Measuring individual differences in reaction norms in field and experimental studies: a power analysis of random regression models

摘要: Summary 1. Interest in measuring individual variation reaction norms using mixed-effects and, more specifically, random regression models have grown apace the last few years within evolution and ecology. However, these are data hungry methods, little effort to date has been put into understanding how much what kind of we need collect order apply usefully reliably. 2. We conducted simulations address three central questions. First, is best sampling strategy sufficient test for models? Second, on occasions when precision difficult assess, can be confident that a failure detect significant variance plasticity represents biological reality rather than lack statistical power? Finally, does common practice censoring individuals with one or repeated measures improve reduce power estimate regressions? 3. also developed series easy-to-use functions ‘pamm’ package R, which freely available, will allow researchers conduct similar analyses tailored specifically their own data. 4. Our results reveal potentially useful rules thumb: large sets (N > 200) needed evaluate individual-specific slopes; number ⁄ observations per ratio approximately 0AE5 consistently yielded highest effects; should not generally censored as this reduces plasticity. 5. discuss wider implications remaining challenges suggest new way standardize would better facilitate comparison findings across empirical studies.