Coefficient Alpha

Article View: bit.listserv.stat-l

Article #3548

Back to bit.listserv.stat-l Previous Next

#3548

From: PAVTROD
Date: Sun, 16 Feb 1992 04:56

9 lines
1203 bytes

I have a simple question about Coefficient Alpha.  I'm running
alphas as a follow up to a factor analysis (one alpha per factor)
to assess the internal consistency of each factor.  I have two questions.
First, several of my factors contained a small number of items (3-4).  The
alphas for those factors were lower than those factors which contained more
items.  Does the alpha tend to be lower for factors with low items counts?
My second question concerns the alpha level.  What reliability level should
factors have in a factor analysis.  Two out of my seven are below .80.  Is
this a matter of serious concern?  Thanks for your assistance.

Technical Details

Message-ID: <STAT-L%92021603211037@VM1.MCGILL.CA>

Path: arc1-ca.pugleaf.net!sparky!uunet!paladin.american.edu!auvm!UMCVMB.BITNET!PAVTROD

Show Full Headers

Complete Article Headers

X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: f5efb,99a4f483a35f9b53,start
X-Google-Attributes: gidf5efb,public
X-Google-ArrivalTime: 1992-02-16 06:43:48 PST
Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
Path: arc1-ca.pugleaf.net!sparky!uunet!paladin.american.edu!auvm!UMCVMB.BITNET!PAVTROD
Message-ID: <STAT-L%92021603211037@VM1.MCGILL.CA>
Newsgroups: bit.listserv.stat-l
Date:         Sat, 15 Feb 1992 22:56:38 CST
Sender:       "STATISTICAL CONSULTING" <STAT-L@MCGILL1.BITNET>
From:         PAVTROD@UMCVMB.BITNET
Subject:      Coefficient Alpha