosis, and therapeutic interven- tions. The present study lends little support to the idea that these types of results affect one's level of psychological discomfort. Support for intrapsychic and interpersonal components of disconfirming one's preferred view was found to be inconclu- sive. A critique of the methodology of this study reveals that several changes can be made to find more conclusive results regarding the effects that preferred views have on how people view personality assessment. The study of the role that preferred views can play in people's lives is an area of research which can ultimately be of distinct benefit to clinicians. 58 methods References Anderson, N. H. (1968). Likableness ratings of 555 person- ality-trait words. Journal of Personality and Social Psychology, 9,272-279. Denzine, G. M. (1998). The use of Q-methodology in student affairs research and practice. Student Affairs Journal-Online [Online], 37 paragraphs. Available: http://sajo.org [1998, April 3]. Elliot, A J. (personal communication). As used in Elliot, A.J. & Devine, RG. (1994). On the motivational nature of cognitive dissonance: Dissonance as psychological discomfort. Journal of Personality and Social Psychology, 67, 382-394. Eron, J. B. & Lund, T. W. (1993). How problems evolve and dissolve: Integrating narrative and strategic concepts. Family Process, 32,291-309. Eron, J.B. & Lund, T.W. (1996). Narrative solutions in brief therapy^ New York: The Guilford Press. Fossum, T.A., Weyant, S.A., Etter, L., Feldman Barrett, L. (1998, May). A short and simple alternative to long personality questionnaires. Poster session presented at the annual meeting of the American Psychological Society, Washington, D.C. Higgins, E.T. (1987). Self-discrepancy: Atheory relating self and affect. Psychological Review, 94, 319-340. Higgins, E.T., Bond, R.N., Klein, R. & Strauman, T. (1986). Self-discrepancies and emotional vulnerability: How magnitude, accessibility, and type of discrepancy influence affect. Journal of Personality and Social Psychology, 51, 5-15. Isaacson, G. S. & Landfleld, A. W. (1965). Meaningfulness of personal versus common constructs. Journal of Individual Psychology, 21,160-166. Kelly, G. A. (1955). Psychology of personal constructs (2 volumes). New York: Norton. Laing, R. D. (1969). Self and others. London: The Camelot Press. Preferred View and Therapy 59 Laing, R. D., Phillipson, H. & Lee, A. R. (1966). Interper- sonal perception: A theory and a method of research. London: Tavistock Publications. Lund, T. W. & Eron, J. B. (1998). The narrative solutions approach for retelling children's stories. In M. F. Hoyt (Ed.), The handbook of constructive therapies: Innovative approaches from leading practitioners. San Francisco: Jossey-Bass. Markus, H. & Nurius, P. (1986). Possible selves. American Psychologist, 41,954-969. Rogers, C. R. (1951). Client-centered therapy. Boston: Houghton Mifflin. Rogers, C. R. (1961). On becoming a person: A therapist's view of psychotherapy. Boston: Houghton Mifflin. Stephenson, W. (1953). The study of behavior: Q-technique and its methodology. Chicago: The University of Chicago Press. Swann„W. B., De la Ronde, C. & Hixon, J. G. (1994). Au- thenticity and positi vity strivings in marriage and courtship. Jour- nal of Personality and Social Psychology, 66(5), 857-869. Acknowledgements The authors would like to thank Kristin Renstrom and Katy Sampson for their assistance with this project, and Glenn Geher, Larry Leitner, Doug Maynard and Caroline Stanley for their comments on earlier drafts of this paper.