The relationship between counselor verbal response modes and the working alliance in career counseling. (Articles).
The authors examined the use of counselor response modes in career counseling, including the overall proportion of each response mode and the relationship of each response mode to the working alliance. Participants were 19 counselors-in-training and 26 clients in 78 sessions, making this the largest reported study of career counselor response modes. Seventy-nine percent of the counselor responses were information, paraphrase, or close-ended questions. None of the categorized counselor responses was- significantly and positively related to the working alliance, although counselor self-disclosure had a significant negative correlation with the alliance. The type of self-disclosure used was largely irrelevant or self-deprecating.
As Swanson (1995) noted in her review of career counseling process research, "We know a great deal about the effectiveness of career interventions in general, we know considerably less about career counseling specifically, and we know almost nothing about career counseling process" (p.217). Increasingly sophisticated met-a-analytic studies (e.g., Oliver & Spokane, 1988; Ryan, 1999; Spokane & Oliver, 1983; Whiston, Sexton, & Lasoff, 1998) have consistently found career counseling to demonstrate strong effect sizes, but little is known about what accounts for these positive outcomes.
In contrast to the specialization of career counseling, psychotherapy process research has been actively pursued since the 1940s. As Hill and Williams (2000) noted in their recent review of the psychotherapy process literature, "The first thing we learned from our literature review is that process research is alive and well with literally hundreds of articles to draw from" (p. 671). This accumulated empirical research forms a foundation to explain specifically what it is about the process of psychotherapy that makes it effective. Recent empirical evidence provided through content analyses of counselor case notes indicated that there is substantial overlap between the content of career counseling and that of psychotherapy (Anderson & Niles, 1995). Given this overlap, it is possible that the variables that have been found to be the most critical to outcome in psychotherapy might also be important in the conduct of career counseling.
One of the most consistently critical factors in the psychotherapy process literature is the working alliance, which has come to be seen as a necessary component of counseling, regardless of theoretical framework (Gelso & Fretz, 1992; Hartley & Strupp, 1983; Robbins, 1992). In fact, the working alliance has been shown to account for 30% to 50% of the variance in outcome measures across a range of studies (see, for example, Gelso & Carter, 1985; Horvath & Luborsky, 1993).
In addition, the importance and role of the working alliance in career counseling have begun to gain attention and recognition. Meara and Patton (1994) discussed the working alliance as a key element in enhancing the counseling of individuals who have career concerns. In 1995, Heppner and Hendricks investigated the role of the working alliance with an undecided and an indecisive college student, using a single-subject methodology; Heppner and Hendricks found that both clients endorsed the importance of the working alliance in helping them meet their counseling goals. In addition, two studies have used hierarchical linear modeling to explore the role of the working alliance in predicting career counseling outcomes (Heppner, Multon, Gysbers, Ellis, & Zook, 1998; Multon, Heppner, Gysbers, Zook, & Ellis-Kalton, 2001). Although they provide important information about the positive linear progression of the working alliance over time, none of these studies have delineated what occurs within career counseling that l eads to the development of a strong working alliance.
Although a number of definitions exist for the working alliance, the one that we used in this investigation is Bordin's (1979), which is the most heuristic, spawning numerous theoretical and empirical works (Kivlighan & Shaughnessy, 1995). Bordin defined the working alliance as a combination of three related components that determine the quality and strength of all helping alliances. These components are (a) goals, client and counselor agreement on the goals of treatment; (b) tasks, client and counselor agreement on the tasks to achieve these goals; and (c) bond, the development of a personal bond between the client and counselor.
Several studies have shown that there is a relationship between the working alliance and therapy outcome (see Gaston, 1990; Horvath & Symonds, 1991). In addition, although the importance of the working alliance is expected to wax and wane over the course of counseling, most researchers and counselors agree that the quality of the working alliance is especially important in the early phase of counseling (Gelso & Carter, 1985). Specifically, research has shown that the strength of the alliance after the third session relates positively to outcome (Hartley & Strupp, 1983; Horvath & Greenberg, 1986; Morgan, Luborsky, Crits-Christoph, Curtis, & Solomon, 1982; Saltzman, Luetgert, Roth, Creaser, & Howard, 1976). If the working alliance is not sound in this early phase, a poor outcome is likely to occur (Kokotovic & Tracey, 1990).
Thus, given the importance of quickly developing a strong working alliance, the question for career counselors becomes, What contributes to the development of the working alliance in the early phase of career counseling? Specifically, what career counselor interventions facilitate a strong working alliance?
One of the most productive ways of examining parallel questions in psychotherapy process research has been to study various counselor techniques. Counselor techniques can be defined as methods used by a counselor to facilitate positive behavior change in clients (Harper & Bruce Sanford, 1981). This definition implies that techniques are measurable entities and that techniques are used intentionally by counselors (Hill, 1992a, 1992b) The most common method of operationalizing counselor interventions has been through response modes (e.g., Hill, 1978; Porter, 1943; Robinson, 1950; Snyder, 1945; Stiles, 1979; Strupp, 1973), which refer to the counselor's intervention, independent of the topic or content of the speech (Hill, 1982). Response modes are measured in sentences or speaking turns (Hill, 1992a, 1992b).
One measure frequently used to assess and investigate counselor verbal response modes is the Hill Counselor Verbal Response Modes Category System (Hill, 1978, 1985, 1986). Developed from 11 existing response taxonomy systems, Hill's (1985, 1986) system includes 12 pantheoretical, nominal, mutually exclusive therapist verbal response modes.
The purpose of our study, which was primarily exploratory, was to use methodologies from psychotherapy research to investigate the following five research questions:
1. What verbal responses do counselors-in-training use in career counseling?
2. How similar are these verbal responses to those found in the one other career counseling study that was undertaken to analyze counselor response modes?
3. Given the importance of the working alliance to counseling outcome, which of the verbal response modes were related to the strength of the working alliance in career counseling?
4. How do findings about helpful counselor response modes in the career counseling setting compare with those found in psychotherapy process research?
5. Are any counselor response modes negatively correlated with the working alliance?
The career counseling clients were community residents who voluntarily came to the university's career center with a primary career issue (e.g., career dissatisfaction, career indecision, loss of job), and many also had related personal concerns (e.g., financial issues, conflict with coworkers, feelings of anxiety, death of a spouse). To increase the generalizability of the results, clients were not screened. However, for our study, we only included clients who (a) agreed that the first three sessions could be audiotaped, (b) were audible on all three audiotaped sessions, and (c) completed a Working Alliance Inventory after the third session; in addition, the client and his or her counselor both consented to participate in he study. Although the practicum counselors were asked to audiotape all of their sessions, sometimes they did not tape one of the first three sessions, or the tapes were inaudible, thus eliminating a potential client. For example, rapes from 6 potential participants were discarded because t hey were inaudible.
The 26 clients (14 women, 12 men) who met the inclusion criteria ranged in age from 16 to 59 years; 1 participant did not report age (M = 30.32, SD 8.42). Regarding clients' racial/ethnic background, 18 self-identified as European American, 2 as other (e.g., biracial), and 1 self-identified as Latino/a American; 5 participants did not report racial data. Eleven were employed full-time, 7 were employed part-time, 6 were not employed, and 2 did not report employment status. Ten clients were enrolled as fulltime students, 4 were enrolled as part-time students, and 10 were not enrolled in school; 2 participants did not report their educational status. Three clients had master's degrees, 11 had bachelor's degrees, 2 had high school diplomas, 4 had completed some high school education, and 1 had a grade school education. Five participants did not report their level of education.
Counselors of the clients who met the inclusion criteria were 19 counseling psychology graduate students (14 women, 5 men) enrolled in a career counseling practicum at a large, midwestern university who volunteered to participate in this study. The age of the counselors ranged from 23 to 55 years (M = 29.66, SD = 8.31,); 13 were European American, 3 were African American, 2 were Latino/a, and 1 was Asian American. For all counselors, this was their first practicum that focused specifically on career-related concerns. They had all taken a graduate-level course in career information and process. The philosophy of the practicum has been outlined in Career Counseling: Process, Issues, and Techniques (Gysbers, Heppner, & Johnston, 1998). Essentially, a holistic approach to the client is emphasized, and the counselors learn that the development of the working alliance is a core component of the career counseling process.
Raters of the verbal response modes were 12 (8 women, 4 men) master's-and doctoral-level graduate students (ns = 7 and 5, respectively). These raters were selected from all of the students who volunteered (N = 16), after they had been interviewed by either the first or second author. Eleven of the participants were students in counseling psychology, and 1 was a graduate student in higher education who had taken several courses in counseling. Six had completed only a prepracticum course, 1 had completed one practicum, and the remaining students had completed two or more practica. It should be noted that in the prepracticum course, each graduate student saw one undergraduate client for a total of four sessions, and each session was observed live by a supervisor and two other prepracticum students. Therefore, each prepracticum student conducted four 50-minute sessions and observed eight 50-minute counseling sessions. (See Multon, Kivlighan, & Gold, 1996, for a description of how the prepracticum course is conduc ted.) Thus, all raters had some counseling experience. Raters completed their tasks as four groups of three.
Measures and Rating System
Demographic questionnaires. Separate demographic questionnaires were developed for both counselors and clients. The counselor form included ive questions that elicited information on participants' sex, age, race, and previous practicum experience. The client form included six questions that elicited information on participants' age, sex, race, employment status, student status, and educational level.
Working Alliance Inventory--Short Form (WAI-S). The short version of Horvath and Greenberg's (1986) 36-item WAI was used for this study. The WAI assesses the working alliance according to Bordin's (1979) formulations. Items for the short version (WAI-S) were obtained from a factor analysis of the WAI. Tracey and Kokotovic (1989) selected the four highest loading items from each of the three WAI subscales: Bond, Tasks, and Goals. The WAI-S has 12 items and exists in two parallel forms: counselor and client. Each item is rated on a 7-point, fully anchored response scale, ranging from never (1) to always (7). The items are summed, with higher scores indicating a stronger working alliance. Total scores range from 12 to 84. Alpha coefficients for the WAI-S Total scores from a sample of 124 client--counselor pairings were .98 for client ratings (Tracey & Kokotovic, 1989).
Tracey and Kokotovic (1989) found that the factor structure of the WAI-S was found to be the same as that of the original WAI. Thus, the WAI-S assesses a general overriding alliance dimension, as well as Bordin's (1979) specific Bond, Tasks, and Goal variables. The three subscales have been found to be highly intercorrelated (Tracey & Kokotovic, 1989); thus, we used only the Total score in our investigation. Only the client version of the WAI--S was used for two reasons. First, it was our intent to study the client's appraisal of the alliance because their appraisals are related to various counselor intentions. Second, the client version has been shown to be a better predictor of outcome than the counselor version (Horvath & Symonds, 1991). Good construct validity for the WAI has been established through multitrait-multimethod analyses (Horvath & Greenberg, 1986). Evidence of concurrent and predictive validity has been provided by significant correlations of WAI scores with other counseling relationship measu res of therapeutic outcome (Horvath & Greenberg, 1986; Horvath & Symonds, 1991).
Revised Hill Counselor Verbal Response Modes Category System (Hill, 1985, 1986). This system includes 12 pantheoretical, nominal, mutually exclusive therapist verbal response modes: approval, information, direct guidance, closed question, open question, paraphrase (which includes restatement, reflection, summary, and nonverbal referent), interpretation, confrontation, self-disclosure, minimal encourager, silence, and other. Elliott et al. (1987) found concurrent validity for these categories with those of other response modes systems. The categories in the Hill (1985) system can be subdivided into six larger clusters: minimal responses (minimal encouragers, silence), supportive interventions (approval), directive interventions (information, direct guidance), questions (closed questions, open questions) paraphrase (paraphrase), and interpretive interventions (interpretation, confrontation, self-disclosure). To correct for different amounts of talking, proportions rather than frequencies of response modes were used in all analyses (Hill, 1978; Hill et al., 1988).
Data collection. Counselors met with clients over the course of one semester. Before seeing any clients, counselors were informed of the study, and they completed a consent form and a demographic questionnaire. Either before or during their first session, clients also completed a consent form and a demographic questionnaire. To comply with the inclusion criteria, all clients in this study also had audible audiotapes from the first three sessions, and they completed a measure (WAI-S) of the perceived strength of the working alliance.
Preparation of the audiotapes. Because the unit of investigation in this study is aggregated data and not individual sessions, it is recommended that researchers use segments of sessions rather than entire tapes (Friedlander, Ellis, Siegel, Raymond, & Haase, 1988). Friedlander and her colleagues used a variety of data sets and process measures to test segment generalizability in different formats of counseling with different coding systems (including the system used in this study). They found that when sessions were examined as a group, or aggregate, even fairly small segments could represent the full session with a reasonable degree of accuracy. Whole sessions are necessary only when conducting individual case studies.
The audiotapes were cued to begin at a point that would allow the raters to listen to the middle 20 minutes of the tape. This segment was chosen because it represented the central part of the counseling session. The 78 audiotapes (i.e., three initial sessions per client) were randomly assigned to rating teams.
Rating of verbal response modes. Before completing ratings, raters received approximately 15 hours of training by the second author on using the Hill Counselor Verbal Response Modes Category System to code counselor responses. This training included the following five steps: (a) reading the manual for Hill's (1985) system (b) reviewing definitions of categories, (c) reviewing instructions given in the manual for categorizing, (d) completing practice transcripts provided in the manual, and (e) completing practice transcripts using three audiotapes that were collected for but not used in the present study.
Raters completed their tasks in four groups of three individuals. Two groups completed ratings on one half of the data; the other two groups completed ratings on the other half of the data. The following rating process was used. Each week, the members of each team listened to three 20-minute segments of three different sessions for one counselor - client pair, prepared transcripts of the counselor's responses for those sessions, and then rated all responses from the sessions while relistening to the audiotapes. The following week, the prepared transcripts and audiotapes were given to a second team member for rating. This rotation process was again repeated during a 3rd week. At the end of 3 weeks, all three team members met to discuss discrepancies and to decide on final response categories. In accordance with the manual, if two raters provided the same category for a response, the category was accepted. If all three raters disagreed about the category for a response, the response was discussed, and the team selected a category. The transcripts were only prepared once for each set of audiotapes to ensure that each member was rating the same list of responses. During the discussion in the 3rd week, any disagreements concerning response units in the transcripts were presented and examined. Any necessary alterations in responses and category assignments were completed at this time.
To investigate the representativeness of the chosen segment, counselor verbal responses for the chosen segment (middle 20 minutes) were correlated with responses for the whole session for one audiotape. Ratings for this analysis were performed by the second author and by two of the raters in accordance with rating procedures outlined in the manual for Hill's (1985) system. Proportions of each category were used to calculate the correlation of .97 (p < .01, df = 6). In addition, the relationship between the two rating teams was investigated. Specifically, interrater reliability was calculated using a conservative test, the kappa statistic (Tinsley & Weiss, 1975). The result was K = 0.82, indicating good interrater reliability.
In the career counseling sessions, the most frequent counselor verbal response was information (M = .35, SD = .14), with paraphrase (M = .17, SD = .09) and closed question (M = .17, SD = .09) the second most frequent responses. The remaining verbal responses were as follows: open question (M = .10, SD = .05), approval (M = .10, SD = .06), self-disclosure (M = .03, SD = .04), direct guidance (M = .05, SD = .05), interpretation (M = .02, SD = .02), and confrontation (M = .00, SD = .00). Client WAI-S total ratings ranged from 57 to 84 (M = 71.54, SD = 7.80), with the item level mean of 5.96 (SD = .65).
Correlations were performed to examine the relationship between counselor verbal responses and clients' WAI-S post--third session scores. The data from all three sessions for each client were collapsed into one observation for each category. Table 1 presents these correlations.
In order to examine the relation of various types of verbal responses to a potentially relevant characteristic of the counselor (i.e., experience level), proportions of counselor verbal responses and the clients' perceptions of the working alliance (i.e., the WAI-S) were correlated with counselors' previous practicum experience. Previous practicum experience was operationalized as the number of semesters enrolled in practicum. For the 19 counselors-in-training, previous practicum experience ranged from zero to three semesters (M = .57, SD = .83). Statistically significant positive correlations were found between counselors' previous practicum experience and open question (r = .65, p < .01) as well as the use of paraphrase (r .56, p < .01); that is, counselors with more experience tended to ask more open-ended questions and to paraphrase more. However, the correlation between the counselors previous practicum experience and the clients' perception of the working alliance after the third session was not signifi cant (r -.08, p .71).
Results of this investigation are important both from a methodological and an empirical perspective. From a methodological perspective, only one other study has been published, to date, that examined career counselor response modes (Nagel, Hoffman, & Hill, 1995). Nagel et al. used 4 counselor-client pairs and examined two sessions. Thus, the current investigation, which used 19 counselors-in-training and 26 clients across 78 client sessions, represents a substantial methodological advance. In addition, in order to add to the generalizability of our findings, the clients in this study were not recruited but rather were adults in career transition who voluntarily came to a career center. This also represents a substantial contribution in that actual career clients were studied in an actual career counseling setting. As Swanson (1995) noted after her review of the career counseling process and outcome research, one of the most surprising features of the career counseling literature was the
brevity of many of thc interventions and the frequent use of recruited or no-choice participants rather than actual clients. In fact, researchers have nor yet adequately evaluated career counseling. Studies that appear to be evaluations of career counseling often consist of a single session, are structured or didactic in nature, and focus on test interpretation or information giving. IN short, they do not seem to resemble career counseling. (p. 251)
In addition to methodological advances, this study also offers a number of important empirical findings. For example, we used clinically relevant, established measures and a categorization scheme that can be compared across studies. Thus, these results begin to form a knowledge base documenting what actually occurs during the career counseling process.
One major finding in this study is that in a relatively large sample of career counseling sessions, the three most frequently used counselor response modes (provision of information, paraphrase, and closed question) accounted for 79% of the total counselor responses. It should be noted, however, that counselor verbal responses were analyzed for the first three sessions of career counseling. Some verbal responses, such as interpretation, would probably not be used that early in the counseling process, especially by novice counselors. However, because three or four sessions are typical of much of career counseling (Gysbers et al., 1998), it may be that these more advanced verbal response modes are used very infrequently in career counseling.
In examining this data, it would seem that having a larger repertoire of verbal responses might give the counselor-in-training greater flexibility in working with his or her clients. We do have evidence from this investigation that none of the most frequently used responses were significantly related to the working alliance. Whether they contribute to positive client outcome will require further investigation. Recently published data indicated that a full 60% of adult clients who sought career counseling were assessed as experiencing psychological distress, predominately high levels of depression, anxiety, and interpersonal sensitivity (Multon et al., 2001). Given this level of severity of symptoms along with complex career concerns, it would seem that having an expanded repertoire of verbal response modes might be very important. These results lend support to researchers who have long argued for greater competency in training on the integration of career counseling and psychotherapy (Manuele-Adkins, 1992; Ni les & Pate, 1989). One can question whether providing information, paraphrasing the client, and making extensive use of close-ended questions would be helpful with this complex constellation of client issues.
Our second research question was related to how comparable these data are with data from the only other study (Nagel et al., 1995) that examined counselor verbal response modes in career counseling. We compared our findings with those of Nagel et al. (1995), examining similarities and differences in career counselors' verbal response modes across two different samples. This comparison indicated that providing information was the most frequent response mode in both studies, with 42% in the Nagel et al. study and 35% in our investigation. In addition, paraphrasing also accounted for a large percentage of responses in both studies (i.e., 14% in the Nagel et al. study and 17% in the current investigation). Nagel et al. reported a greater use of direct guidance (i.e., 18% compared with 5% in our study). We also found that counselors used more close-ended questions than did the counselors in the Nagel et al. study (i.e., 17% in the current study compared with 9% in the Nagel et al. study). The three most infrequent ly used counselor response modes (i.e., interpretation, confrontation, and self-disclosure) were also the same across the two investigations. These three response modes comprised between 5% and 7% of the verbal responses in both studies. Nagel et al. concluded that the career counselors in their study were most similar to active, directive helpers (e.g., family practice lawyers, Albert Ellis, and radio talk show psychologists), which they determined on the basis of the number of response modes identified in other studies. Overall, they stated that the master's-level career counselors in their investigation focused primarily on action-oriented interventions with less focus on more intrapsychic interventions, at least in middle sessions. The current investigation, with a much larger sample of career counseling sessions, offers support for Nagel et al.'s conclusions. Counselors in our study were very active and directive, giving information, paraphrasing the client's feelings and content, as well as using an int erviewing style that included close-ended questions.
Next, given the importance of the working alliance to counseling outcome, we were interested in determining which of the verbal response modes were positively related to the strength of the working alliance in career counseling, and in comparing those findings to those in psychotherapy process research. We found that none of the verbal response modes categorized in this investigation were significantly and positively related to the strength of the working alliance in career counseling. Although a strong working alliance seems to have been built between client and counselors in this study, this relationship was apparently unrelated to most of the counselors' response modes. Thus, it seems that variables other than counselor verbal responses were responsible for building the working alliance in a positive direction.
The fourth research question asked for a comparison of counselor response modes to ones that have been found to be helpful within psychotherapy process research; the fifth research question asked whether any counselor response modes were negatively correlated with the working alliance. In her review, Hill (1992a, 1992b) indicated that interpretation, self-disclosure, paraphrase, and approval seemed most consistently helpful to clients. In contrast, the current study found no significant relationship between the working alliance and paraphrase, interpretation, or approval. Furthermore, self-disclosure was negatively correlated with the clients' rating of the working alliance. Thus, our findings suggest that the same type of counselor response modes that may be helpful in psychotherapy may be less so in career counseling. It is important to note, however, that the Hill Counselor Verbal Response Modes Category System (Hill, 1985, 1986) was not designed specifically for career counseling and may not be sensitive to some aspects of career counseling. In addition, Hill (1992a, 1992b) cautioned readers that findings are tentative and should not be interpreted as hard and fast rules about the use of counselor techniques. It is noted that some findings are from studies that have methodological shortcomings; thus, findings need to be replicated in investigations that use more rigorous designs. Overall, according to Hill 1992a, 1992b), although interpretations are helpful, ample evidence indicates that interpretations should be of moderate depth and related to a client's interpersonal difficulties. In essence, then, it is the kind of interpretations rather than the number that is important. Also, Hill (1992a, 1992b) stated that there is tentative evidence to indicate that an effective counseling technique is to use supportive interpretations after confrontation. Counselors in the present study appeared to use no confrontations across the 78 sessions. Perhaps this was because confrontation is an advanced skill, and all couns elors in this study were novice career counselors.
Another important contrast between what has been demonstrated to be helpful in psychotherapy process research and the findings in the current investigation is in the area of counselor self-disclosure. The use of counselor self-disclosure has been investigated in many studies within psychotherapy process research (see Hill & Williams, 2000). Self-disclosure is defined as "any statement in which the therapist reveals something personal about herself or himself to the client" (Hill & Williams, 2000, p. 675). In their review of the self-disclosure literature, Hill and Williams reported that across studies, self-disclosure was used infrequently. They reported that self-disclosure ranged from 1% to 4% of the verbal responses across studies, which is comparable to our investigation in which it was used 3% of the time and Nagel et al.'s (1995) study, in which it was used 2% of the time. In their summary of the combined studies within psychotherapy research Hill and Williams reported the following.
The evidence indicates that therapist self-disclosure is often effective, resulting in high ratings of helpfulness, high levels of client experiencing, and new insight or perspective. Furthermore, self-disclosure can make therapists seem more real or human, improve the therapeutic relationship, make clients feel more normal or reassured, lead to symptom relief, and lead to greater lilting of the therapist. (p. 676)
Given this glowing summary of the effectiveness of counselor self-disclosure in psychotherapy research, it was somewhat perplexing to find a strong, significant negative correlation between counselor self-disclosure and the strength of the working alliance in the third session. Because this was the first investigation to examine the relationship of career counselor self-disclosure and the working alliance, it is difficult to know whether this finding was unique to this study. It may be a difference in what career clients, as compared with psychotherapy clients, find useful in building an alliance with their counselor.
Although future studies will need to examine the impact of counselor self-disclosure on the formation of the working alliance in career counseling, we can hypothesize about some possible explanations for this finding. For example, one possible explanation for this apparent incongruity may be that, as Hill (1992a) argued, "helpfulness of different response modes appears to be dependent on the therapist intention, the type of client, the client's presenting problem, and client's experiencing level immediately before the therapists intervention" (p. 729).
In addition, there seem to be different forms of self-disclosure, some perhaps more helpful than others. Using categories from previous research, Hill, Mahalik, and Thompson (1989) found that clients rated reassuring disclosures as more helpful than challenging disclosures. An examination of the specific self-disclosures that were used by the counselors in our study revealed that few of the disclosures seemed to be intended to reassure or confront. Most of them seemed to be what Hill and O'Brien (1999) discussed as self-disclosures that helpers use to satisfy their urge to reveal themselves, rather than to intentionally help clients gain insight. Greenberg, Rice, and Elliott (1993) called such impulsive helper openness "promiscuous self-disclosure." The result of using such self-disclosure is that the clients feel uncomfortable and lose respect for the helper. Some examples of nonhelpful self-disclosures from our study include "That's where I did my undergrad, I remember how fun it was"; "I'm personally going to register with them just because I'd love to hear about something somebody says that gets them the perfect job"; "I have another client whose family is from Phoenix"; and "I think I am putting this spin on it because I've experienced the same thing in my life." This kind of self-serving self-disclosure may very well make the client feel uncomfortable and that perhaps their counselor is not attending to them.
Perhaps even more disturbing in examining the actual transcripts were counselors who used self-deprecating self-disclosures. Examples include: "Oh, I'm distracted, I mean fragmented. My brain is going in so many different directions, I forgot where we were." "I'm a real big word person, like real anal retentive." "I'm a little uncomfortable critiquing people's work because it's like, what do I know?" It is clear that this kind of counselor self-disclosure would not lead to faith in the counselor and likely accounted for some of the variance in the negative correlation between self-disclosure and formation of the working alliance. Future investigations will need to explore whether these findings are generalizable to other career counseling samples.
There was a significant positive relationship between counselors who had previous practicum experiences in another setting and their use of both more paraphrasing and more open-ended questions. Thus, counselor experience was related to use of certain counselor response modes.
This investigation also provides potentially effective data for integrating science and practice (Heppner, Multon, & Gysbers, 2001). Findings from this study could serve to inform the training of career counselors. For example, the data on self-disclosure alone could prompt an excellent discussion of the appropriate uses of self-disclosure. Emphasis could be placed on previous findings that indicate that self-disclosure has been found to be useful in forming a solid working alliance, especially when self-disclosures are used for reassuring the client. Data from the current study could also be presented to demonstrate inappropriate self-disclosure and how such self-disclosure can lead to a poorer alliance formation. In addition, some research has been conducted on counselors' covert thoughts or self-talk during sessions. Williams, Judge, Hill, and Hoffman (1997) examined what beginning graduate students describe as their covert self-talk, or reactions during counseling sessions and how they cope with these rea ctions. Some of the self-deprecating verbal responses that were used by counselors in this investigation might be discussed in this context. That is, beginning counselors are often nervous and have a range of self-talk, including the expression of feelings of inadequacy. However, there are other ways to manage these reactions, aside from revealing them to clients in the manner that was apparently done by some counselors in this study. Thus, it is expected that investigations of this type can provide nor only useful information from a research perspective, but they can also serve as a training tool to help inform career counseling practice.
Thus, it was the purpose of this study to examine the counselor verbal responses that are used predominately by counselors-in-training during career counseling. The findings from this relatively large sample study reinforce some of the findings from the one previous verbal response mode study in the career counseling literature in terms of the most used and least used response modes; however, the findings also diverged from some of the earlier findings. The most surprising finding was the negative relationship between counselor self-disclosure and the formation of the working alliance. Future research will be needed to further examine contributions to the working alliance in career counseling, including the impact of the counselors' level of experience in providing career counseling. We hope future research of this type will provide useful information concerning variables in the career counseling process that lead to the effective outcomes that have been reported consistently in the career counseling literatu re.
TABLE 1 Correlations Between Counselors' (N = 19) Verbal Responses and Clients' (N = 26) Working Alliance Inventory-- Short Form (WAI-S) Post--Third Session Scores Varbal Response Correlation approval -.00 Information -.11 Direct guidance .26 Closed question .27 Open question -.05 Paraphrase .11 Interpretation .01 Confrontation .00 Self-disclosure -.58 ** Note. df = 24. ** p < .01.
Anderson, W. P., Jr., & Niles, S. G. (1995). Career and personal concerns expressed by career counseling clients. The Career Development Quarterly, 43, 240-245.
Bordin, E. S. (1979). The generalizabililty of the psychoanalytic concept of the working alliance. Psychotherapy: Theory, Research and Practice, 16, 252-260.
Elliott, R., Hill, C. E., Stiles, W. B., Friedlander, M. L., Mahrer, A. R., & Margison, F. R. (1987). Primary therapist response modes: Comparison of six rating systems. Journal of Consulting and Clinical Psychology, 55, 218-223.
Friedlander, M. L., Ellis, M. V., Siegel, S. M., Raymond, L., & Haase, R. F. (1988). Generalizing from segments to sessions: Should it be done? Journal of Counseling Psychology, 35, 243-250.
Gaston, L. (1990). The concept of the alliance and its role in psychotherapy: Theoretical and empirical considerations. Psychotherapy, 27, 143-153.
Gelso, C. J., & Carter, J. A. (1985). The relationship in counseling and psychotherapy: Components, consequences, and theoretical antecedents. The Counseling Psychologist, 13, 155-243.
Gelso, C. J., & Fretz, B. R. (1992). Counseling psychology. Fort Worth, TX: Harcourt Brace Jovanovich.
Greenberg, L. S., Rice, L. N., & Elliott, R. (1993). Facilitating emotional change. New York: Guilford Press.
Gysbers, N. C., Heppner, M. J., & Johnston, J. A. (1998). Career counseling: Process, issues, and techniques. Boston: Allyn & Bacon.
Harper, F. D., & Bruce-Sanford, G. C. (1981). Counseling techniques: An outline and overview. Alexandria, VA: Douglass.
Hartley, D. E., & Strupp, H. H. (1983). The therapeutic alliance: Its relationship to outcome in brief psychotherapy. In J. Masling (Ed.), Empirical studies in analytic theories (pp. 1-37). Hillsdale, NJ: Erlbaum.
Heppner, M. J., & Hendricks, F. (1995). A process and outcome study examining career indecision and indecisiveness. Journal of Counseling & Development, 73, 426-437.
Heppner, M. J., Multon, K. D., & Gysbers, N. C. (2001, August). Informing career counseling training through programmatic empirical research. In K. M. O'Brien & M. Heppner (Co-Chairs), Career counseling training: Empirical models, social justice settings, diverse populations. Symposium presented at the annual meeting of the American Psychological Association, San Francisco.
Heppner, M. J., Multon, K. D., Gysbers, N. C., Ellis, C. A., & Zook, C. E. (1998). The relationship of trainee self-efficacy to the process and outcome of career counseling. Journal of Counseling Psychology, 45, 393-402.
Hill, C. E. (1978). Development of a counselor verbal response category system. Journal of Counseling Psychology, 25, 461-468.
Hill, C. E. (1982). Counseling process research: Philosophical and methodological dilemmas. The Counseling Psychologist, 10, 7-19.
Hill, C. E. (1985). Manual for Counselor Verbal Response Modes Category System (Rev. ed.). Unpublished manuscript, University of Maryland at College Park.
Hill, C. E. (1986). An overview of the Hill Counselor and Client Verbal Response Modes Category Systems. In L. S. Greenberg & W. M. Pinsof (Eds.), The psycho-therapeutic process: A research handbook (pp. 131-160). New York: Guilford.
Hill, C. E. (1992a). An overview of four measures developed to test the Hill process model: Therapist intentions, therapist response modes, client reactions, and client behaviors. Journal of Counseling & Development, 70, 729-737.
Hill, C. E. (1992b). Research on therapist techniques in brief individual therapy: Implications for practitioners. The Counseling Psychologist, 20, 689-711.
Hill, C. E., Helms, J. E., Tichenor, V., Spiegel, S. B., O'Grady, K. E., & Perry, E. S. (1988). Effects of therapist response modes in brief psychotherapy. Journal of Counseling Psychology, 35, 222-233.
Hill, C. E., Mahalik, J. K., & Thompson, B. J. (1989). Therapist self-disclosure. Psychotherapy, 26, 290-295.
Hill, C. E., & O'Brien, K. M. (1999). Helping skills: Facilitating exploration, insight, and action. Washington, DC: American Psychological Association Press.
Hill, C. E., & Williams, E. N. (2000). The process of individual therapy. In S. Brown & R. Lent (Eds.), Handbook of counseling psychology (3rd ed., pp. 670-710)). New York: Wiley.
Horvath, A. O., & Greenberg, L. 5. (1986). The development of the Working Alliance Inventory. In L. S. Greenberg & W. Pinsof (Eds.), The psychotherapeutic process: A resource handbook (pp. 529-556). New York: Guilford Press.
Horvath, A. O., & Luborsky, L. (1993). The role of the therapeutic alliance in psychotherapy. Journal of Consulting and Clinical Psychology, 61, 56 1-573.
Horvath, A. O., & Symonds, B. D. (1991). Relation between working alliance and outcome in psychotherapy: A meta-analysis. Journal of Counseling Psychology, 38, 139-149.
Kivlighan, D. M., Jr., & Shaughnessy, P. (1995). Analysis of the development of the working alliance using hierarchical linear modeling. Journal of Counseling Psychology, 42, 338-349.
Kokotovic, A. M., & Tracey, T. J. (1990). Working alliance in the early phase of counseling. Journal of Counseling Psychology, 37, 16-21).
Manuele-Adkins, C. (1992). Career counseling is personal counseling. The Career Development Quarterly, 40, 313-323.
Meara, N. M., & Patton, M. J. (1994). Contributions of the working alliance in the practice of career counseling. The Career Development Quarterly, 43, 161-177.
Morgan, R., Luborsky, L., Crits-Christoph, P., Curtis, H., & Solomon, J. (1982). Predicting the outcome of psychotherapy by the Penn Helping Alliance Rating Method. Archives of General Psychiatry, 39, 397-402.
Multon, K. D., Heppner, M. J., Gysbers, N. C., Zook, C. E., & Ellis-Kalton, C. (2001). Client psychological distress: An important factor in career counseling. The Career Development Quarterly, 49, 324-335.
Multon, K. D., Kivlighan, D. M., Jr., & Gold, P. B. (1996). Changes in counselor adherence over the course of training. Journal of Counseling Psychology, 43, 356-363.
Nagel, D. P., Hoffman, M. A., & Hill, C. E. (1995). A comparison of verbal response modes used by master's-level career counselors and other helpers. Journal of Counseling & Development, 74, 101-104.
Niles, S. G., & Pate, R. H. (1989). Competency in training issues related to the integration of career counseling and mental health counseling. Journal of Career Development, 16, 63-71.
Oliver, L. W., & Spokane, A. R. (1988). Career intervention outcome: What contributes to client gain? Journal of Counseling Psychology, 35, 447-462.
Porter, E. H., Jr. (1943). The development and evaluation of a measure of counseling interview procedures. Educational and Psychological Measurement, 3, 105-126.
Robbins, S. B. (1992). The working alliance. In M. Patton & N. Meara (Eds.), Psychoanalytic counseling (pp. 97-121). Chichester, England: Wiley.
Robinson, F. P. (1950). Principles and procedures in student counseling. New York: Harper.
Ryan, N. E. (1999). Career counseling and career choice goal attainment: A meta-analytically derived model for career counseling practice. Unpublished doctoral dissertation, Loyola University, Chicago.
Saltzman, C., Luetgert, M. J., Roth, C. H., Creaser, J., & Howard, L. (1976). Formation of a therapeutic relationship: Experiences during the initial phase of psychotherapy as predictors of treatment duration and outcome. Journal of Consulting and Clinical Psychology, 44, 546-555.
Snyder, W. U. (1945). An investigation of the nature of nondirective psychotherapy. Journal of General Psychology, 33, 193-223.
Spokane, A. R., & Oliver, L. W. (1983). The outcomes of vocational interventions. In W. B. Walsh & S. H. Osipow (Eds.), Handbook of vocational psychology (pp. 99-116). Hillsdale, NJ: Erlbaum.
Stiles, W. B. (1979). Verbal response modes and psychotherapeutic technique. Psychiatry, 42, 49-62.
Strupp, H. H. (1973). Foreword. In D. J. Kiesler (Ed.), The process of psychotherapy (pp. xiii-xvi). Chicago: Aldine.
Swanson, J. L. (1995). The process and outcome of career counseling. In W. B. Walsh & S. H. Osipow (Eds.), Handbook of vocational psychology: Theory, research, and practice (2nd ed., pp. 217-259). Mahwah, NJ: Erlbaum.
Tinsley, H. E. A., & Weiss, D. J. (1975). Interrater reliability and agreement of subjective judgements. Journal of Counseling Psychology, 22, 358-376.
Tracey, T. J., & Kokotovic, A. M. (1989). Factor structure of the Working Alliance Inventory. Psychological Assessment: A Journal of Consulting and Clinical Psychology, 1, 207-210.
Whiston, S. C., Sexton, T. L., & Lasoff, D. L. (1998). Career intervention outcome: A replication and extension of Oliver and Spokane (1988). Journal of Counseling Psychology, 45, 150-165.
Williams, E. N., Judge, A. B., Hill, C. E., & Hoffman, M. A. (1997). Experience of novice therapists in prepracticum: Trainees', clients', and supervisors' perceptions of therapists' personal reactions and management strategies. Journal of Counseling Psychology, 44, 390-399.
Karen D. Multon, Maryf. Heppner, and Norman C. Gysbers, Department of Educational, School, and Counseling Psychology, University of Missouri-Columbia; Carrie A. Ellis-Kalton, Neuropsychological Diagnostics, Inc., St. Louis, Missouri. Correspondence concerning this article should be addressed to Karen D. Multon, Department of Educational, School, and Counseling Psychology, 16 Hill Hall, University of Missouri, Columbia, MO 65211 (e-mail: MultonK@missouri.edu)