SMITH, Herbert L <br/><br/>
Introducing new contraceptives in rural china :  a field experiment 

 - Thousand Oaks  :  SAGE,  May 2005 
<br/><br/> The project on INtroducing New Contraceptives in Rural China (INCR) was carried out between 1991 an d1996 in four counties of rural north  China. The experimental componet involved the random assignment of a multipronged treatment to four townships in each county. Two townships per county served as controls. The scale of the project made it nearly impossible to maintain the integrity of the experimental model that was at the core of the project's design. In spite of the massive numbers of people in the catchment areas of the study (>100,000), the experimental design was of low power, with but eight controls on sixteen treatments; when random assignment is at higher levels of analysis, masses of individual observations only create greater precision in the estimation of higher-order observations. However, the study would not have been made, were it not conceived as an experiment.