Re: [R] Split-split plot ANOVA
] Sent: Thursday, February 03, 2005 5:40 AM Subject: Re: [R] Split-split plot ANOVA Hi Mike, Do you have a schematic drawing of how exactly your treatments were applied? In split-plot experiments, it is generally very important to clearly define the sequence of plot sizes, because if you don´t do this properly, then the output will be confusing. Checking if your degrees of freedom at each level are correct should give you a good idea about whether you´ve specified the model in the right way. Generally, I see some problem with your model specification as you seem to have two (not one) treatments in some of your subplots. If I got it right, the sequence of terms should be something like Block/Whole.plot/Caging/Competition/Species at least if it´s a full split-plot. Can you send me some more details on the design? Regards, Christoph [EMAIL PROTECTED] wrote: I have been going over and over the examples in MASS and the Pinheiro and Bates example, but cannot get my model to run correctly with either aov or lme. Could someone give me a hand with the correct model statement? It would help to see some of the things you have tried already ... First a description of the design. We are studying germination rates for various species under a variety of treaments. This is a blocked split-split plot design. The levels and treatments are: Blocks: 1-6 Whole plot treatment: Overstory: Yes or No Split plot treatments: Caging (to protect against seed predators): Yes or No Herbaceous competition (i.e., grass): Yes or No Split-split plot treatment: Tree species: 7 kinds The response variable is Lag, which is a indication of when the seeds first germinated. I would try somthing like lme (fixed= Lag ~ Caging + herbaceous + tree, data= your.data, random= ~ 1 | Overstory/split/splitsplit) Perhaps you want/need to add some interactions as well. Overstory, split and splitsplit would be factors with specific levels for each of the plots, split plots and split-split plots, respectively. Thus what I attempted here is to separate the variables of the hierarchical design of data gathering (which go into the random effects) and the treatments (which go into the fixed effects). The degrees of freedom for the fixed effects are automatically adjusted to the correct level in the hierarchy. Did you try that? What did not work out with it? Lastly, I have unbalanced data since some treatment combinations never had any germination. In principle, the REML estimates in lme are not effected by unbalanced data. BUT I do not think that the missing germinations by themselves lead to an unbalanced data set: I assume it is informative that in some treatment combinations there was no germination. Thus, your lag there is something close to infinity (or at least longer than you cared to wait ;-). Thus, I would argue you have to somehow include these data points as well, otherwise you can only make a very restricted statement of the kind: if there was germination, this depended on such and such. Since the data are highly nonnormal, I hope to do a permutations test on the F-values for each main effect and interaction in order to get my p-values. As these are durations a log transformation of your response might be enough. Regards, Lorenz - Lorenz Gygax, Dr. sc. nat. Centre for proper housing of ruminants and pigs Swiss Federal Veterinary Office agroscope FAT Tänikon, CH-8356 Ettenhausen / Switzerland __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R
Re: [R] Split-split plot ANOVA
Hi Mike, Do you have a schematic drawing of how exactly your treatments were applied? In split-plot experiments, it is generally very important to clearly define the sequence of plot sizes, because if you don´t do this properly, then the output will be confusing. Checking if your degrees of freedom at each level are correct should give you a good idea about whether you´ve specified the model in the right way. Generally, I see some problem with your model specification as you seem to have two (not one) treatments in some of your subplots. If I got it right, the sequence of terms should be something like Block/Whole.plot/Caging/Competition/Species at least if it´s a full split-plot. Can you send me some more details on the design? Regards, Christoph [EMAIL PROTECTED] wrote: I have been going over and over the examples in MASS and the Pinheiro and Bates example, but cannot get my model to run correctly with either aov or lme. Could someone give me a hand with the correct model statement? It would help to see some of the things you have tried already ... First a description of the design. We are studying germination rates for various species under a variety of treaments. This is a blocked split-split plot design. The levels and treatments are: Blocks: 1-6 Whole plot treatment: Overstory: Yes or No Split plot treatments: Caging (to protect against seed predators): Yes or No Herbaceous competition (i.e., grass): Yes or No Split-split plot treatment: Tree species: 7 kinds The response variable is Lag, which is a indication of when the seeds first germinated. I would try somthing like lme (fixed= Lag ~ Caging + herbaceous + tree, data= your.data, random= ~ 1 | Overstory/split/splitsplit) Perhaps you want/need to add some interactions as well. Overstory, split and splitsplit would be factors with specific levels for each of the plots, split plots and split-split plots, respectively. Thus what I attempted here is to separate the variables of the hierarchical design of data gathering (which go into the random effects) and the treatments (which go into the fixed effects). The degrees of freedom for the fixed effects are automatically adjusted to the correct level in the hierarchy. Did you try that? What did not work out with it? Lastly, I have unbalanced data since some treatment combinations never had any germination. In principle, the REML estimates in lme are not effected by unbalanced data. BUT I do not think that the missing germinations by themselves lead to an unbalanced data set: I assume it is informative that in some treatment combinations there was no germination. Thus, your lag there is something close to infinity (or at least longer than you cared to wait ;-). Thus, I would argue you have to somehow include these data points as well, otherwise you can only make a very restricted statement of the kind: if there was germination, this depended on such and such. Since the data are highly nonnormal, I hope to do a permutations test on the F-values for each main effect and interaction in order to get my p-values. As these are durations a log transformation of your response might be enough. Regards, Lorenz - Lorenz Gygax, Dr. sc. nat. Centre for proper housing of ruminants and pigs Swiss Federal Veterinary Office agroscope FAT Tänikon, CH-8356 Ettenhausen / Switzerland __ R-help@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 __ R-help@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
Re: [R] Split-split plot ANOVA
Jesus and the rest of the R-help community: Thanks for your help. I have been going over and over the examples in MASS and the Pinheiro and Bates example, but cannot get my model to run correctly with either aov or lme. Could someone give me a hand with the correct model statement? First a description of the design. We are studying germination rates for various species under a variety of treaments. This is a blocked split-split plot design. The levels and treatments are: Blocks: 1-6 Whole plot treatment: Overstory: Yes or No Split plot treatments: Caging (to protect against seed predators): Yes or No Herbaceous competition (i.e., grass): Yes or No Split-split plot treatment: Tree species: 7 kinds The response variable is Lag, which is a indication of when the seeds first germinated. I will be doing this analysis for a couple other response variables as well in separate analyses. I have had mixed results using the examples as a guide to build my statement; I am unsure how to specify the crossed factors at the split-plot level. Lastly, I have unbalanced data since some treatment combinations never had any germination. Since the data are highly nonnormal, I hope to do a permutations test on the F-values for each main effect and interaction in order to get my p-values. Thanks for your help in advance, Mike Mike Saunders Research Assistant Forest Ecosystem Research Program Department of Forest Ecosystem Sciences University of Maine Orono, ME 04469 207-581-2763 (O) 207-581-4257 (F) - Original Message - From: Jesus Frias [EMAIL PROTECTED] To: Mike Saunders [EMAIL PROTECTED]; R Help r-help@stat.math.ethz.ch Sent: Tuesday, February 01, 2005 10:57 AM Subject: RE: [R] Split-split plot ANOVA Hi Mike, *An example of the use of aov() for a split-plot is in MASS library(MASS) example(Oats) The book also gives a detailed explanation *pp 45-52 of the Pinheiro and Bates book gives you an example of the use of lme() on a split-plot. If you have a non balanced design, lme() from the nlme library might be a better tool than aov(). Also, if you have the lme4 library installed you'll have a lot more flexibility on the formulation of your random effects. regards, Jesus -- Jesús María Frías Celayeta School of Food Sci. and Env. Health. Faculty of Tourism and Food Dublin Institute of Technology Cathal Brugha St., Dublin 1. Ireland t +353 1 4024459 f +353 1 4024495 w www.dit.ie/DIT/tourismfood/science/staff/frias.html -- -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Mike Saunders Sent: 01 February 2005 14:36 To: R Help Subject: [R] Split-split plot ANOVA Does someone out there have an example of R-code for a split-split plot ANOVA using aov or another function? The design is not balanced. I never set up one in R before and it would be nice to see an example before I tackle a very complex design I have to model. Thanks, Mike Mike Saunders Research Assistant Forest Ecosystem Research Program Department of Forest Ecosystem Sciences University of Maine Orono, ME 04469 207-581-2763 (O) 207-581-4257 (F) [[alternative HTML version deleted]] __ R-help@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 -- This message has been scanned for content and viruses by the DIT Information Services MailScanner Service, and is believed to be clean. http://www.dit.ie -- This message has been scanned for content and viruses by the DIT Information Services MailScanner Service, and is believed to be clean. http://www.dit.ie __ R-help@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
RE: [R] Split-split plot ANOVA
I have been going over and over the examples in MASS and the Pinheiro and Bates example, but cannot get my model to run correctly with either aov or lme. Could someone give me a hand with the correct model statement? It would help to see some of the things you have tried already ... First a description of the design. We are studying germination rates for various species under a variety of treaments. This is a blocked split-split plot design. The levels and treatments are: Blocks: 1-6 Whole plot treatment: Overstory: Yes or No Split plot treatments: Caging (to protect against seed predators): Yes or No Herbaceous competition (i.e., grass): Yes or No Split-split plot treatment: Tree species: 7 kinds The response variable is Lag, which is a indication of when the seeds first germinated. I would try somthing like lme (fixed= Lag ~ Caging + herbaceous + tree, data= your.data, random= ~ 1 | Overstory/split/splitsplit) Perhaps you want/need to add some interactions as well. Overstory, split and splitsplit would be factors with specific levels for each of the plots, split plots and split-split plots, respectively. Thus what I attempted here is to separate the variables of the hierarchical design of data gathering (which go into the random effects) and the treatments (which go into the fixed effects). The degrees of freedom for the fixed effects are automatically adjusted to the correct level in the hierarchy. Did you try that? What did not work out with it? Lastly, I have unbalanced data since some treatment combinations never had any germination. In principle, the REML estimates in lme are not effected by unbalanced data. BUT I do not think that the missing germinations by themselves lead to an unbalanced data set: I assume it is informative that in some treatment combinations there was no germination. Thus, your lag there is something close to infinity (or at least longer than you cared to wait ;-). Thus, I would argue you have to somehow include these data points as well, otherwise you can only make a very restricted statement of the kind: if there was germination, this depended on such and such. Since the data are highly nonnormal, I hope to do a permutations test on the F-values for each main effect and interaction in order to get my p-values. As these are durations a log transformation of your response might be enough. Regards, Lorenz - Lorenz Gygax, Dr. sc. nat. Centre for proper housing of ruminants and pigs Swiss Federal Veterinary Office agroscope FAT Tänikon, CH-8356 Ettenhausen / Switzerland __ R-help@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
[R] Split-split plot ANOVA
Does someone out there have an example of R-code for a split-split plot ANOVA using aov or another function? The design is not balanced. I never set up one in R before and it would be nice to see an example before I tackle a very complex design I have to model. Thanks, Mike Mike Saunders Research Assistant Forest Ecosystem Research Program Department of Forest Ecosystem Sciences University of Maine Orono, ME 04469 207-581-2763 (O) 207-581-4257 (F) [[alternative HTML version deleted]] __ R-help@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
RE: [R] Split-split plot ANOVA
Hi Mike, *An example of the use of aov() for a split-plot is in MASS library(MASS) example(Oats) The book also gives a detailed explanation *pp 45-52 of the Pinheiro and Bates book gives you an example of the use of lme() on a split-plot. If you have a non balanced design, lme() from the nlme library might be a better tool than aov(). Also, if you have the lme4 library installed you'll have a lot more flexibility on the formulation of your random effects. regards, Jesus -- Jesús María Frías Celayeta School of Food Sci. and Env. Health. Faculty of Tourism and Food Dublin Institute of Technology Cathal Brugha St., Dublin 1. Ireland t +353 1 4024459 f +353 1 4024495 w www.dit.ie/DIT/tourismfood/science/staff/frias.html -- -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Mike Saunders Sent: 01 February 2005 14:36 To: R Help Subject: [R] Split-split plot ANOVA Does someone out there have an example of R-code for a split-split plot ANOVA using aov or another function? The design is not balanced. I never set up one in R before and it would be nice to see an example before I tackle a very complex design I have to model. Thanks, Mike Mike Saunders Research Assistant Forest Ecosystem Research Program Department of Forest Ecosystem Sciences University of Maine Orono, ME 04469 207-581-2763 (O) 207-581-4257 (F) [[alternative HTML version deleted]] __ R-help@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 -- This message has been scanned for content and viruses by the DIT Information Services MailScanner Service, and is believed to be clean. http://www.dit.ie -- This message has been scanned for content and viruses by the DIT Information Services MailScanner Service, and is believed to be clean. http://www.dit.ie __ R-help@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