When is tempo effect a tempo distortion?
Presentation at Population Association of America Annual Meeting, Detroit, 30 April
Publication date: 2009
Our aim is to clarify the fundamental conceptual difference between a tempo effect and a tempo distortion. We conclude that tempo effects are widespread in period measures, but that the need for adjustment depends on the use to which the indicator will be put. When the objective is to measure current conditions, as is usually the case, two approaches exist. In the classical approach current event rates are assumed to reflect fully current conditions by rising and falling over time, and tempo effects exist but are ignored. In contrast, we argue that current rates and the tempo effects in measures derived from them do not reflect current conditions when tempo changes are driven by changing conditions that shift event rate schedules to higher or lower ages. In this case, tempo effects are distortions that need to be corrected to get an accurate measurement of current conditions.