| 英文摘要 |
Counseling supervision serves as the cornerstone of professional training in the counseling profession. Its operational quality is not only crucial for the development of counselors' practical competence but also directly influences client welfare. Supervision acts as a critical gatekeeping mechanism, ensuring that therapeutic services meet ethical standards and that the well-being of clients is protected. However, achieving these ideal functions is fraught with challenges. The supervision process involves complex interpersonal dynamics, power differentials, and contextual pressures, which can create significant difficulties for supervisors. Consequently, the issue of negative impacts resulting from supervision-such as harmful supervision-has gradually gained attention in recent literature. Research indicates that if difficulties within the supervision process are not properly addressed, they can lead to detrimental outcomes, impeding supervisee growth and potentially harming clients. Although previous studies have extensively explored supervision competence or the supervisory relationship, few have simultaneously focused on specific 'supervision difficulties' and the supervisors' 'coping efficacy' when facing these difficulties. Most existing measures focus on general self-efficacy rather than situational coping. In reality, whether a supervisor possesses the self-efficacy, defined as the belief in one's ability to effectively handle the situation, when facing difficulties is the key to transforming difficulties into opportunities for growth. To bridge this gap, this study aimed to develop the 'Counseling Supervision Difficulties and Coping Efficacy Scale' using the Double Matrix Model of Supervision as the theoretical framework. By employing this comprehensive model, which covers the therapeutic system, the supervisory system, and wider contextual factors, this study sought to develop a scale with good reliability, validity, and adaptability to the local culture, assisting supervisors in identifying difficulties and enhancing their confidence in coping. This study adopted an exploratory sequential mixed research design conducted in three phases to ensure the scale reflected local practices and possessed rigorous psychometric properties. In the preparation phase, the primary goal was to construct the content of the scale based on qualitative data. Six counseling supervisors were selected through purposive and snowball sampling for semi-structured in-depth interviews. The participants included five females and one male. In terms of educational background, five held Master's degrees in counseling, and one was a doctoral student. To capture a comprehensive range of experiences, the sample included supervisors at different developmental stages: two novice supervisors (with 1 and 3 years of experience), two intermediate supervisors (with 6 and 7 years of experience), and two expert supervisors (with 13 and 16 years of experience). The formal interview guide covered three main areas: (1) Clarification of the definitions of supervision difficulties and coping strategies; (2) An overview of the participants' supervision experience; and (3) A request for participants to recall at least 4 to 5 specific supervision experiences, describing the specific difficulty encountered, the impact it had on their work, and how they handled or coped with the situation. Thematic analysis yielded 59 meaning codes, 27 sub-themes, and 7 core themes. Integrating these qualitative results with literature, the researchers constructed an initial 72-item scale based on the framework of the Double Matrix Model of Supervision. Subsequently, five experts were invited to assess content validity (CVI), resulting in the removal of ambiguous items and a revised 61-item pilot scale. In the pilot phase, the revised scale was administered to counselors in Taiwan who had counseling supervision experience, yielding 97 valid samples. To ensure the quality of the scale, rigorous item analysis was conducted using multiple indices. The criteria for item retention included: (1) Normal distribution tests, where items were required to have an absolute skewness value less than 5 and an absolute kurtosis value less than 15; (2) Extreme group comparisons (Critical Ratio), where high and low scoring groups (top and bottom 27%) were compared to ensure significant discrimination; and (3) Homogeneity tests, which examined communality and the change in Cronbach'sαif the item was deleted Following item analysis, Exploratory Factor Analysis (EFA) was performed using principal component analysis with varimax rotation. Items with poor discrimination or insufficient factor loadings (less than .50) were removed, resulting in a formal scale of 56 paired items. Finally, in the formal phase, the scale was administered to a larger sample, yielding 160 valid samples. Confirmatory Factor Analysis (CFA) was used to test construct validity (including convergent and discriminant validity), and criterion-related validity was analyzed using the 'Counseling Supervision Professional Perception Scale' as the criterion. Statistical analysis results showed that the scale possesses excellent psychometric properties. Exploratory Factor Analysis (EFA) verified the factor structure for both the difficulties and coping efficacy scales. For the Supervision Difficulties Scale, seven factors were extracted, explaining a total of 56.54% of the variance. Specifically, Factor 1 (Supervisor Difficulties) explained 9.75%, Factor 2 (Supervisory Relationship Difficulties) explained 9.17%, Factor 3 (Case Understanding and Conceptualization Difficulties) explained 8.95%, Factor 4 (Supervisee Difficulties) explained 8.41%, Factor 5 (Contextual Factor Difficulties) explained 7.79%, Factor 6 (Counseling Strategies and Interventions Difficulties) explained 6.75%, and Factor 7 (Counseling Relationship Difficulties) explained 5.73%. Similarly, the Coping Efficacy Scale also yielded seven factors, explaining a total of 75.87% of the variance. This structure corresponds to the seven foci of the Double Matrix Model of Supervision. The formal scale demonstrated excellent internal consistency. The Cronbach'sαcoefficients were .95 for both the total Supervision Difficulties Scale and the total Coping Efficacy Scale. The subscale coefficients ranged from .82 to .89, indicating high stability across all dimensions. Regarding construct validity, the standardized factor loadings for all items were greater than .50, and both Average Variance Extracted (AVE) and Composite Reliability (CR) met the standards (AVE > .36, CR > .70), indicating good convergent validity. Discriminant validity analysis showed significant distinction between dimensions, as the confidence intervals of correlations between factors did not include 1. In terms of criterion-related validity, the analysis revealed a significant moderate negative correlation between overall supervision difficulties and overall professional perception (r = -.55, p < .001). This suggests that higher perceived difficulties are associated with lower professional self-perception. Conversely, a significant moderate positive correlation was found between overall coping efficacy and overall professional perception (r = .61, p < .001). This indicates that supervisors who believe in their ability to cope with difficulties tend to have a higher perception of their professional competence, validating the scale's effectiveness in measuring the intended constructs. Beyond developing a practical assessment tool, this study revealed unique cultural phenomena in Taiwanese supervision practices. The findings indicated a distinct pattern in how supervisors experience difficulties versus their efficacy in handling them. While supervisors felt challenged by difficulties in the dimensions of 'Professional Skills' (e.g., case conceptualization, strategies) and 'Regulations/Systems' (e.g., contextual factors), they still maintained relatively high coping efficacy in these areas. They possessed the 'Reason' (theoretical knowledge) and understood the 'Law' (institutional rules) to manage these tasks. However, once issues involved 'People' and 'Relationships'-specifically tension in the supervisory relationship, handling supervisee's personal issues, or managing countertransference-their coping efficacy was significantly lower. This result presents a hierarchy of efficacy: Reason > Law > Emotion. This hierarchy reflects the cultural scripts of 'Renqing' (human feelings) and 'Harmony' valued in Chinese society. In this cultural context, maintaining interpersonal harmony is paramount. The conflict between 'getting things done' (professional tasks) and 'maintaining relationships' (interpersonal harmony) creates a unique internal struggle for supervisors. Managing interpersonal conflict and relational tension represents the deepest challenge because specific cultural scripts for directly addressing conflict without damaging the relationship are often lacking. This scale contributes to the field by expanding the Double Matrix Model from a general analytical framework into a tool that reflects local cultural dynamics. It shifts the focus from merely listing problems to evaluating the 'confidence gap' in handling them. Practically, this scale can serve as a tool for supervisor self-awareness and reflection, helping them identify specific areas where their confidence may be low despite their competence. For supervision training institutions, it is recommended that future training expand from a purely competency-based approach (focusing on skills and theories) to a culture-relationship based approach. Training should explicitly assist supervisors in navigating the tension between professional gatekeeping and relational harmony, providing them with strategies to handle the 'Renqing' dynamics that are unique to the local context. |
本站僅提供期刊文獻檢索。
