Psychometric properties of the Social Support Scale (SSS) in two Aboriginal samples

Abstract

In Australia, despite social support increasingly being reported as playing an important role in influencing health outcomes of Aboriginal and Torres Strait Islanders, measures of social support have not yet been validated for Aboriginal people. The current study aimed to evaluate the validity and reliability of the Social Support Scale in an Aboriginal and/or Torres Strait Islander population. The Social Support Scale (SSS) is a 4-item psychological instrument that was designed to evaluate four social support functions, instrumental, informational, emotional and appraisal support. Data included participants from two different samples: (1) Teeth Talk Study (n = 317), an oral-health randomized controlled trial (RCT) conducted with Aboriginal adults; and (2) the South Australian Aboriginal Birth Cohort Study (n = 367), a prospective longitudinal birth cohort study in which pregnant Aboriginal women were interviewed at baseline. The SSS psychometric properties were examined with Graphical Loglinear Rasch Models (GLLRM). The overall fit to a GLLRM was established (χ2(96)sample1= 52.7, p = 0.06; χ2(25)sample2= 22.2, p = 0.62) after accounting for local dependence between items 3 and 4. Item 2 displayed differential item functioning by employment status in Sample 1. Regarding dimensionality, the SSS was unidimensional in both samples (γobs1= 0.80; γexp1= 0.78, p = 0.65; γobs2= 0.75, γexp2= 0.77, p = 0.16). The instrument also displayed good reliability (Rsample1= 0.82, Rsample2= 0.84). Despite a few identified limitations (such as poor targeting), the findings indicated that the SSS is a promising instrument to provide culturally-valid and reliable measurement of social support among Aboriginal and/or Torres Strait Islander adults. Future studies should further investigate the instrument psychometric properties in other Aboriginal samples and the development and inclusion of culturally-sensitive items are also recommended.