Research on online teacher professional development (PD) increasingly draws on multiple sources of evidence, including self-report questionnaires, platform-based participation metrics, and, more recently, cognitive-task measures. While each of these approaches captures a different aspect of teacher learning and transfer, it remains unclear how they should be interpreted together within a single analytic framework. This concise paper addresses that question through a study of in-service Lithuanian informatics teachers enrolled in a nationwide online PD platform.

The study brings together three types of indicators: (1) Learning Transfer System Inventory (LTSI) dimensions reflecting perceived transfer conditions, motivation, and constraints; (2) cognitive-task measures indexing working memory, associative learning, and generalization; and (3) platform traces capturing patterns of uptake and engagement during the PD process. Preliminary analyses suggest that stronger cognitive performance is not reflected in uniformly higher transfer scores. Instead, associations appear at the level of particular LTSI dimensions, indicating that cognitive-task outcomes map onto selected aspects of transfer readiness rather than overall self-reported transfer conditions. This pattern is important methodologically: it suggests that collapsing across questionnaire dimensions or across cognitive-task outcomes may obscure the structure of the relationships under study. It also raises the broader question of how platform-based engagement indicators should be interpreted when they do not align neatly with either self-report or performance-based measures.

The discussion therefore focuses on how to build multi-indicator designs for studying teacher learning in digital environments. The paper argues that combining cognitive, self-reported, and behavioral data is valuable not because it produces a single definitive measure, but because it reveals different layers of the learning and transfer process. This has implications both for theory-building and for the design of online PD systems that aim to support a wider range of learner profiles.