Hypothesis Testing how is the test statistic useful to determine if we reject/or not the null hypothesis?
What do you need help with on hypothesis testing? There are several types of tests and it depends on what you are given. If you want I can give you a handy guide I made on hypothesis testing when I took AP Stat.
did you have to use excel or the table to find p-value?
So you know that you obtain test statistic based on the type of hypothesis test you are doing, such as z-test, t-test, etc, but generally it is \( \frac{statistic - parameter}{s}\) Once you obtain your test statistic, if your p-value of the test statistic is within the acceptance region, or p-value is greater than alpha, then it is not statistically significant and we fail to reject the null. We *cannot* say we approve the null. If it is less than alpha, we reject the null. Most of the time, the general alpha value is 0.05 but it depends on your confidence level required.
When I took stat we didn't use excel to find the p-value, I had a calculator that would do it for me (the TI 84 plus)
so for 1 class, I can't use the table and for other class, I have to use table to get the value. Table are based on "degree of freedom" so I was wondering why do we even need to calculate test statistics if we can just find the value from the table by using Degree of freedom.
The table used to confuse me a ton for t-tests, but basically you need to use in reverse when you find the test statistic. You would go to the appropriate df, then find the range where your test statistic exists. Then you go up and you see the corresponding areas (right tailed because this is t test!). That is where your p-value would most likely exist. For instance, if your df=6 and you obtained a test statistic of 1.5, your area is most likely between 0.1 and 0.05. Then you can go on with your conclusion
Also examiners typically ask for the test statistic as well even if the p value is correct
Does your class let you use a graphing calculator? For me, the exam allowed graphing calculator so using the table was not necessary either.
after looking at my notes I think we calculate test statistic to compare result. However, if we just have to make a decision we just need df, alpha value, and the table.
we use computers...
I mean software such as excel and many more..
You would do it just the same as a z-test, just that t-test has different calculations to find the p value. So you still need to obtain p value and do your conclusion from there
Ah okay. I don't know inferential stat in excel bc we did it on Ti 84s
do you know if t(critical) is the same as p-value ??
ofc they aren't. 🤦
thanks!
Your welcome :)
You're*
Join our real-time social learning platform and learn together with your friends!