Marianne Welmers

Chapter 6 136 Alliance Complexity: The Matter of Perspective Another key finding that emerges from the studies included in this dissertation is that – especially in family treatment – the working alliance is not a simple, unilateral, objective construct, but rather a matter of perspective: establishing a judgement on the strength of the alliance depends on whose perspective is measured. Afirstmatter of perspective regards different informants (therapists, observers, and family members) reporting on the alliance. In line with prior research in individual youth and adult psychotherapy (for meta-analytic reviews see Roest et al., 2021b; Tryon et al., 2007), this dissertation (Chapter 3) showed that therapists generally reported weaker alliances as compared to familymembers and observers. These different ratingsmay reflect different frames of reference: therapists probably compare relationships with clients to those with previous clients or to a theoretical ideal of the therapeutic relationship, whereas clients may rather compare the therapeutic relationship to their interaction with friends, family members, teachers, etc. (Hartmann et al., 2015; Roest et al., 2021b; Tryon et al., 2007). Although these different (interpretation) frameworks may be a given, prior research in individual therapy indicates the importance of a more or less shared perspective on the strength of the alliance to promote treatment effectiveness (Bachelor, 2013; Fjermestad et al., 2016; Kivlighan, 2007; Rozmarin et al., 2008; Zilcha-Mano et al., 2017). Moreover, a recent family therapy study found that a shared alliance perception between the therapist and all family members in treatment was a predictor of improvement in child functioning and achieving family goals (Escudero et al., 2021). It was interesting to find in the meta-analysis (Chapter 2) that alliance informant did not moderate the alliance-outcome association, indicating that the effect of alliance on outcome is independent of whose perspective (i.e., therapist, observer, client) on the alliance is measured. However, the study presented in Chapter 5 did indicate an informant-effect: family members’ unbalanced observed early treatment alliances were positively associated with treatment outcome, whereas unbalanced therapist- or self- reported alliances were not. This finding implies that alliance discrepancies between family members as observed during an early treatment session (the interpersonal aspect of alliance) may be a marker of potential treatment success – as will be discussed further on in this chapter – whereas discrepancies in their inner perceptions of the alliance (the intrapersonal aspect) are not. Apparently, different reports, reflecting different aspects of the alliance, in turn may have different effects on outcome. The second important ‘matter of perspective’ in this dissertation is unique to systemic family treatment, namely: multiple alliances with different family members, each with their own perspective on the alliance. In Chapter 4, we found that the strength of the