t-Test one sided

[Manfred]

How do I compute a one sided t-Test with SPSS? I only get a two sided result
in the SPSS output.
I have the following data (computed by R):



t = 0.1771, df = 277, p-value = 0.5702
alternative hypothesis: true mean is less than 175.5
95 percent confidence interval:
     -Inf 176.5394
sample estimates:
mean of x
 175.6007

> t.test(data$GRO, mu=175.5)

        One Sample t-test

data:  data$GRO

t = 0.1771, df = 277, p-value = 0.8596
alternative hypothesis: true mean is not equal to 175.5
95 percent confidence interval:
 174.4811 176.7204
sample estimates:
mean of x
 175.6007




--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/t-Test-one-sided-tp5714605.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Zdaniuk, Bozena-3]

Bozena Zdaniuk will be out of reach till August 14, 2012. Thank you for your patience.

[Swank, Paul R]

To do a one-tailed test, divide the p value by two and make sure the result is in the predicted direction.

Paul R. Swank, Ph.D.
Professor, Department of Pediatrics
Medical School
Adjunct Professor, Health Promotions and Behavioral Sciences
School of Public Health
University of Texas Health Science Center at Houston


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of abcdefg
Sent: Tuesday, August 07, 2012 3:39 AM
To: [hidden email]
Subject: t-Test one sided

How do I compute a one sided t-Test with SPSS? I only get a two sided result in the SPSS output.
I have the following data (computed by R):



t = 0.1771, df = 277, p-value = 0.5702
alternative hypothesis: true mean is less than 175.5
95 percent confidence interval:
     -Inf 176.5394
sample estimates:
mean of x
 175.6007

> t.test(data$GRO, mu=175.5)

        One Sample t-test

data:  data$GRO

t = 0.1771, df = 277, p-value = 0.8596
alternative hypothesis: true mean is not equal to 175.5
95 percent confidence interval:
 174.4811 176.7204
sample estimates:
mean of x
 175.6007




--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/t-Test-one-sided-tp5714605.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Rich Ulrich]

To provide a fuller version of what Paul posted --

The one-tailed p-value from R was 0.57.  Or, if
you considered both of the possible one tailed tests,
the two values would be 0.43 and 0.57, since 0.43
is what is needed to sum to 1.0.

The p-value from SPSS is 0.86, which is twice
the (smaller) value of 0.43. 

To recover the same 0.57 as you see in R, you can
divide the 0.86 by 2, to get 0.43; and then you note
that the direction of the observed difference is actually
in the wrong direction, so that instead of using p/2,
you want to report 1-p/2 .

--
Rich Ulrich

> Date: Tue, 7 Aug 2012 01:39:20 -0700

> From: [hidden email]
> Subject: t-Test one sided
> To: [hidden email]
>
> How do I compute a one sided t-Test with SPSS? I only get a two sided result
> in the SPSS output.
> I have the following data (computed by R):
>
>
>
> t = 0.1771, df = 277, p-value = 0.5702
> alternative hypothesis: true mean is less than 175.5
> 95 percent confidence interval:
> -Inf 176.5394
> sample estimates:
> mean of x
> 175.6007
>
> > t.test(data$GRO, mu=175.5)
>
> One Sample t-test
>
> data: data$GRO
>
> t = 0.1771, df = 277, p-value = 0.8596
> alternative hypothesis: true mean is not equal to 175.5
> 95 percent confidence interval:
> 174.4811 176.7204
> sample estimates:
> mean of x
> 175.6007
>
> ...