000			01727naa a2200181uu 4500
001			5082516122417
003			OSt
005			20190211160052.0
008			050825s2005 xx ||||gr |0|| 0 eng d
100	1		_aCOOK, Thomas D. _915852
245	1	0	_aEmergent principles for the design, implementation, and analysis of cluster-based experiments in social science
260			_aThousand Oaks : _bSAGE, _cMay 2005
520	3		_aIn experimentally designed research, many good reasons exist for assigning groups or clusters to treatments rather than individuals. This article identifies them and offers some principles about them. One emphasizes how statistical power and sample size estimation depend on intraclass correlations, particulary after conditioning on the use of cluster-level covariates. Another stress assigning experimental units at the lowest level of aggregation possible, provide this does not subtly change the research question. A third emphasizes the utility of minimizing and measuirng interunit communication, though neither is easy to achieve. A fourth advises against experiments that are totally black box and so leave program implementaion and process more salient. The last principle involves the utility of describing treatment heterogeneity and estimating its consequences, though causal conclusions about the heterogeneity will be less well warranted compared to conclusions about the intended treatment, every experiment's major focus.
773	0	8	_tThe Annals of The American Academy of Political and Social Science _g599, p. 176-198 _dThousand Oaks : SAGE, May 2005 _xISSN 00027162 _w
942			_cS
998			_a20050825 _b1612^b _cAnaluiza
998			_a20100803 _b1036^b _cCarolina
999			_aConvertido do Formato PHL _bPHL2MARC21 1.1 _c13424 _d13424
041			_aeng