Reliability problem

[Humphrey Paulie]

        Hi,
  I computed cronbach alpha reliability for 14 items 52 cases.
it is -1.71!!!
The SPSS says: "The value is negative due to a negative average covariance among items. This violates reliability model assumptions. You may want to check item codings".
what does it mean and what should i do?
thanks
  Humphrey


---------------------------------
Do you Yahoo!?
 Everyone is raving about the  all-new Yahoo! Mail.

[Mark A Davenport MADAVENP]

I seem to recall am SPSS white paper that came out several years ago that
explained why SPSS sometimes presents negative alphas even if all of the
items are kosher (relatively speaking).  I'll see if I can scrape up the
link. In the mean time, check your item statistics, in particular your
'Alpha if item deleted' stats.  This should give you an immediate
indicator of an offender.  However, it sounds like you may have other
problems.  14 items for 52 cases sounds like you may have stability
problems no matter what you do (short of removing items or gathering more
cases).

Peace,

Mark.

***************************************************************************************************************************************************************
Mark A. Davenport Ph.D.
Senior Research Analyst
Office of Institutional Research
The University of North Carolina at Greensboro
336.256.0395
[hidden email]

'An approximate answer to the right question is worth a good deal more
than an exact answer to an approximate question.' --a paraphrase of J. W.
Tukey (1962)






Humphrey <[hidden email]>
Sent by: "SPSSX(r) Discussion" <[hidden email]>
10/25/2006 05:46 AM
Please respond to
Humphrey <[hidden email]>


To
[hidden email]
cc

Subject
Reliability problem






        Hi,
  I computed cronbach alpha reliability for 14 items 52 cases.
it is -1.71!!!
The SPSS says: "The value is negative due to a negative average covariance
among items. This violates reliability model assumptions. You may want to
check item codings".
what does it mean and what should i do?
thanks
  Humphrey


---------------------------------
Do you Yahoo!?
 Everyone is raving about the  all-new Yahoo! Mail.

[Mark A Davenport MADAVENP]

Here is the link to the white paper.  It's on their support site so you
might have to log in.  Thanks to David Nichols for a great paper!

http://support.spss.com/Tech/Products/SPSS/Documentation/Statistics/articles/negalpha.htm

***************************************************************************************************************************************************************
Mark A. Davenport Ph.D.
Senior Research Analyst
Office of Institutional Research
The University of North Carolina at Greensboro
336.256.0395
[hidden email]

'An approximate answer to the right question is worth a good deal more
than an exact answer to an approximate question.' --a paraphrase of J. W.
Tukey (1962)






Humphrey <[hidden email]>
Sent by: "SPSSX(r) Discussion" <[hidden email]>
10/25/2006 05:46 AM
Please respond to
Humphrey <[hidden email]>


To
[hidden email]
cc

Subject
Reliability problem






        Hi,
  I computed cronbach alpha reliability for 14 items 52 cases.
it is -1.71!!!
The SPSS says: "The value is negative due to a negative average covariance
among items. This violates reliability model assumptions. You may want to
check item codings".
what does it mean and what should i do?
thanks
  Humphrey


---------------------------------
Do you Yahoo!?
 Everyone is raving about the  all-new Yahoo! Mail.

[statisticsdoc]

Stephen Brand
www.statisticsdoc.com

Humphrey,

Mark has supplied a link to a great paper.  Your alpha is not only negative,
it is in excess of 1 in absolute value.  I would  check to see that your
items are all keyed in the same direction as you want the item total to be
scored.  If you have items that are negatively worded (i.e., where lower
scores rather than higher ones indicate higher levels of the construct being
measured), then reverse score them before computing alpha (and before
computing the scale total).

HTH,

Stephen Brand

For personalized and professional consultation in statistics and research
design, visit
www.statisticsdoc.com


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]]On Behalf Of
Humphrey
Sent: Wednesday, October 25, 2006 5:47 AM
To: [hidden email]
Subject: Reliability problem


        Hi,
  I computed cronbach alpha reliability for 14 items 52 cases.
it is -1.71!!!
The SPSS says: "The value is negative due to a negative average covariance
among items. This violates reliability model assumptions. You may want to
check item codings".
what does it mean and what should i do?
thanks
  Humphrey


---------------------------------
Do you Yahoo!?
 Everyone is raving about the  all-new Yahoo! Mail.