• single rater bias;
  • multi-stakeholder constructs;
  • buyer-supplier relationships;
  • measurement error;
  • social capital;
  • dyadic data;
  • cross-prediction

As the global competitive landscape intensifies, firms have looked to their supply chain organizations to improve cost, visibility, and cycle time performance across functions, products, and markets. As a result, the scope of supply chain related operations have increasingly cut across organizational boundaries. To understand and capture such cross-organizational activities, researchers have broadened the focus of their studies and included multiple stakeholders in their analysis (e.g., integration, sustainability, and buyer-supplier relationships). However, multi-stakeholder research has also increased the complexity and effort required to conduct studies across organizational boundaries. Unfortunately, many studies that use multi-stakeholder constructs fail to fully address their multi-sided nature during both construct conceptualization and data collection. Several studies suggest that neglecting the multi-sided nature of certain constructs can affect the research validity and reliability and may invalidate research inferences and results, although such concerns have not been empirically demonstrated. The current study addresses this gap by performing a series of tests using data from 105 matched pairs of buyers and their suppliers to illustrate key methodological considerations for conducting multi-stakeholder research. This study also offers practical guidance regarding assumptions routinely made in single rater research and proposes when single rater data may be appropriate for multi-stakeholder research.