PSYCHOMETRIC PROPERTIES OF STUDENT-TEACHER RELATIONSHIP SCALE – SHORT FORM: CROATIAN VALIDATION STUDY FROM PRESCHOOL EDUCATION CONTEXT
Having in mind the importance of the relationship between children and their preschool teachers, the research about the quality of its measurement should present one of the main research focuses in overall early and preschool care and education studies. Previous studies clearly demonstrated that positive and close relationships between children and teachers resulted with higher level of socio-emotional and cognitive development of children in a long run. On the other side, the child-teacher relationship that is full of conflicts and bad feelings very negatively influence on children learning outcomes and developed behavioural patterns. Generally, children mental health is significantly determined by these early relationships with significant others in the educational context. Therefore, it is very important to develop and validate the measures that could contribute to understanding and measurement of children-preschool teachers’ relationship. One of that measure is Student-Teacher Relationship Scale, short form (STRS; Pianta, 2002). It is created with the purposes of measuring overall quality of the relationship between students (aged from 3 to 8) and their teachers. The relationship patterns that are originally should measure within the long form (28 items) are closeness, conflict and dependency. The short form is consisting of 15 items, which are measuring closeness (positive relationship) and conflict (negative relationship).
In this study, after expert translation, STRS-short form has been applied in the sample of 161 preschool teachers who rated their relationship with the vulnerable preschool children, i.e. with the preschool children living in risk life conditions: children who are members of discriminated groups (i.e. Roma minority), children from the social care, children from the violent families, children from low stimulating families, children with developmental disabilities and special needs, and children from families with chronically ill member(s). The sample of preschool teachers was from different kindergartens in Primorsko-Goranska County in Croatia.
The exploratory and confirmatory factor analysis with extraction method Principal Components and Varimax rotation with Kaiser Normalization were run. EFA resulted with four factors, which explained 59.555% of total variance. However, when two factors were specified, CFA results were psychometrically satisfactory. So, CFA confirmed hypothesized two-factor structure of STRS and explained 44.487% of total variance (the first factor Conflict explained after rotation 22.315% and the second factor Closeness explained after rotation 22.172% of the total variance). Rotated component matrix showed the lowest loading of the items four (“This child is uncomfortable with physical affection or touch from me”) what should be taken into account in future studies. Determined reliability levels Cronbach alpha were satisfying: for Conflict subscale α = .766 and for Closeness subscale α = .804. The determined findings have discussed within the suggested guidelines for future research studies: applying adjustments in translation; running focus groups with preschool teachers with the aim of creating culturally original STRS-items; and the application of the long form of STRS scale.