Heterogeneity's ruses: some surprising effects of selection on population dynamics.

By James W. Vaupel, A. Yashin. 1985

"As a cohort of people, animals, or machines ages, the individuals at highest risk tend to die or exit first. This differential selection can produce patterns of mortality for the population as a whole that are surprisingly different from the patterns for subpopulations or individuals. Naive acceptance of observed population patterns may lead to erroneous policy recommendations if an intervention depends on the response of individuals. Furthermore, because patterns at the individual level may be simpler than composite population patterns, both theoretical and empirical research may be unnecessarily complicated by failure to recognize the effects of heterogeneity."

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. J. Lebreton
2. Olivier Gimenez
3. R. Pradel
4. C. Barbraud
5. Ashley B. Pitcher
6. J. Gaillard
7. E. Cam
8. T. Coulson

Citations in the corpus

• Using Simpson's Paradox to Discover Interesting Patterns in Behavioral Data (2018, background): “For example, a study (Vaupel and Yashin 1985) of recidivism showed that the rate at which released convicts return to prison declines over time.”
• Can you Trust the Trend?: Discovering Simpson's Paradoxes in Social Data (2018, background): “Vaupel and Yashin [20] give a compelling illustration of how survivor bias can shift the composition of data, distorting the conclusions drawn from it.”
• Computational social scientist beware: Simpson’s paradox in behavioral data (2017, background): “Vaupel and Yashin [23] give several illustrations of this effect.”
• Repeat burglary victimisation: a tale of two theories (2008, background): “Vaupel and Yashin 1985 ) that the analysis of waiting times for heterogeneous populations (with timestable risks) can produce surprising effects not dissimilar to that of the time course of repeat victimisation.”
• Maternal age affects offspring lifespan of the seed beetle, Callosobruchus maculatus (2003, methodology): “The observed increase in offspring lifespan is not an artefact of genetic heterogeneity within this population ( Vaupel & Yashin 1985 ); we specifically tested for a change in χ1”
• Mortality Modelling and Forecasting: a Review of Methods (2008, background): “Subgroups with higher growth rates come to dominate, resulting in increased average growth rates; similarly, a subcohort with high mortality rates will diminish in relative size, giving rise to declining average rates (Vaupel & Yashin, 1985).”
• On Quitting: Performance and Practice in Online Game Play (2017, unknown): “As a result, the trends observed in aggregated data may be quite different from those of the underlying subgroups (Vaupel and Yashin 1985).”
• Failing firms and successful entrepreneurs: serial entrepreneurship as a temporal portfolio (2013, background): “The problem of determining the rates of failure for entrepreneurs from that for firms is complicated with the usual confounding issues that plague cohort analysis over heterogeneous groups (Haunsperger and Saari 1981; Vaupel and Yashin 1985).”
• Why do higher education students drop out? Evidence from Spain (2008, background): “…is, a step function of t. Failure to control for any unobserved individual-specific effects that may affect the hazard function can produce severe bias in the structural estimates of duration models (see, e.g., Allison, 1982; Lancaster, 1990; Sheps and Menken, 1973; Vaupel and Yashin, 1985).”
• Capture–recapture models with heterogeneity to study survival senescence in the wild (2010, background): “Individual heterogeneity can lead to population-level patterns that are not always representative of the actual relationship at the individual level (Vaupel and Yashin 1985).”
• Exploring Theories of Victimization Using a Mathematical Model of Burglary (2011, background): “However, this perception may be illusory and instead reflect a type of statistical artifact that can occur when data are aggregated for groups of places that experience very different risks (for a more general discussion of heterogeneity's ruses, see Vaupel and Yashin 1985).”
• Measuring senescence in wild animal populations: towards a longitudinal approach (2008, background): “FOR INDIV IDUAL HETEROGENEITY Individuals dying during a given period may not represent a random sub-section of a population (Vaupel, Monton & Stallard 1979; Vaupel & Yashin 1985).”
• On Improving Telemetry-Based Survival Estimation (2006, background): “…may be assessed by plotting the baseline hazard function and determining if hazard declines over time, whereas further tests involve comparing hazard functions across relevant covariates where sample heterogeneity is suspected if hazard rates fail to follow similar trends (Vaupel and Yashin 1985).”