Hello, I am a new user of R and trying to assess the sample size for data that is being collected on water quality at sites across a wide geographic region. A preliminary set of data has been collected and I would like to use it to assess whether we are collecting enough data and in the right places. A factorial approach was initially used to characterize sites by well type, latrine type, distance between well and latrine, and ecological region. Altogether the basic structure has: 3 types of wells 3 types of latrines 4 distance categories 13 regions We define a ?site-type? as: a well-latrine-distance combination. There are 36 of these. A number of replicates (between 1 and 4) of the 36 site-types are included in the set of sites in each of the 13 regions. Some regions have more replicates than others due to complexity in the region. In total there are 936 sites. At this point, I have an ANOVA model with water quality measures and only (these) categorical data. I want to know if I am collecting enough samples (given alpha and beta levels) to see if there are effects for wells, latrines, distances, and region (independently), as well as interactions for well-distance, well-latrine, and well-region. I would like to also perform a power analysis to allow the power vary with sample size. I am working my way through various texts and help functions but thought I would see if anyone else has learned how to do this already. I would appreciate any and all guidance. Best wishes, Chris Christopher Behr Principal Analyst eDesign Dynamics www.edesigndynamics.com Christopher Behr Principal Analyst eDesign Dynamics www.edesigndynamics.com 4024 Calvert St. NW Washington DC 20007 (202) 298-6437 (t/f) (551) 998-4823 (c)
sample size determination
2 messages · Chris Behr, Martin Henry H. Stevens
?power.anova.test is designed for 1 way models. I would be intrigued to found out more about power in complicated designs. Do look into the archives at http://finzi.psych.upenn.edu/search.html In the past, I used power.t.test on individual model coefficients, though I am not sure it was correct. Hank Stevens
On Oct 18, 2005, at 3:50 PM, Chris Behr wrote:
Hello, I am a new user of R and trying to assess the sample size for data that is being collected on water quality at sites across a wide geographic region. A preliminary set of data has been collected and I would like to use it to assess whether we are collecting enough data and in the right places. A factorial approach was initially used to characterize sites by well type, latrine type, distance between well and latrine, and ecological region. Altogether the basic structure has: 3 types of wells 3 types of latrines 4 distance categories 13 regions We define a ?site-type? as: a well-latrine-distance combination. There are 36 of these. A number of replicates (between 1 and 4) of the 36 site-types are included in the set of sites in each of the 13 regions. Some regions have more replicates than others due to complexity in the region. In total there are 936 sites. At this point, I have an ANOVA model with water quality measures and only (these) categorical data. I want to know if I am collecting enough samples (given alpha and beta levels) to see if there are effects for wells, latrines, distances, and region (independently), as well as interactions for well-distance, well-latrine, and well-region. I would like to also perform a power analysis to allow the power vary with sample size. I am working my way through various texts and help functions but thought I would see if anyone else has learned how to do this already. I would appreciate any and all guidance. Best wishes, Chris Christopher Behr Principal Analyst eDesign Dynamics www.edesigndynamics.com Christopher Behr Principal Analyst eDesign Dynamics www.edesigndynamics.com 4024 Calvert St. NW Washington DC 20007 (202) 298-6437 (t/f) (551) 998-4823 (c)
______________________________________________ R-help at stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting- guide.html
Dr. Martin Henry H. Stevens, Assistant Professor 338 Pearson Hall Botany Department Miami University Oxford, OH 45056 Office: (513) 529-4206 Lab: (513) 529-4262 FAX: (513) 529-4243 http://www.cas.muohio.edu/~stevenmh/ http://www.muohio.edu/ecology/ http://www.muohio.edu/botany/ "E Pluribus Unum"