gms | German Medical Science

GMS Psycho-Social-Medicine

Corporate journal of German scientific societies in psychosocial medicine

ISSN 1860-5214

Perceived self-competence and relationship experiences in inpatient psychotherapy: a pilot study

Kompetenz- und Beziehungserleben in stationärer Psychotherapie: eine Pilotstudie

Research Article

  • corresponding author Isa Sammet - Department of Psychosomatic Medicine and Psychotherapy, Medical Clinic, University of Tübingen, Germany
  • Steffen Häfner - Department of Psychosomatic Medicine and Psychotherapy, Medical Clinic, University of Tübingen, Germany
  • Eric Leibing - Clinic for Psychosomatics and Psychotherapy, University of Göttingen, Germany
  • Tim Lüneburg - Center of Psychiatry Herisau, Switzerland
  • Henning Schauenburg - Clinic for Psychosomatic and General Clinical Medicine, University of Heidelberg, Germany

GMS Psychosoc Med 2007;4:Doc04

The electronic version of this article is the complete one and can be found online at:

Published: April 5, 2007

© 2007 Sammet et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( You are free: to Share – to copy, distribute and transmit the work, provided the original author and source are credited.


Objective: The patient’s sense of capability in mastering future challenges (“self-competence”) represents an important therapeutic target. To date, empirical findings concerning the influence of the therapeutic relationship on perceived self-competence remain scarce. Against this backdrop, mutual associations between perceived self-competence, symptom distress and various relationship experiences within inpatient psychotherapy are investigated.

Methods: 219 inpatients with heterogeneous diagnoses completed the SCL-90-R, the Relationship Questionnaire RQ1 and the Inventory of Interpersonal Problems IIP prior to therapy. Self-competence and relationships to the individual therapist, therapeutic team and fellow patients were assessed weekly using an inpatient questionnaire (SEB).

Results: As expected, there were significant negative correlations between self-competence and symptom distress. Patients with more “fearfully avoidant” behavior upon admission experienced relationships during therapy as significantly more negative. Conversely, the quality of relationships to the individual therapist and fellow patients was predictive of a significant part of variance in self-competence upon discharge.

Conclusions: A model of mutual interactions is proposed for the variables under investigation. Results suggest that the positive association between the therapeutic relationship and symptom reduction could partly be explained by an improvement in perceived self-competence.

Keywords: self-competence, attachment, therapeutic relationship, interpersonal behavior, inpatient psychotherapy


Hintergrund: Handlungskompetenz zur Meisterung künftiger Probleme gilt als wichtige therapeutische Zielvariable. Über den Einfluss der therapeutischen Beziehung auf das Kompetenzerleben gibt es bisher kaum empirisch gesicherte Erkenntnisse. Vor diesem Hintergrund werden die Zusammenhänge zwischen Kompetenzerleben, Symptombelastung und Beziehungserfahrungen in stationärer Psychotherapie untersucht.

Methodik: Vor Therapie wurden bei 219 Patienten heterogener Diagnosen die Symptombelastung (SCL-90-R), der Bindungsstil (Relationship Questionnaire RQ1) und das interpersonelle Verhalten (Inventar Interpersonaler Probleme IIP) erfasst. Wöchentlich wurde das Erleben von Kompetenzerleben und Beziehungen zu Einzeltherapeut, therapeutischem Team und Mitpatienten (Stationserfahrungsbogen SEB) erhoben.

Ergebnisse: Es fanden sich -wie erwartet- negative Zusammenhänge zwischen Kompetenzerleben und Symptombelastung. Patienten mit stärker vermeidendem interpersonellen Verhalten zu Therapie-Beginn nahmen die Qualität der Beziehungen während der Therapie negativer wahr. Umgekehrt konnte die Qualität der Beziehungen zu Einzeltherapeut und Mitpatienten während der Therapie einen Teil der Varianz des Kompetenzerlebens zu Therapie-Ende voraussagen.

Schlussfolgerung: Vorgeschlagen wird ein Modell gegenseitiger Wechselwirkungen zwischen den erfassten Merkmalen. Unter anderem sprechen die Ergebnisse dafür, dass ein positiver Zusammenhang zwischen therapeutischer Beziehung und Therapieergebnis teilweise über die Verbesserung des Kompetenzerlebens zustande kommt.

Schlüsselwörter: Kompetenzerleben, Bindungsstil, therapeutische Beziehung, interpersonelle Probleme, stationäre Psychotherapie


The appraisal of one’s own competence in solving problems has a major impact on mental state. According to Bandura’s concept of self-efficacy [1], which conceptually corresponds to competence expectancy, expectancies can be domain-specific or of a generalized nature. Generalized competence expectancies influence behavior in specific situations and are, in turn, themselves influenced by specific situations [2]. The significant effect of generalized or domain-specific competence expectancies on diverse areas of human behavior (e.g. educational, occupational, sport or health-related behaviors) has been substantiated by an impressive number of studies [3], [4].

In the clinical setting, self-competence has long been recognized as one of the best predictors of mental state [5]. This is exemplified by the results of several recent studies: Improvement in self-competence is the most significant predictor of therapy success in the case of social anxiety [6]. Low levels of self-competence combined with perfectionism have a predictive value for episodes of binge-eating [7]. For alcohol-dependent patients, the willingness to commit to positive change is significantly associated with self-competence [8], as is the outcome of therapy in the case of adolescent drug addicts [9]. In the context of inpatient psychotherapy, positive competence expectancy is predictive of improvement in daily functioning [10], as well as reduction in the frequency and duration of hospital treatment [11]. Furthermore, the course of symptom development in psychotherapy inpatients is significantly related to the development of self-competence [12], with successful patients showing a greater increase in the confidence to solve own problems [13]. Already in 1994, after reviewing the current state of psychotherapy research, Grawe, Donati and Bernauer [14] came to the conclusion that the most effective factor of psychotherapy lies in the development of self-competence.

Against this backdrop of widespread evidence for the beneficial influence of competence expectancy, the question arises as to how an increase in self-competence can be achieved in psychotherapy. Bandura [1] assumes that self-competence is formed through learning experiences gained in mastering tasks and situations and comparing one’s own competence with that of others. In this context, significant others are thought to play a supportive role. They can provide emotional support, relay values, serve as role models in demonstrating constructive coping strategies and encourage an active exploration of the environment. Bowlby’s Attachment Theory [15] further contributes to an understanding of the role of significant others. The theory postulates that positive experiences of support in exploring the environment promote a positive “internal working model” with a self-image that embodies competence. This working model affects the building of relationships as well as the ability to pursue one’s own goals in the face of emotional strain. The development of attachment is thus a central element in the development of both social and generalized competence expectancy [16].

A number of studies have empirically investigated the relationship between attachment and self-competence. A comparably high level of self-competence was, for example, found in securely attached sixth-graders [17]. In studies based on student samples, social self-competence has been shown to mediate between attachment anxiety and feelings of loneliness [18] or attachment anxiety and avoidance [19]. In patients with depressive or agoraphobic disorders [20], one dimension of insecure attachment, namely the need for approval, has been found to be significantly correlated with self-competence.

It is in connection with this very aspect of positive experiences of support that the psychotherapist comes into play. The assumption that competence expectancy can be influenced by the therapeutic relationship, is found e.g. in Mallinckrodt’s Social Competencies in Interpersonal Process (SCIP) model [21], although the model also accepts that competence expectancy in turn influences the therapeutic relationship. However, as yet this has been subject to insufficient investigation. As significant others, therapists fulfil important functions in promoting current perceptions of self-competence as a foundation for competence expectancy in mastering future problems.

Research questions

The present study aims at addressing the mutual associations between perceived self-competence, symptom distress and relationship experiences in psychotherapy against the background of theories and empirical research. Inpatient therapy represents a suitable research setting, since it also permits an investigation of non-therapeutic relationships with fellow patients that are known to play a significant role in successful therapy. For the purpose of ensuring inter-individual comparability of results, perceived generalized self-competence with respect to “solving one’s own problems” is examined. This represents a key component of specific problem-solving competence [22], [23].

The first part of the present investigation can be classified as a replication study examining the presumed association between generalized self-competence and symptom distress. (Hypothesis 1a: Perceived generalized self-competence and symptom distress are significantly negatively related both before and after therapy. Hypothesis 1b: Pre-post differences in symptom distress and perceived self-competence are significantly negatively correlated). The second part investigates the nature of the link between perceived generalized self-competence and attachment style as well as current relationship experiences, which both have so far only rarely been addressed. (Hypothesis 2a: Interpersonal behavior at the start of therapy is significantly related to perceived self-competence during therapy. Hypothesis 2b: Patients’ relationship experiences with their therapist, the therapeutic team and fellow patients are significantly associated with perceived self-competence upon therapy completion). The third part of the investigation is devoted to examining the associations between the patients’ perceived self-competence, the relationship to their therapist and symptom reduction. (Hypothesis 3: The association between relationship experiences with the therapist and pre-post symptom relief is weaker when effects of perceived self-competence on symptom relief are controlled for). Positive effects of the patient’s relationship to his/her therapist on therapy success have been demonstrated in numerous studies [5]. The way in which these effects occur has, however, received little empirical attention. Results which support Hypothesis 3 could indicate that the therapeutic relationship has a positive effect on therapy success when for example it promotes perceived self-competence.



The hypotheses were tested using data routinely collected in the context of quality assurance efforts from inpatients prior to, during and after therapy (course of 12 weeks on average). All 228 patients receiving inpatient therapy at a University Clinic for Psychosomatic Medicine and Psychotherapy for at least 8 weeks over a period of 3 years (2002-2004) were included in the investigation. The minimal length of treatment (8 weeks), which was considered to represent an adequately long observation period, constituted the only exclusion criterion. Since data collection was integrated into ward routine and reliably carried out by the nursing team, complete sets of data were obtained for 96.1% of patients. Accordingly, N=219 patients were included in the final study sample. The data sets for the 9 excluded patients were incomplete due to a leave of absence or a temporary transfer to a different medical department on at least one of the survey occasions. Inpatient treatment in the clinic under investigation comprises a combination of group therapy led by a psychotherapist (2 x 90-minute sessions per week) and individual psychotherapy (2 x 50-minute sessions per week) with a psychodynamic emphasis and integrated aspects of behavior therapy and creative approaches (creative arts and body therapy; each 2 x 60-minute group sessions per week) which are led by specialized therapists. The therapeutic team further comprises nursing staff, that offers daily rounds for the purpose of reflecting upon the day (morning and evening rounds, each of half hour duration) and who also attend the patients’ ward meetings. Outside of therapy sessions, patients have the opportunity to meet in a recreation room. They are also able to plan and carry out joint leisure activities.

Sample characteristics are displayed in Table 1 [Tab. 1]. As is typical of inpatient-psychosomatic treatment, patients were predominantly female. The most frequent primary diagnoses were depressive and anxiety disorders, although a large proportion of multiple diagnoses were observed (on average 2.1 diagnoses per patient). The low proportion of university graduates/graduates from colleges of higher education and the high proportion of high school graduates is presumably due to the relatively high share of students (“still in education”).


Perceived self-competence and relationship experiences were assessed weekly using the “Stationserfahrungsbogen SEB”, a German inpatient questionnaire [24]. This questionnaire is routinely employed on the ward under investigation for the purpose of monitoring the therapeutic process of the patients. Three of the SEB scales assess relationship experiences during therapy over the previous seven days. These three scales comprise Alliance with the Individual Therapist (emotional closeness and the feeling of being understood), Alliance with the Therapeutic Team (general therapeutic atmosphere imparted by the team) and Perceived Attention from Fellow Patients (feeling of being supported by the group). Internal consistencies are stated as being “satisfactory to good” (Cronbach’s alpha αAlliance Therapist=.71; αAlliance Team=.85; αAttention Patients=.80).

A further scale measures self-competence using six-point items extracted from Schwarzer’s Generalized Self-Efficacy Scale [24] (Cronbach’s alpha α=.78). In light of the fact that the SEB is designed to evaluate the process of inpatient therapeutic experience in terms of a weekly account, Schwarzer’s original items were modified to address self-competence perceptions over the past week as opposed to self-competence expectations for the future. (Examples of the three items with the highest factor loadings: “I faced difficulties in a calm manner, because I knew I could rely on my coping abilities”. “Even in the case of unexpected events, I was confident that I could deal with things”. “I had the feeling that I was able to solve difficult problems”). This corresponds to the question of interest in the present investigation, pertaining to the connection between experienced relationships and perceived self-competence in the past (in each case in the previous week) based on the assumption that this perceived self-competence will have a major influence on self-competence expectations in the future.

At the onset and end of therapy (first and last days of therapy), the following self-rating instruments were applied: The German version of the Symptom Checklist SCL-90-R [25] for the assessment of general symptom distress, the Inventory of Interpersonal Problems IIP [26] for the self-evaluation of interpersonal behavior, and the Relationship Questionnaire RQ1 [27]. The RQ1 is based on a classification of self and other-models as positive or negative. Four attachment styles are defined: Secure (self and other-model are positive), Preoccupied (self-model is negative, other-model is positive), Dismissive-Avoidant (self-model is positive, other-model is negative) and Fearful-Avoidant (self and other-model are negative). Patients rate the extent to which each of the corresponding behaviors applies to them on a 7-point scale.

For the purpose of minimizing socially desirable responses, statements were treated anonymously by means of questionnaire coding. These codes were deciphered by the patients themselves on the very last day of therapy. The study was approved by the local ethics commission.

Statistical analysis

Associations between perceived self-competence and symptom distress at the start and end of therapy were evaluated using Pearson’s correlation coefficients. Effect sizes were computed for pre-post changes in symptom distress and self-competence.

The power of perceived self-competence at the onset of therapy in predicting relationship experiences with the individual therapist, the therapeutic team and fellow patients (SEB scales) was tested using regression analyses. For the evaluation of hypotheses pertaining to associations between relationship experiences and perceived self-competence during therapy, the mean value of the weekly measurements of each patient were employed as a basis for further analyses. This procedure was selected owing to the study’s investigative focus on the individual’s typical level and not the situative variability of a respective characteristic, the latter of which depends on factors such as setting (e.g. absence of the therapist due to vacation). An analysis of trends and intraindividual variability during therapy has been conducted elsewhere [12].

It was analyzed by multiple regression modeling in how far the attachment style and interpersonal behavior at onset of therapy and the relation experiences during therapy are predictive of the average perceived self-competence during therapy. Using version 12.0 of the statistical program SPSS, a stepwise regression analysis was carried out, in which variables were included in the regression model according to the strength of their correlation with the dependent variable as long as they proved to be relatively independent of other regressors and thus substantially contributed to an increase in predictive information.

In order to estimate the influence of perceived self-competence on the expected association between therapeutic relationships and changes in symptoms, partial correlations controlling for self-competence at the end of therapy were calculated between mean experienced alliance with the individual therapist, therapeutic team and fellow patients on one hand and symptom distress at the end of therapy on the other.


Perceived self-competence and symptom distress

Patients’ mean symptom distress (General Severity Index of the SCL-90-R) was 1.42 (SD=0.64) upon being admitted and 0.87 (SD=0.62) prior to being discharged. The resulting effect size of 0.9 can be interpreted as reflecting successful inpatient therapy. Women demonstrated significantly greater symptom distress at the onset of therapy (MWomen= 1.51, SD=0.60; MMen=1.19; SD=0.60; T=3.561; p<.005). No significant differences were observed at the end of therapy.

Self-competence values at the onset and end of therapy are displayed in Table 2 [Tab. 2] according to sex and educational qualification. The average value for perceived self-competence based on the entire sample at the onset of therapy was slightly below 3.50 on the 6-point Likert scale (values ranged from 1-6). This value represents a tendency towards somewhat unfavorable self-ratings in terms of perceived self-competence. Only a relatively small improvement (effect size ES=0.35) was found when comparing ratings at the end of therapy, although this improvement reflected a shift towards ratings of self-competence which tended to be positive. While women commenced therapy with a significantly lower level of perceived self-competence than male patients, they also tended (although not significantly) to profit more from therapy (ESWomen=.40; ESMen=.20), so that no significant differences were found between men and women upon completion of therapy. Patients with a lower level school leaving certificate also began therapy with significantly less perceived self-competence; this sub-sample was, however, not able to catch up with the more highly educated patients despite also improving over the course of therapy. With respect to primary diagnoses, no significant differences in initial or final levels of self-competence were found. This result was not surprising given the number of multiple diagnoses and associated multi-morbidity of patients within diagnosis groups.

Perceived self-competence in the total sample at the onset of therapy was slightly correlated and during therapy moderately correlated with symptom distress GSI of the SCL-90-R (rStart= -.314**; rEnd= -.580**). The change in perceived self-competence was moderately correlated with the pre-post-therapy change in symptom distress (rChange= .349**), i.e. symptom distress tended to abate more when self-competence increased and vice versa. These results support Hypotheses 1a and 1b.

Perceived self-competence and interpersonal behavior

Total values of interpersonal problems according to the IIP at onset of therapy (M=1.61; SD=0.57) and dismissive attachment behavior according to RQ1 (M=4.67; SD=2.10) were significantly negatively correlated with perceived self-competence at the start of therapy (rIIP-Total= -.308; p=.000; rDismissiveAtt= -.297; p=.000). Secure attachment (M=3.49; SD=1.88) was significantly positively correlated (rSecureAtt=.341; p<.005) with self-competence.

A stepwise regression analysis was performed in order to examine whether RQ1 and IIP scales at the onset of therapy helped to explain the variance in mean perceived self-competence during therapy. The resulting regression model is presented in Table 3 [Tab. 3]. As can be seen, only one sub-scale of the IIP (“FG” overly introverted/avoidant) and one scale of the RQ1 (“secure attachment style”) proved to be suitable predictors, although these alone were able to explain 22.9% of the variance in perceived self-competence. In summary, patients who at the onset of therapy rated themselves as securely attached, showed less avoidant relationship behavior, claimed to have less interpersonal problems and perceived themselves as more self-competent during the course of therapy (Hypothesis 2a).

On average, relationship experiences with the individual therapist and therapeutic team as well as attention received from fellow patients were rated positively with relatively small standard deviations (MIndTherRel=4.61, SD=0.75; MTeamRel=4.39, SD=0.73; MPatientAtt=4.38, SD=0.72 in each case on a scale from 1= very negative to 6= very positive). According to regression analyses, perceived self-competence at the onset of therapy contributed very little although significantly (p=.000) to the variance of these relationship ratings during therapy: R2 IndTherRel=.110; R2 TeamRel=.099; R2 PatientAtt=.107). Mean self-competence during therapy was significantly correlated (p=.000) with mean relationship experience with individual therapist (r=.439), relationship with the team (r=.540) and attention received from fellow patients (r=.502). A multiple regression analysis revealed that 28.2% of the variance in perceived self-competence was explained by the three relationship dimensions, with attention from fellow patients making the largest contribution (Table 4 [Tab. 4]). Regarding the variance of perceived self-competence at the end of therapy (Table 5 [Tab. 5]), relationship experiences contributed relatively little (18.9%) although significantly, i.e. patients who perceived relationships during therapy (in particular the attention received from fellow patients and alliance with the individual therapist) as less positive also experienced themselves during therapy as less self-competent and showed relatively low levels of perceived self-competence upon conclusion of therapy (Hypothesis 2b). With respect to perceived self-competence at the end of therapy, the influence of fellow patients diminished in comparison with that of the relationship with the individual therapist.

Partial correlations between symptom distress and mean perceived quality of relationships with the individual therapist, therapeutic team and fellow patients when controlling for self-competence are displayed in Table 6 [Tab. 6]. As expected, relationship quality was negatively correlated with symptom distress. This association remained significant, although reduced, when controlling for self-competence. This result is compatible with Hypothesis 3.


The present investigation has the character of a pilot study which examines associations between perceived self-competence, interpersonal characteristics and symptom distress. Prior to a discussion of the findings, methodological limitations will be addressed. First, the results are based on self-rating instruments and thus show only one side of the story. Therefore, future studies should include observer ratings to complement and complete the presented results. Second, while perceived self-competence and relationship experiences were assessed in an adequately reliable manner in the present study, future research should examine validity by employing additional measurement instruments. As an instrument designed to monitor changes by means of repeated measurement, the SEB (Experience of Inpatient Therapeutic Process Questionnaire) measures target constructs using relatively few items for the purpose of time economy. In light of this, it is to be particularly emphasized that the present results and their interpretations may only serve to generate hypotheses which must be further addressed in future studies. The necessity for further research is even greater, as the correlations in the present study are significant but low and thus only represent trends.

A statistically significant - though in terms of effect size (0.35) small - improvement in generalized self-competence was achieved across the therapy period. This result replicates the findings of a study by Waelte et al. [28], who obtained an effect size of 0.39 in a comparable inpatient therapeutic setting. These findings suggest that generalized perceived self-competence is a characteristic, which - to a limited extent - can be influenced in the short time frame of inpatient therapy (in the present study 12 weeks). It should be noted that the therapeutic promotion of self-competence in patients with a lower level school leaving certificate merits particular attention in future research; according to the present data, this group of patients began therapy with a comparably low level of perceived self-competence and failed to reach the initial level of patients with a higher level of education despite marked improvement by the end of therapy. In contrast, initial sex-specific differences to the detriment of female patients were compensated over the course of therapy. This finding appears to be indicative of particular therapeutic efficiency in the case of women.

The altogether relatively marginal improvement in perceived self-competence was, nonetheless, significantly correlated with pre-post symptom reduction. This relationship could provide an indication of the significance of perceived self-competence. Upon conclusion of therapy, the association between symptom distress and perceived self-competence proved larger, which could be explained by a therapeutically desirable sensitization with respect to the patient’s perception of his/her subjective self-competence experiences. The correlative results support the previously established assumptions regarding the relationship between perceived self-competence and mental state. Each of these two variables is likely to have a mutual influence on the other: On one hand the expectation that problems can be solved through one’s own competence can lead to a reduction in symptom distress. On the other hand, (a lack of) perceived self-competence can also result from (high) symptom distress, for example when strong symptoms cause patients to feel that they are not able to deal with problems at hand. Indeed, it has been empirically established that an abatement of depressive symptoms is paralleled by an increase in perceived self-competence [29].

The present findings tend to support assumptions derived from attachment theory concerning the relationship between perceived self-competence and interpersonal behavior or attachment characteristics. Moreover, the rather moderate amount of variance explained by the employed regression models would appear to signal the existence of other major determinants which were not incorporated in this study. Therapy-induced changes in the patient’s psychosocial environment could, for example, play a role (e.g. occupational reintegration measures, initiation of successive outpatient therapy, clarification of living situation etc.). This would correspond to a direct experience of success in coping with issues to be solved [1] and would thus constitute one of probably many factors which can improve perceived self-competence.

With respect to those aspects of relationship investigated in the present study, the following conclusions can be made: Patients perceive their self-competence in solving problems during therapy as comparatively lower, if they initially had relatively lower scores on the scale “Secure Attachment Style” and higher scores on the scale “Fearful-Avoidant Attachment Style” as well as more interpersonal problems, in particular in terms of fearful-avoidant behavior. From the perspective of attachment theory, this can be explained as follows: Fearful-avoidant patients have less opportunities to fall back on the support of others. This most probably exerts an inhibitory effect on the development of perceived self-competence, since situations within a social context - i.e. situations which provide the opportunity to receive affirmation and gather experiences through contact with others – are less frequented.

Remarkably, though not surprisingly, attention from fellow patients during therapy had the greatest predictive value for mean perceived self-competence as compared with relationships to the therapeutic team and the individual therapist. This underscores the fact that fellow patients represent a specific factor within inpatient therapy. This is presumably not exclusively limited to well-established group therapy effects (for example joint problem solving), but also ascribable to informal interactions outside of therapy sessions. With respect to perceived self-competence at the end of therapy, the influence of fellow patients is relativised to the extent that the significance of the relationship to the individual therapist at this stage of therapy is greater. In light of the fact that the three relationship scales (individual therapist, team, fellow patients) of the SEB address the “feeling of being supported and understood” in the social setting of the hospital ward, the results are collectively in line with attachment theory, according to which an atmosphere of relationship security forms the foundation for perceptions of self-competence [16]. The results of a study conducted with alcohol-dependent patients also proved consistent with this line of reasoning [30]. In this particular sample of patients, who received psychodynamically oriented group therapy, perceived self-competence was also seen to increase even when an improvement in this domain did not constitute the explicit therapeutic goal. This can be explained by drawing upon the assumption that perceived self-competence is best developed within an empathic atmosphere in which scope for personal responsibility, self-determination and tolerance of ambivalence is expansively provided [31].

On the other hand, indications that perceived self-competence at the onset of therapy influences perceptions of alliance in therapeutic (and non-therapeutic) relationships can also be found, although these are based on small correlations. Results of a study by Tschacher et al. [32] also support this relationship. Applying time series analysis methods to 91 courses of psychotherapy, the authors revealed that perceived self-competence governed subjective experiences of the therapeutic relationship. However, it was also observed that therapy outcome is positively influenced by those very process variables that increase self-competence. The therapeutic relationship was shown to be such a variable. Evidence that it has a positive effect on therapy success partly due to the promotion of perceived self-competence was also found in the present study. Correlations between the therapeutic relationship and symptom distress were reduced when the association of each of the two variables with perceived self-competence was controlled for using partial correlation analyses. Similar results stem from a study on alcohol-dependent patients [33] in which correlations between therapy success and both therapeutic empathy and attention were found to be no longer significant when perceived self-competence was partialled out.


Based on the findings of the present study, it can be argued that relatively small though significant associations exist between perceived self-competence and symptom distress on one hand and perceived self-competence and relationship experiences on the other. Regression analyses reveal that relationship experiences are able to partly predict variance in perceived self-competence and vice versa. The assumption of mutual effects between investigated variables appears to be plausible, as shown in detail in the discussion section above. Against this backdrop and as a result of the present pilot study, a hypothetical model can be proposed which requires systematic testing in subsequent studies.

This hypothetical model can be summarized as follows:

Perceived self-competence is a characteristic which to a limited extent can be developed within brief therapeutic processes. Though the change in perceived self-competence is small, it proves symptomatically relevant. While positive expectations with respect to one’s own competence can reduce symptom distress, high symptom distress can also lead to lower estimations of self-competence. In assuming such a circular effect, the promotion of perceived self-competence can be viewed as a central target variable within the therapeutic process.

One of the numerous factors which can contribute to promoting perceived self-competence during the therapeutic process is the therapeutic relationship. After all, perceived self-competence is formed in part in the context of interpersonal experiences according to the feedback principle: Fearful-avoidant behavior impedes positive perceptions of self-competence and negatively perceived self-competence leads to fearful avoidance. The therapeutic relationship represents a special social setting which offers favorable conditions for the development of self-competence through the establishment of a “good” atmosphere with basic acceptance and scope for personal responsibility and self-determination. The relationship between past perceptions of self-competence and future self-competence expectations is to be examined.

Independent of the validity of these still-to-be-assessed hypotheses pertaining to associations between relevant variables, clinical practice can benefit from the conclusion that, on account of its relationship with symptom distress, perceived self-competence represents a relevant indicator of mental state, which should thus receive adequate attention within the therapeutic process.


Conflicts of interest

None declared.


Bandura A. Self efficacy: toward a unifying theory of behavioural change. Psychol Rev. 1977;84:191-215.
Shelton S. Developing the construct of general self-efficacy. Psychol Rep. 1990;66:987-94
Bandura A. Self-efficacy. The Exercise of control. New York: WH Freeman & Company; 1997.
Schwarzer R, Jerusalem M. Das Konzept der Selbstwirksamkeit. Z Päd. 2002;48:28-53.
Bergin AE, Garfield SL. Handbook of psychotherapy and behaviour change. New York: Wiley & Sons; 1994.
Gaudiano BA, Herbert JD. Preliminary psychometric evaluation of a new self-efficacy scale and its relationship to treatment outcome in social anxiety disorder. Cognit Ther Res. 2003;27:537-55.
Bardone-Cone AM, Abramson LY, Vohs KD, Heatherton TF, Joiner TE. Predicting bulimic symptoms: An interactive model of self-efficacy, perfectionism, and perceived weight status. Behav Res Ther. 2006;44:27-42.
Demmel R, Beck B, Richter D, Reker T. Readiness to change in a clinical sample of problem drinkers: Relation to alcohol use, self-efficacy and treatment outcome. Eur Addic Res. 2004;10:133-8.
Burleson JA, Kaminer Y. Self-efficacy as a predictor of treatment outcome in adolescent substance use disorders. Addict Behav. 2005;30:1751-64.
Fliege H, Rose M, Bronner, Klapp BF. Prädiktoren des Behandlungsergebnisses stationärer psychosomatischer Therapie. Psychother Psychosom Med Psychol. 2002;52:47-55.
Fliege H, Rogalinski A, Schmid G, Kocalevent R-D, Lohmann R, Danzer G, Klapp BF. Evaluation stationärer psychosomatischer Behandlung anhand subjektiver und objektiver Kriterien. Z Psychosom Med Psychother. 2006;52:63-80.
Sammet I, Schauenburg H, Voges M. Gibt es unterschiedliche Verlaufstypen in der Stationären Psychotherapie. In: Vandieken R, Häckl E, Mattke D, editors. Was tut sich in der stationären Psychotherapie. Gießen: Psychosozial Verlag; 1998. p. 82-97.
Liebler A, Biella M, Schulz W, Hartmann U. Differenzielle Therapieverläufe und Behandlungserfolg in der stationären Gruppenpsychotherapie. Gruppenpsychother Gruppendyn. 2004;40:1-21.
Grawe K, Donati R, Bernauer F. Psychotherapie im Wandel. Von der Konfession zur Profession. Hogrefe: Göttingen; 1994.
Bowlby J. Eine Analyse der Mutter-Kind-Beziehung. München: Kindler; 1975.
Zimmermann P. Von Bindungserfahrungen zur individuellen Emotionsregulation: das entwicklungspsychologische Konzept der Bindungstheorie. In: Strauß B, Buchheim A, Kächele H. Klinische Bindungsforschung. Stuttgart: Schattauer; 2002. p. 147-61.
Coleman PK. Perceptions of parent-child attachment, social self-efficacy and peer relationships in middle childhood. Infant Child Develop. 2003;12:351-68.
Wei M, Russell DW, Zakalik RA. Adult attachment, social self-efficacy, self-disclosure, loneliness, and subsequent depression for freshman college students: A longitudinal study. J Couns Psychol. 2005;52:602-14.
Mallinckrodt B, Wei M. Attachment, social competencies, social support and psychological distress. J Couns Psychol. 2005;52:358-67.
Strodl E, Noller P. The relationship of adult attachment dimensions to depression and agoraphobia. Pers Relatsh. 2003;10:171-85.
Mallinckrodt B. Attachment, social competencies, social support and interpersonal process in psychotherapy. Psychother Res. 2000;10:239-66.
Schwarzer R. Measurement of perceived self-efficacy. Psychometric scales for cross-cultural research. Berlin: Freie Universität Berlin; 1993.
Scholler G, Fliege H, Klapp, B. Fragebogen zu Selbstwirksamkeit, Optimismus und Pessimismus. Psychother Psychosom Med Psychol. 1999;49:275-83.
Schauenburg H, Sammet I. Der Stationserfahrungsbogen - SEB. Fragebogen zur Erfassung des Verlaufes stationärer Psychotherapien. Psychother Psychosom Med Psychol. 1999;49:476-83.
Franke GH. SCL-90-R. Die Symptom-Checkliste von Derogatis - Deutsche Version. Göttingen: Beltz; 1995.
Horowitz LM, Strauß B, Kordy H. Inventar zur Erfassung Interpersonaler Probleme - Deutsche Version. Weinheim: Beltz; 1994.
Bartholomew K, Horowitz LM. Attachment styles among young adults: A test of a four-category model. J Pers Soc Psychol. 1991;61:226-44.
Wälte D, Ludwig-Becker F, Brandenburg U. Veränderung der kognitiven Selbstregulation bei Patienten mit psychosomatischen Störungen im Verlauf von stationärer Psychotherapie- eine kontrollierte Studie. In: Bassler M, editor. Leitlinien zur stationären Psychotherapie - Pro und Contra. Gießen: Psychosozial-Verlag; 2000. p. 335-62.
Weinmann M, Bader JP, Endrass J, Hell D. Self-efficacy beliefs and locus of control orientation in depressed inpatients. Z Klin Psychol Psychother. 2001;30:153-8.
Sandahl C, Gerge A, Herlitz K. Does treatment focus on self-efficacy result in better coping? Paradoxical findings from psychodynamic and cognitive-behavioural group treatment of moderately alcohol-dependent patients. Psychother Res. 2004;14:388-97.
Miller WR, Rollnick S. Motivational interviewing. New York: Guilford Press; 1991.
Tschacher W, Baur N, Grawe K. Temporal interaction of process variables in psychotherapy. Psychother Res. 2000;10:296-308.
Ritter A, Bowden S, Murray T. The influence of the therapeutic relationship in treatment for alcohol dependency. Drug Alcohol Rev. 2002; 21:261-8.