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

Julien G. A. Martin, Daniel H. Nussey, Alastair J. Wilson, Denis Réale

Research output: Contribution to journalArticle

140 Citations (Scopus)

Abstract

1. Interest in measuring individual variation in reaction norms using mixed-effects and, more specifically, random regression models have grown apace in the last few years within evolution and ecology. However, these are data hungry methods, and little effort to date has been put into understanding how much and what kind of data we need to collect in order to apply these models usefully and reliably.
2. We conducted simulations to address three central questions. First, what is the best sampling strategy to collect sufficient data to test for individual variation using random regression models? Second, on occasions when precision is difficult to assess, can we be confident that a failure to detect significant variance in plasticity using random regression represents a biological reality rather than a lack of statistical power? Finally, does the common practice of censoring individuals with one or few repeated measures improve or reduce power to estimate individual variation in random regressions?
3. We have also developed a series of easy-to-use functions in the ‘pamm’ statistical package for R, which is freely available, that will allow researchers to conduct similar power analyses tailored more specifically to their own data.
4. Our results reveal potentially useful rules of thumb: large data sets (N > 200) are needed to evaluate the variance of individual-specific slopes; a number of individuals/number of observations per individual ratio of approximately 0·5 consistently yielded the highest power to detect random effects; individuals with one or few observations should not generally be censored as this reduces power to detect variance in plasticity.
5. We discuss the wider implications of these simulations and remaining challenges and suggest a new way to standardize results that would better facilitate the comparison of findings across empirical studies.
Original languageEnglish
Pages (from-to)362-374
Number of pages13
JournalMethods in Ecology and Evolution
Volume2
Issue number4
Early online date22 Dec 2010
DOIs
Publication statusPublished - Aug 2011

Fingerprint

reaction norm
individual variation
experimental study
simulation
plasticity
researchers
ecology
sampling
field study
analysis
measuring
testing
effect
methodology

Keywords

  • individual variation
  • mixed-effects model
  • phenotypic plasticity
  • random regression
  • reaction norms

Cite this

Measuring individual differences in reaction norms in field and experimental studies : a power analysis of random regression models. / Martin, Julien G. A. ; Nussey, Daniel H. ; Wilson, Alastair J.; Réale, Denis.

In: Methods in Ecology and Evolution, Vol. 2, No. 4, 08.2011, p. 362-374.

Research output: Contribution to journalArticle

Martin, Julien G. A. ; Nussey, Daniel H. ; Wilson, Alastair J. ; Réale, Denis. / Measuring individual differences in reaction norms in field and experimental studies : a power analysis of random regression models. In: Methods in Ecology and Evolution. 2011 ; Vol. 2, No. 4. pp. 362-374.
@article{65c12e1738314650870cefb0b0d6334f,
title = "Measuring individual differences in reaction norms in field and experimental studies: a power analysis of random regression models",
abstract = "1. Interest in measuring individual variation in reaction norms using mixed-effects and, more specifically, random regression models have grown apace in the last few years within evolution and ecology. However, these are data hungry methods, and little effort to date has been put into understanding how much and what kind of data we need to collect in order to apply these models usefully and reliably.2. We conducted simulations to address three central questions. First, what is the best sampling strategy to collect sufficient data to test for individual variation using random regression models? Second, on occasions when precision is difficult to assess, can we be confident that a failure to detect significant variance in plasticity using random regression represents a biological reality rather than a lack of statistical power? Finally, does the common practice of censoring individuals with one or few repeated measures improve or reduce power to estimate individual variation in random regressions?3. We have also developed a series of easy-to-use functions in the ‘pamm’ statistical package for R, which is freely available, that will allow researchers to conduct similar power analyses tailored more specifically to their own data.4. Our results reveal potentially useful rules of thumb: large data sets (N > 200) are needed to evaluate the variance of individual-specific slopes; a number of individuals/number of observations per individual ratio of approximately 0·5 consistently yielded the highest power to detect random effects; individuals with one or few observations should not generally be censored as this reduces power to detect variance in plasticity.5. We discuss the wider implications of these simulations and remaining challenges and suggest a new way to standardize results that would better facilitate the comparison of findings across empirical studies.",
keywords = "individual variation, mixed-effects model, phenotypic plasticity, random regression, reaction norms",
author = "Martin, {Julien G. A.} and Nussey, {Daniel H.} and Wilson, {Alastair J.} and Denis R{\'e}ale",
year = "2011",
month = "8",
doi = "10.1111/j.2041-210X.2010.00084.x",
language = "English",
volume = "2",
pages = "362--374",
journal = "Methods in Ecology and Evolution",
issn = "2041-210X",
publisher = "WILEY-BLACKWELL",
number = "4",

}

TY - JOUR

T1 - Measuring individual differences in reaction norms in field and experimental studies

T2 - a power analysis of random regression models

AU - Martin, Julien G. A.

AU - Nussey, Daniel H.

AU - Wilson, Alastair J.

AU - Réale, Denis

PY - 2011/8

Y1 - 2011/8

N2 - 1. Interest in measuring individual variation in reaction norms using mixed-effects and, more specifically, random regression models have grown apace in the last few years within evolution and ecology. However, these are data hungry methods, and little effort to date has been put into understanding how much and what kind of data we need to collect in order to apply these models usefully and reliably.2. We conducted simulations to address three central questions. First, what is the best sampling strategy to collect sufficient data to test for individual variation using random regression models? Second, on occasions when precision is difficult to assess, can we be confident that a failure to detect significant variance in plasticity using random regression represents a biological reality rather than a lack of statistical power? Finally, does the common practice of censoring individuals with one or few repeated measures improve or reduce power to estimate individual variation in random regressions?3. We have also developed a series of easy-to-use functions in the ‘pamm’ statistical package for R, which is freely available, that will allow researchers to conduct similar power analyses tailored more specifically to their own data.4. Our results reveal potentially useful rules of thumb: large data sets (N > 200) are needed to evaluate the variance of individual-specific slopes; a number of individuals/number of observations per individual ratio of approximately 0·5 consistently yielded the highest power to detect random effects; individuals with one or few observations should not generally be censored as this reduces power to detect variance in plasticity.5. We discuss the wider implications of these simulations and remaining challenges and suggest a new way to standardize results that would better facilitate the comparison of findings across empirical studies.

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

KW - individual variation

KW - mixed-effects model

KW - phenotypic plasticity

KW - random regression

KW - reaction norms

U2 - 10.1111/j.2041-210X.2010.00084.x

DO - 10.1111/j.2041-210X.2010.00084.x

M3 - Article

VL - 2

SP - 362

EP - 374

JO - Methods in Ecology and Evolution

JF - Methods in Ecology and Evolution

SN - 2041-210X

IS - 4

ER -