Quantcast

How to run an ICC in SPSS 15.0

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

How to run an ICC in SPSS 15.0

Marc L. Steinberg
I apologize if this is too broad a question, but I need to run an ICC to
check on inter-rater reliability between three coders rating multiple
constructs (listening to therapy tapes).

I set up the dataset so each coder is a nominal variable.
Some constructs are given a rating of 1 (low) to 5 (high).
Some constructs are pure behavioral counts (e.g. how many times did the
therapist do behavior X).

When I run the ICC, I'm just entering two variables at a time: coder and the
variable of interest.  Is this correct?


Any thoughts?  thanks!  marc

=====================
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
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: How to run an ICC in SPSS 15.0

Meyer, Gregory J
Mark, you will need a column for each coder and construct. e.g.,
Construct1_Coder1, Construct1_Coder2, Construct1_Coder3,
Construct2_Coder1, etc. The rows will be the tapes or tape segments. The
ICC allows you to examine all 3 coders simultaneously, not just as
pairwise comparisons.

Greg

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Marc L. Steinberg
Sent: Wednesday, January 02, 2008 12:04 PM
To: [hidden email]
Subject: How to run an ICC in SPSS 15.0

I apologize if this is too broad a question, but I need to run an ICC to
check on inter-rater reliability between three coders rating multiple
constructs (listening to therapy tapes).

I set up the dataset so each coder is a nominal variable.
Some constructs are given a rating of 1 (low) to 5 (high).
Some constructs are pure behavioral counts (e.g. how many times did the
therapist do behavior X).

When I run the ICC, I'm just entering two variables at a time: coder and
the
variable of interest.  Is this correct?


Any thoughts?  thanks!  marc

=====================
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
Loading...