Programming the statistical procedures from SAS

When I use proc univariate for hypothesis testing the t value and p values do not make sense

Accepted Solution Solved
Reply
New Contributor
Posts: 3
Accepted Solution

When I use proc univariate for hypothesis testing the t value and p values do not make sense

The data comes from the data table: marchflights.sas7bdat   (airplane flights and the corresponding weights of "freight" - the variable)

 

Freight (variable)

Mean=333.95, Standard deviation=98.11.  Number of observations=635

 

Below is the code:

 

/**Hypothesis Testing**/
proc univariate data=certprep.marchflights mu0 = 333.95;
     var freight;
run;

 

I believe that the t statistic should be much higher than 0.000283,

and the p value should be very low, looking like 0.0040 or something like that, but it is 0.9998  

 

Results are below:

 

Tests for Location: Mu0=333.95

0.0002830.9998
t statistic  p value
  

 

Can the results of the t statistic and the p value actually be correct? I don't think so.

Is there anything I'm missing here? Is there something wrong with the code?

 

Thank you.

 

jtflynn33


Accepted Solutions
Highlighted
Solution
2 weeks ago
Super User
Posts: 21,478

Re: When I use proc univariate for hypothesis testing the t value and p values do not make sense

Posted in reply to jtflynn33

I think you're misunderstanding hypothesis testing.


The null hypothesis is that the mean is equal to 333.95. If p<Alpha then you would reject the null hypothesis.

Since the P-value is 1, this means you fail to reject the null hypothesis and the mean is not significantly different than 333.95. 

 

Usually you're looking to reject a null hypothesis, in this case you're looking to confirm your null hypothesis. 

 


jtflynn33 wrote:

The data comes from the data table: marchflights.sas7bdat   (airplane flights and the corresponding weights of "freight" - the variable)

 

Freight (variable)

Mean=333.95, Standard deviation=98.11.  Number of observations=635

 

Below is the code:

 

/**Hypothesis Testing**/
proc univariate data=certprep.marchflights mu0 = 333.95;
     var freight;
run;

 

I believe that the t statistic should be much higher than 0.000283,

and the p value should be very low, looking like 0.0040 or something like that, but it is 0.9998  

 

Results are below:

 

Tests for Location: Mu0=333.95

0.000283 0.9998
t statistic   p value
   

 

Can the results of the t statistic and the p value actually be correct? I don't think so.

Is there anything I'm missing here? Is there something wrong with the code?

 

Thank you.

 

jtflynn33


 

 

View solution in original post


All Replies
Super User
Posts: 10,313

Re: When I use proc univariate for hypothesis testing the t value and p values do not make sense

Posted in reply to jtflynn33

It looks right.  If you want p value be small (significant) , then t statistic must be big enough. 

t statistic measured the deviation of H0 statistical (when H0 is assuming right) .

New Contributor
Posts: 3

Re: When I use proc univariate for hypothesis testing the t value and p values do not make sense

Thanks, your answer really helped me.
Respected Advisor
Posts: 2,155

Re: When I use proc univariate for hypothesis testing the t value and p values do not make sense

[ Edited ]
Posted in reply to jtflynn33

t = (xbar - mu) / (s / sqrt(n))

 

xbar and mu are both 333.95 (well, xbar probably has a few more decimal places), so t close to zero is correct.

 

When you have a t close to zero, you have a p value close to 1

--
Paige Miller
New Contributor
Posts: 3

Re: When I use proc univariate for hypothesis testing the t value and p values do not make sense

Posted in reply to PaigeMiller
Thanks for your help!!
Highlighted
Solution
2 weeks ago
Super User
Posts: 21,478

Re: When I use proc univariate for hypothesis testing the t value and p values do not make sense

Posted in reply to jtflynn33

I think you're misunderstanding hypothesis testing.


The null hypothesis is that the mean is equal to 333.95. If p<Alpha then you would reject the null hypothesis.

Since the P-value is 1, this means you fail to reject the null hypothesis and the mean is not significantly different than 333.95. 

 

Usually you're looking to reject a null hypothesis, in this case you're looking to confirm your null hypothesis. 

 


jtflynn33 wrote:

The data comes from the data table: marchflights.sas7bdat   (airplane flights and the corresponding weights of "freight" - the variable)

 

Freight (variable)

Mean=333.95, Standard deviation=98.11.  Number of observations=635

 

Below is the code:

 

/**Hypothesis Testing**/
proc univariate data=certprep.marchflights mu0 = 333.95;
     var freight;
run;

 

I believe that the t statistic should be much higher than 0.000283,

and the p value should be very low, looking like 0.0040 or something like that, but it is 0.9998  

 

Results are below:

 

Tests for Location: Mu0=333.95

0.000283 0.9998
t statistic   p value
   

 

Can the results of the t statistic and the p value actually be correct? I don't think so.

Is there anything I'm missing here? Is there something wrong with the code?

 

Thank you.

 

jtflynn33


 

 

☑ This topic is solved.

Need further help from the community? Please ask a new question.

Discussion stats
  • 5 replies
  • 221 views
  • 7 likes
  • 4 in conversation