Measuring Cognition in a Multi-Mode Context

Although multi-mode studies may present complications for a variety of survey measures, tests of cognitive ability are particularly problematic. Measures of cognitive ability have been incorporated in many surveys, and they are especially common in longitudinal surveys of health and ageing. Despite these challenges, there are few mode comparisons that focus on cognitive measures, and the implications of mixed-mode design decisions for the measurement of cognition in longitudinal surveys remain largely unclear. In general, cognitive assessments in surveys share several qualities that complicate implementation across modes. If comparability of tests is a priority, it is important to identify tests that are suitable for administration across different modes. Regardless of how comparable the tests are in terms of administration protocol, however, there are still likely to be some measurement differences by mode.