I ran a study asking participants to choose between options A, B, and C based on various parameters.
For example: Which of the following did you find to be most inspirational?
My data looks something like this:
ID | Inspiration |
---|---|
1 | A |
2 | C |
3 | B |
4 | C |
5 | B |
6 | C |
7 | B |
8 | C |
9 | A |
10 | B |
11 | A |
12 | B |
I have calculated the relative frequencies so my data now looks like this:
ID | Inspiration | Proportion |
---|---|---|
1 | A | .25 |
2 | C | .33 |
3 | B | .42 |
4 | C | .33 |
5 | B | .42 |
6 | C | .33 |
7 | B | .42 |
8 | C | .33 |
9 | A | .25 |
10 | B | .42 |
11 | A | .25 |
12 | B | .42 |
My question is: how do I test to see if the relative frequency of each choice is significantly different from one another? That is, how do I know if the frequency of which people chose Option A is significantly different from the frequency of which they chose either B or C?
I have tried running t-tests, anovas, chi-squared tests, and two-proportion z-tests, but none of these options seem to be exactly what I'm looking for.
First use dput()
to provide your data:
dta <- structure(list(ID = 1:12, Inspiration = c("A", "C", "B", "C",
"B", "C", "B", "C", "A", "B", "A", "B")), class = "data.frame", row.names = c(NA, -12L))
Now construct a table of the relative frequencies:
tbl <- table(dta$Inspiration)
tbl
# A B C
# 3 5 4
Now the null hypothesis that each choice is selected equally (ie frequency of A == B == C):
chisq.test(tbl)
#
# Chi-squared test for given probabilities
#
# data: tbl
# X-squared = 0.5, df = 2, p-value = 0.7788
#
# Warning message:
# In chisq.test(tbl) : Chi-squared approximation may be incorrect
No significant difference between the categories.
The technical post webpages of this site follow the CC BY-SA 4.0 protocol. If you need to reprint, please indicate the site URL or the original address.Any question please contact:yoyou2525@163.com.