one-sample t test question

[Salbod]

Dear Friends,



I am trying to compare CBCL scores to the normative sample on which the CBCL
was standardized.  For example, if one of my scales has a mean of 54.38 and
an SD of  5.15, based on an N = 30, can I compare this to the normative
sample with the mean value of 50 with this formula:

T = (50-54.3750)/(5.15/sqrt(30))



What hypothesis is tested with SPSS's one-sample t test?



Any explanation regarding the above tests would be greatly appreciated.





TIA.



Stephen Salbod, Pace University, NYC









PS My earlier question on pretest comparisons and posttest comparisons was
made moot-the person switched to ANCOVA.

=====================
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

[Swank, Paul R]

Technically, it should be 54.375 - 50 since 50 is the supposed
population value. However, 50 is not really the population value, it is
the mean of the norm group (standardized). To be completely correct, you
should use a two sample model using the standard deviation (10) from the
norm group and the norm group sample size in the two group formula.
However, since the norm group is most likely fairly large, the
difference probably won't be very noticeable. The null hypothesis for
the one sample model is the mean of the population from which the sample
was drawn is 50. You have evidence that this hypothesis may not be true.

Paul R. Swank, Ph.D.
Professor and Director of Research
Children's Learning Institute
University of Texas Health Science Center - Houston


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Stephen Salbod
Sent: Sunday, March 23, 2008 11:53 AM
To: [hidden email]
Subject: one-sample t test question

Dear Friends,



I am trying to compare CBCL scores to the normative sample on which the
CBCL
was standardized.  For example, if one of my scales has a mean of 54.38
and
an SD of  5.15, based on an N = 30, can I compare this to the normative
sample with the mean value of 50 with this formula:

T = (50-54.3750)/(5.15/sqrt(30))



What hypothesis is tested with SPSS's one-sample t test?



Any explanation regarding the above tests would be greatly appreciated.





TIA.



Stephen Salbod, Pace University, NYC









PS My earlier question on pretest comparisons and posttest comparisons
was
made moot-the person switched to ANCOVA.

=====================
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

[valferes]

Hi Stephen,
 

If you want to perform the solution proposed by Paul Swank, you can run the following syntax:

 
http://pages.infinit.net/rlevesqu/Syntax/T-Test/TTestEffectSizeNonoverlapAndPower.txt <http://pages.infinit.net/rlevesqu/Syntax/T-Test/TTestEffectSizeNonoverlapAndPower.txt>
(Method 2)
 
Valentim R. Alferes
http://www.fpce.uc.pt/niips/spss_prc/index.htm
 
<http://www.fpce.uc.pt/nucleos/niips/>
 
 

________________________________

From: SPSSX(r) Discussion on behalf of Stephen Salbod
Sent: Sun 2008-03-23 16:53
To: [hidden email]
Subject: one-sample t test question



Dear Friends,



I am trying to compare CBCL scores to the normative sample on which the CBCL
was standardized.  For example, if one of my scales has a mean of 54.38 and
an SD of  5.15, based on an N = 30, can I compare this to the normative
sample with the mean value of 50 with this formula:

T = (50-54.3750)/(5.15/sqrt(30))



What hypothesis is tested with SPSS's one-sample t test?



Any explanation regarding the above tests would be greatly appreciated.





TIA.



Stephen Salbod, Pace University, NYC









PS My earlier question on pretest comparisons and posttest comparisons was
made moot-the person switched to ANCOVA.

=====================
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