Upgrade to access all content for this subject
In what way is the chi-square distribution different from the $t$-distribution?
There are many chi-square distributions based on the number of degrees of freedom, but only one $t$-distribution.
The chi-square distribution is used to write both confidence intervals and hypothesis tests, but the $t$-distribution is only used to write hypothesis tests.
Chi-square values are always positive while $t$-values can be positive or negative.
The chi-square distribution becomes more nearly normal as the degrees of freedom gets smaller while the $t$-distribution becomes more nearly normal as the degrees of freedom gets larger.
For low degrees of freedom, the chi-square distribution is symmetric while the $t$-distribution is skewed right.