Changes between Version 4 and Version 5 of SOPs/anova


Ignore:
Timestamp:
05/24/13 14:24:35 (12 years ago)
Author:
gbell
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • SOPs/anova

    v4 v5  
    55See the [http://www.graphpad.com/support/faqid/1745/ Prism help page] for some general considerations.
    66
    7 ==== Reading in Data ====
     7==== Reading in data ====
    88
    99
     
    2828}}}
    2929 
    30 ==== Creating an ANOVA Table ====
     30==== Creating an ANOVA table ====
    3131  * Use the command //anova// or //aov// with summary.  The first argument is the dependent variable, followed by ~, and then by independent variable(s).
    3232  * So if we want to set up a model where weight is a function of the group (e.g., the weight potentially depends on the group)
     
    4040}}}
    4141
    42 == Post-Test: Comparing All Pairs of Means ==
     42== Post-test: Comparing all pairs of means ==
    4343
    4444==== Tukey ====
     
    6767Two-way ANOVA should be used for experiments where two different factors are being studied (such as comparing different treatments of different genotypes of mice).
    6868
    69 ==== Reading in Data ====
     69==== Reading in data, plotting, and summarizing ====
    7070
    7171
     
    8080# ...
    8181strains = read.delim("brain_weights.txt",header=TRUE)
     82
     83# Plot the data by group
     84boxplot(weight ~ paste(genotype, treatment), data=strains)
     85stripchart(weight ~ paste(genotype, treatment), data=strains, vert=T, method="jitter", jitter = 0.4, pch=19, cex=2, add=T)
     86
     87# Summarize the data by group
     88tapply(strains$weight, paste(strains$genotype, strains$treatment), mean)
    8289}}}
    8390
     
    98105Note that the p-value for each factor depends on the order of the factors in the above formulas.
    99106
    100 == Post-Test: Comparing All Pairs of Means ==
     107== Post-test: Comparing all pairs of means ==
    101108
    102109As before, with 1-way ANOVA,