I need to determine the critical t values for one-sided tails of 75% and 99%, for 40 degrees of freedom.
The following is code for a two-sided 99% critical t values:
qt(0.01, 40)
but how can I determine for a one-sided critical t value?
This question is related to
r
The code you posted gives the critical value for a one-sided test (Hence the answer to you question is simply:
abs(qt(0.25, 40)) # 75% confidence, 1 sided (same as qt(0.75, 40))
abs(qt(0.01, 40)) # 99% confidence, 1 sided (same as qt(0.99, 40))
Note that the t-distribution is symmetric. For a 2-sided test (say with 99% confidence) you can use the critical value
abs(qt(0.01/2, 40)) # 99% confidence, 2 sided
Josh's comments are spot on. If you are not super familiar with critical values I'd suggest playing with qt, reading the manual (?qt
) in conjunction with looking at a look up table (LINK). When I first moved from SPSS to R I created a function that made critical t value look up pretty easy (I'd never use this now as it takes too much time and with the p values that are generally provided in the output it's a moot point). Here's the code for that:
critical.t <- function(){
cat("\n","\bEnter Alpha Level","\n")
alpha<-scan(n=1,what = double(0),quiet=T)
cat("\n","\b1 Tailed or 2 Tailed:\nEnter either 1 or 2","\n")
tt <- scan(n=1,what = double(0),quiet=T)
cat("\n","\bEnter Number of Observations","\n")
n <- scan(n=1,what = double(0),quiet=T)
cat("\n\nCritical Value =",qt(1-(alpha/tt), n-2), "\n")
}
critical.t()
Extending @Ryogi answer above, you can take advantage of the lower.tail
parameter like so:
qt(0.25/2, 40, lower.tail = FALSE)
# 75% confidence
qt(0.01/2, 40, lower.tail = FALSE)
# 99% confidence
Source: Stackoverflow.com