ABSTRACT
Background: Although the importance of the therapeutic alliance in the treatment process and health outcomes is recognized, so far, there has been no evaluation in the Peruvian context that considers possible individual differences that could influence this assessment. Purpose: This study assessed the psychometric properties of the WAI-S-P in a sample of individuals from Peru who are receiving psychological therapy. Furthermore, a network analysis was conducted to investigate the direct relationships between the therapeutic alliance and several relevant sociodemographic variables. Methods: The short version of the Working Alliance Inventory was used in a sample of 241 participants (Mage=32.58, SD=12.67) that had attended less than 6 sessions. Three models were considered, including a three-factor and a two-factor correlated model, as well as a bifactor model. In addition, a network of partial associations was created including the overall therapeutic alliance, sex, age, and number of psychotherapeutic sessions. Results: The bifactor model, with an overall therapeutic alliance factor and two specific factors ("contact" and "contract"), better fit the data. Invariance of the structure by sex and age showed equitable measurement. On the other hand, network analysis revealed a positive correlation between total session attendance and therapeutic alliance. Men reported higher therapeutic alliance, while women had higher total session attendance. Conclusion: The results of this study suggest that the therapeutic alliance is better represented by a bifactor model and demonstrates invariance across sex and age in Peruvian adults. Additionally, findings indicate that differences in life experiences and the sex of patients may need to be verified in future studies to better understand nuanced needs in forming therapeutic alliances at least in the early stages of session attendance.
