Make sure you have dplyr installed and loaded into your library for this step.
# create a new table called "abr"
# that groups results based on condition and result, our two IV
abr = dta%>%group_by(Condition, Result)%>%summarize(
# create a column called M that gives me the mean of my Abr variable
#(from the dataset) and that ignores missing data (na.rm = TRUE)
M = mean(Abr, na.rm = TRUE),
# do the same for standard deviation
SD = sd(Abr, na.rm = TRUE),
# tell me the sample size of each cell
n=n()
)
# calculate the error on the estimate of the mean and add it as a column.
# we need this term for plotitng
abr$error = 1.96*(abr$SD / sqrt(abr$n))
# print this table in the output
print(abr)
Words in pink are variables you can re-name in this code to whatever you want
Words in green are files and variables specific to your existing R environment that should already exist (i.e., they might vary from my code to yours, but you are not creating them in this code)
Words in black are non-negotiable parts of the code
| Term | What it’s doing |
|---|---|
| abr = | Create a new table called “abr” |
| I highly encourage you to choose a new name for each table you make. It prevents us from saving-over important data and allows us to notice errors in the code much faster. | |
| dta%>% | working within my data file called “dta” (replace this with the name of your data file) |
| group_by(Condition, Result)%>% | Report all results grouped by the different levels of my IVS |
| Condition and Result are the names of the IV variables in my dataset | |
| summarize( | summarize the data you have using the following summary statistics |
| M = | Create a new column in the table called “M” (where we will put the means) |
| mean(Abr, na.rm = TRUE), | Create the mean of this variable in my dataset called Abr (change to your variable name) |
| na.rm = TRUE: ignore missing cases (otherwise it will return a mean of NA) | |
| SD = sd(Abr, na.rm = TRUE), | Same as we did for the means but for standard deviation |
| n=n()) | Create a new column in the table with the n of each cell in the 2x2 (or whatever you have) |
| abr$error | Create a new column called “error” in the table we previously called “abr” |
| 1.96*(abr$SD / sqrt(abr$n)) | This step is necessary if you want to create a graph with error bars, as we call this variable in the plotting code below. |
| For context, the equation for standard error is 1.96*(SD/sqrt(n)) | |
| abr$SD = Calling the “SD” column of the abr table | |
| sqrt(abr$n) = Getting the square root of the n column in the abr table | |
| print(abr) | Print the new table so we can see our results! |
This can be a really nice block to set echo=FALSE
p1 = ggplot(abr, aes(x=Condition, y=M, fill=Result)) +
geom_bar(position=position_dodge(), stat="identity", color="black")+
scale_fill_manual(values=c("white","gray87", "gray43"))+
geom_errorbar(aes(ymin=M-error, ymax=M+error), width=.2,
position=position_dodge(.9))+
ylab("Brand Attitude") + xlab("Condition")+ #re-label the axes
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(), #remove background
panel.background = element_blank(), axis.line = element_line(colour = "black"), legend.text = element_text(size = 14), axis.text = element_text(size = 14), axis.title = element_text(size = 20), legend.title = element_text(size = 20))
print(p1)
t.test(Abr ~ Result, data = dta)
m1 = aov(Abr ~ Condition, data = dta) # run the ANOVA
summary(m1) # print the ANOVA results
emmeans(m1, pairwise~Condition)$contrasts #run post hoc tests if necessary
m1 = aov(Abr ~ Condition*Result, data = dta) # run the ANOVA
summary(m1) # print the ANOVA results
emmeans(m1, pairwise~Result|Condition)$contrasts # run the simple effects one way
emmeans(m1, pairwise~Condition|Result)$contrasts# run the simple effects the other way