Development and validation of the workplace COVID-19 knowledge and stigma scale (WoCKSS) and its application in analysing stigma dynamics among factory workers using social network analysis

The COVID-19 pandemic not only posed major public health challenges but also intensified social issues such as stigma, particularly within workplace settings. In Malaysia, the manufacturing sector was disproportionately affected, creating an important context for examining workplace stigma. This study aimed to develop, validate and perform reliability assessment of the Workplace COVID-19 Knowledge and Stigma Scale (WoCKSS) and to investigate how workplace COVID-19 stigma is shaped and distributed through social networks. The study was conducted in two phases using a cross-sectional design. Phase 1 focused on the development and validation of WoCKSS, a Malay-language instrument comprising two domains, namely knowledge and stigma. The stigma domain was structured around three constructs, which were stereotype, fear and prejudice. Content and face validation were conducted with expert panels and target respondents. Construct validity was assessed using Item Response Theory (IRT) for the knowledge domain and Exploratory and Confirmatory Factor Analyses (EFA and CFA) for the stigma domain. The final instrument retained 14 knowledge items and 7 stigma items, each demonstrating strong psychometric performance, including acceptable item difficulty and discrimination, unidimensionality for the knowledge domain, and robust internal consistency based on McDonald’s omega. Phase 2 applied the validated instrument among 179 factory workers in Negeri Sembilan and incorporated both egocentric and sociocentric perspectives from Social Network Analysis (SNA). In the egocentric analysis, multiple logistic regression (MLogR) identified age and job position as significant predictors of workplace stigma, with younger and administrative employees reporting higher stigma levels. Binary logistic multilevel modelling (MLM) further indicated that the influence of alters on ego’s stigma-related judgements was shaped primarily by relationship type, with family members exerting the strongest influence. Sociocentric analysis was conducted using Exponential Random Graph Modelling (ERGM) and the Quadratic Assignment Procedure (QAP). ERGM results showed that tie formation was influenced by reciprocity and homophily based on gender, job position and stigma status. QAP demonstrated that dyads with similar eigenvector centrality and coreness were more likely to share the same stigma status, indicating that structural influence and embeddedness within the network core were important correlates of stigma similarity. In conclusion, the findings show that workplace COVID-19 stigma is shaped not only by individual attributes but also by relational structures and positional dynamics within workplace networks. The integration of psychometric validation with both egocentric and sociocentric network approaches provides a comprehensive understanding of how stigma is socially structured and sustained within organisational settings.