Mediators’ Skills for Trust Building Inventory. A Psychometric Networks Approach

Abstract

Trust building depends both on therapeutic alliance construction and the quality of the intervention. Although trust-building has received much attention in the literature, there are hardly any validated tools to assess it. In order to cover this gap, the main goal of this work is to present a psychometric inventory for measuring both the skills for building a therapeutic alliance and the skills for intervention through a mediation process. 170 subjects, mediators and clients, voluntarily participated in the validation study. An advanced approach by means of psychometric networks was used to estimate and test the two-factor hypothesized model: 1) Skills for building a therapeutic alliance, and 2) Skills for the intervention in the mediation process. An Exploratory Graph Analysis (EGA) with the Triangulated Maximally Filtered Graph (TMFG) was implemented for estimating the inventory latent network with all items. Further bootstrapping techniques were used for assessing the latent structure stability. A Confirmatory Factor Analysis (CFA) over the best fitted EGA model was applied with a robust estimator (WLSMV). Reliability analyses for the best-fitted model were also implemented. Results indicated a stable and well-fitted two-factor latent network model, which was also confirmed by CFA: (2scaled = 216.84, df = 169, p = .008, RMSEA = .039, CFI = .986, TLI = .984). All reliability indices for the two-factor model obtained adequate values (above .80) and all items provided adequate psychometric behavior. This new inventory can be useful for developing improvement professional practice training programs, and also for enriching mediation-related higher education curricula.