Hello, I am struggling to figure out how to analyse a dataset I have inherited (please note this was conducted some time ago, so the data is as it is, and I know it isn't perfect!). A brief description of the experiment follows: Pots of grass were grown in 1l pots of standad potting medium for 1 month with a regular light and watering regime. At this point they were randomly given 1l of one of 4 different pesticides at one of 4 different concentrations (100%, 75%, 50% or 25% in water). There were 20 pots of grass for each pesticide/concentration giving 320 pots. There were no control (untreated) pots. The response was measured after 1 week and recorded as either: B1 - grass dead B2 - grass affected but not dead B3 - no visible effect I could analyse this as lethal effect vs non-lethal effect (B1 vs B2+B3) or some effect vs no effect (B1+B2 vs B3) binomial model, but I can't see how to do it with three levels. Any pointing in the right direction greatly appreciated! Thanks Matt -------------------------------------------------------------------------- Disclaimer: This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify me at Matt.Ellis@basc.org.uk then delete it. BASC may monitor email traffic. By replying to this e-mail you consent to BASC monitoring the content of any email you send or receive from BASC. Any views expressed in this message are those of the individual sender, except where the sender specifies with authority, states them to be the views of the British Association for Shooting and Conservation. BASC can confirm that this email message and any attachments have been scanned for the presence of computer viruses but recommends that you make your own virus checks. Registered Industrial and Provident Society No.: 28488R. Registered Office: Marford Mill, Rossett, Wrexham, LL12 0HL. -------------------------------------------------------------------------- [[alternative HTML version deleted]]
You could analyze these data with a multinomial logit model, with an ordinal response. I don't have my copy handy, but I know that Agresti (Categorical Data Analysis, Wiley) covers these models. Cheers David Cross d.cross at tcu.edu www.davidcross.us On Jun 22, 2011, at 8:19 AM, Matt Ellis (Research) wrote:> Hello, > I am struggling to figure out how to analyse a dataset I have inherited > (please note this was conducted some time ago, so the data is as it is, > and I know it isn't perfect!). > > A brief description of the experiment follows: > Pots of grass were grown in 1l pots of standad potting medium for 1 > month with a regular light and watering regime. At this point they were > randomly given 1l of one of 4 different pesticides at one of 4 different > concentrations (100%, 75%, 50% or 25% in water). There were 20 pots of > grass for each pesticide/concentration giving 320 pots. There were no > control (untreated) pots. The response was measured after 1 week and > recorded as either: > B1 - grass dead > B2 - grass affected but not dead > B3 - no visible effect > > I could analyse this as lethal effect vs non-lethal effect (B1 vs B2+B3) > or some effect vs no effect (B1+B2 vs B3) binomial model, but I can't > see how to do it with three levels. > > Any pointing in the right direction greatly appreciated! > Thanks > Matt > > -------------------------------------------------------------------------- > Disclaimer: This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify me at Matt.Ellis at basc.org.uk then delete it. BASC may monitor email traffic. By replying to this e-mail you consent to BASC monitoring the content of any email you send or receive from BASC. Any views expressed in this message are those of the individual sender, except where the sender specifies with authority, states them to be the views of the British Association for Shooting and Conservation. BASC can confirm that this email message and any attachments have been scanned for the presence of computer viruses but recommends that you make your own virus checks. Registered Industrial and Provident Society No.: 28488R. Registered Office: Marford Mill, Rossett, Wrexham, LL12 0HL. > -------------------------------------------------------------------------- > > > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code.
The best method would probably be proportional odds regression using polyr() in 'MASS'. At least it's a starting point. At 09:19 AM 6/22/2011, Matt Ellis \(Research\) wrote:>Hello, >I am struggling to figure out how to analyse a dataset I have inherited >(please note this was conducted some time ago, so the data is as it is, >and I know it isn't perfect!). > >A brief description of the experiment follows: >Pots of grass were grown in 1l pots of standad potting medium for 1 >month with a regular light and watering regime. At this point they were >randomly given 1l of one of 4 different pesticides at one of 4 different >concentrations (100%, 75%, 50% or 25% in water). There were 20 pots of >grass for each pesticide/concentration giving 320 pots. There were no >control (untreated) pots. The response was measured after 1 week and >recorded as either: >B1 - grass dead >B2 - grass affected but not dead >B3 - no visible effect > >I could analyse this as lethal effect vs non-lethal effect (B1 vs B2+B3) >or some effect vs no effect (B1+B2 vs B3) binomial model, but I can't >see how to do it with three levels. > >Any pointing in the right direction greatly appreciated! >Thanks >Matt > >-------------------------------------------------------------------------- >Disclaimer: This email and any files transmitted with it are >confidential and intended solely for the use of the individual or >entity to whom they are addressed. If you have received this email >in error please notify me at Matt.Ellis at basc.org.uk then delete it. >BASC may monitor email traffic. By replying to this e-mail you >consent to BASC monitoring the content of any email you send or >receive from BASC. Any views expressed in this message are those of >the individual sender, except where the sender specifies with >authority, states them to be the views of the British Association >for Shooting and Conservation. BASC can confirm that this email >message and any attachments have been scanned for the presence of >computer viruses but recommends that you make your own virus checks. >Registered Industrial and Provident Society No.: 28488R. Registered >Office: Marford Mill, Rossett, Wrexham, LL12 0HL. >-------------------------------------------------------------------------- > > > > [[alternative HTML version deleted]] > >______________________________________________ >R-help at r-project.org mailing list >https://stat.ethz.ch/mailman/listinfo/r-help >PLEASE do read the posting guide http://www.R-project.org/posting-guide.html >and provide commented, minimal, self-contained, reproducible code.===============================================================Robert A. LaBudde, PhD, PAS, Dpl. ACAFS e-mail: ral at lcfltd.com Least Cost Formulations, Ltd. URL: http://lcfltd.com/ 824 Timberlake Drive Tel: 757-467-0954 Virginia Beach, VA 23464-3239 Fax: 757-467-2947 "Vere scire est per causas scire"