The Influence of Sample Size on Parameter Estimates in Three-Level Random-Effects Models

Kerkhoff D, Nussbeck FW (2019)
Frontiers in Psychology 10: 1067.

Zeitschriftenaufsatz | Veröffentlicht | Englisch
 
Download
OA 1.01 MB
Abstract / Bemerkung
In educational psychology, observational units are oftentimes nested within superordinate groups. Researchers need to account for hierarchy in the data by means of multilevel modeling, but especially in three-level longitudinal models, it is often unclear which sample size is necessary for reliable parameter estimation. To address this question, we generated a population dataset based on a study in the field of educational psychology, consisting of 3000 classrooms (level-3) with 55000 students (level-2) measured at 5 occasions (level-1), including predictors on each level and interaction effects. Drawing from this data, we realized 1000 random samples each for various sample and missing value conditions and compared analysis results with the true population parameters. We found that sampling at least 15 level-2 units each in 35 level-3 units results in unbiased fixed effects estimates, whereas higher-level random effects variance estimates require larger samples. Overall, increasing the level-2 sample size most strongly improves estimation soundness. We further discuss how data characteristics influence parameter estimation and provide specific sample size recommendations.
Stichworte
random effects model; sample size; power analysis; three-level model; parameter estimation
Erscheinungsjahr
2019
Zeitschriftentitel
Frontiers in Psychology
Band
10
Art.-Nr.
1067
ISSN
1664-1078
eISSN
1664-1078
Page URI
https://pub.uni-bielefeld.de/record/2936020

Zitieren

Kerkhoff D, Nussbeck FW. The Influence of Sample Size on Parameter Estimates in Three-Level Random-Effects Models. Frontiers in Psychology. 2019;10: 1067.
Kerkhoff, D., & Nussbeck, F. W. (2019). The Influence of Sample Size on Parameter Estimates in Three-Level Random-Effects Models. Frontiers in Psychology, 10, 1067. doi:10.3389/fpsyg.2019.01067
Kerkhoff, D., and Nussbeck, F. W. (2019). The Influence of Sample Size on Parameter Estimates in Three-Level Random-Effects Models. Frontiers in Psychology 10:1067.
Kerkhoff, D., & Nussbeck, F.W., 2019. The Influence of Sample Size on Parameter Estimates in Three-Level Random-Effects Models. Frontiers in Psychology, 10: 1067.
D. Kerkhoff and F.W. Nussbeck, “The Influence of Sample Size on Parameter Estimates in Three-Level Random-Effects Models”, Frontiers in Psychology, vol. 10, 2019, : 1067.
Kerkhoff, D., Nussbeck, F.W.: The Influence of Sample Size on Parameter Estimates in Three-Level Random-Effects Models. Frontiers in Psychology. 10, : 1067 (2019).
Kerkhoff, Denise, and Nussbeck, Fridtjof W. “The Influence of Sample Size on Parameter Estimates in Three-Level Random-Effects Models”. Frontiers in Psychology 10 (2019): 1067.
Alle Dateien verfügbar unter der/den folgenden Lizenz(en):
Creative Commons Namensnennung 4.0 International Public License (CC-BY 4.0):
Volltext(e)
Access Level
OA Open Access
Zuletzt Hochgeladen
2019-08-15T09:16:24Z
MD5 Prüfsumme
c7ac15b89f0aecbcd86fed51dadad0d8

47 References

Daten bereitgestellt von Europe PubMed Central.

Simulation methods in structural equation modeling
Bandalos D., Gagné P.., 2014

Bates D.., 2006

Bates D., Mächler M., Bolker B., Walker S., Christensen R., Singmann H.., 2016
Context factors and student achievement in the IEA studies: evidence from TIMSS.
Caponera E., Losito B.., 2016
The impact of ignoring a level of nesting structure in multilevel mixture model: a monte carlo study.
Chen Q.., 2012
Design effects for sample size computation in three-level designs.
Cunningham TD, Johnson RE., Stat Methods Med Res 25(2), 2012
PMID: 23070588

Davidson R., MacKinnon J.., 1999
A priori power analysis in longitudinal three-level multilevel models: an example with therapist effects.
de Jong K, Moerbeek M, van der Leeden R., Psychother Res 20(3), 2010
PMID: 19946814
Power analyses for moderator effects in three-level cluster randomized trials.
Dong N., Kelcey B., Spybrook J.., 2018
The impact of the number of dyads on estimation of dyadic data analysis using multilevel modeling.
Du H., Wang L.., 2016
Sample size calculation for stepped wedge and other longitudinal cluster randomised trials.
Hooper R, Teerenstra S, de Hoop E, Eldridge S., Stat Med 35(26), 2016
PMID: 27350420

Hox J.., 1995
Multilevel regression and multilevel structural equation modeling
Hox J.., 2013

Hox J., Moerbeek M., van R.., 2017
Dyadic data analysis using multilevel modeling
Kenny D., Kashy D.., 2011

Kenny D., Kashy D., Cook W.., 2006
The power of the test for treatment effects in three-level cluster randomized designs.
Konstantopoulos S.., 2008
The persistence of teacher effects in elementary grades.
Konstantopoulos S., Chung V.., 2011
The influence of misspecification of the heteroscedasticity on multilevel regression parameter and standard error estimates.
Korendijk E., Maas C., Moerbeek M., Van P.., 2008
Classroom and school factors affecting mathematics achievement: a comparative study of Australia and the United States using TIMSS.
Lamb S., Fullarton S.., 2002
Power struggles: estimating sample size for multilevel relationships research.
Lane S., Hennes E.., 2018
Using diary methods to study marital and family processes.
Laurenceau JP, Bolger N., J Fam Psychol 19(1), 2005
PMID: 15796655
Robustness issues in multilevel regression analysis.
Maas C., Hox J.., 2004
Sufficient sample sizes for multilevel modeling.
Maas C., Hox J.., 2005
Understanding and estimating the power to detect cross-level interaction effects in multilevel modeling.
Mathieu JE, Aguinis H, Culpepper SA, Chen G., J Appl Psychol 97(5), 2012
PMID: 22582726
Accommodating Small Sample Sizes in Three-Level Models When the Third Level is Incidental.
McNeish D, Wentzel KR., Multivariate Behav Res 52(2), 2016
PMID: 28010127
A simulation study of sample size for multilevel logistic regression models.
Moineddin R, Matheson FI, Glazier RH., BMC Med Res Methodol 7(), 2007
PMID: 17634107
Power analysis
Murphy K.., 2011
How to use a Monte Carlo study to decide on sample size and determine power.
Muthén L., Muthén B.., 2002

AUTHOR UNKNOWN, 2013
The ecological fallacy.
Piantadosi S, Byar DP, Green SB., Am. J. Epidemiol. 127(5), 1988
PMID: 3282433
Growth curve modeling to studying change: a comparison of approaches using longitudinal dyadic data with distinguishable dyads.
Planalp E., Du H., Braungart-Rieker J., Wang L.., 2017

AUTHOR UNKNOWN, 2016
Estimating statistical power and required sample sizes for organizational research using multilevel modeling.
Scherbaum C., Ferreter J.., 2009
The effect of multicollinearity on multilevel modeling parameter estimates and standard errors.
Shieh Y.-Y., Fouladi R.., 2003
Power and sample size in multilevel linear models
Snijders T.., 2005
Extending the actor-partner interdependence model for binary outcomes: a multilevel logistic approach.
Spain S., Jackson J., Edmonds G.., 2012

AUTHOR UNKNOWN, 2015
Effect sizes in multilevel models
Tymms P.., 2004
Changing relations between phonological processing abilities and word-level reading as children develop from beginning to skilled readers: a 5-year longitudinal study.
Wagner RK, Torgesen JK, Rashotte CA, Hecht SA, Barker TA, Burgess SR, Donahue J, Garon T., Dev Psychol 33(3), 1997
PMID: 9149925
Accounting for variation in science and mathematics achievement: a multilevel analysis of Australian data third international mathematics and science study (TIMSS).
Webster B., Fisher D.., 2000
Longitudinal effects of school context and practice on middle school mathematics achievement.
Zvoch K., Stevens J.., 2006

Export

Markieren/ Markierung löschen
Markierte Publikationen

Open Data PUB

Web of Science

Dieser Datensatz im Web of Science®

Quellen

PMID: 31164847
PubMed | Europe PMC

Suchen in

Google Scholar