Abdolell, Mohamed
2003-Nov-15 03:56 UTC
[R] computing a p-value for a one-way ANOVA given only means and std dev of 5 factor levels
I've been presented with a problem where I don't have the raw data, but I do have summary data. Specifically, I have a response that takes values from 0 thru 10, and a factor that has 5 levels (unordered) ... I have a mean and std dev for each of the 5 levels and need to obtain a p-value indicating whether there is evidence of a difference between these means. Mean Std Dev N Level A mA stdA nA Level B mB stdB nB Level C mC stdC nC Level D mD stdD nD Level E mE stdE nE My current plan is to assume normally distributed responses and use the known mean and std dev to generate data points which can be combined to create a data set similar to the "true" (but unavailable) data set. Then I can use a standard 1-way ANOVA to determine the p-value. Does anyone know of any pre-packaged solution to this type of problem in R? Or perhaps suggest a better solution than that which I have outlined above? Thanks ... Mohamed This e-mail message (and any attachments) may contain confidential and/or privileged information for the sole use of the intended recipient. Any review or distribution by anyone other than the person for whom it was originally intended is strictly prohibited. If you have received this e-mail in error, please contact the sender and delete all copies. Opinions, conclusions or other information contained in this e-mail may not be that of the organization. [[alternative HTML version deleted]]