Purpose: Assessing complex skills is considered important but challenging. This study focused on developing assessment items to evaluate secondary social science students' proficiency in the subskill of causal analysis. Design/methodology/approach: Based on a conceptual framework of social scientific reasoning, we designed formative assessment items known as STAIRs (Social science Teaching Assessment Items of Reasoning). The STAIRs were validated in three focus groups: two groups of assessment experts (N = 7 and N = 3) and one group of social science teachers (N = 10). Additionally, think-aloud interviews were conducted with eight social science students. The quality of the STAIRs was evaluated by administering the items to 338 social science students in 21 Dutch social science classes. Findings: The results showed that it is possible to distinguish between the three performance levels in students' reasoning using the STAIRs. Practical implications: The design principles for the STAIRs may aid teachers in developing additional assessment items.