Error using RELIABILITY and DO REPEAT command

[Norberto Hernandez]

Good afternoon. I'm working in a standarization of 20 sets of affective images (each set has 60 images) and i'm trying to obtain an interjudge reliability between 5 judges. My data looks like the following:

rater1_set1 rater2_set1 rater3_set1 rater4_set1 rater5_set1.... rater4_set20 rater5_set20

1 0 1 1 1 0 1

0 1 1 0 1 0 1

1 1 1 1 1 1 1

Im trying to use the following syntax:

DO REPEAT #i=1 to 20.

RELIABILITY  VARIABLES = !CONCAT('rater1_set',#i) !CONCAT('rater2_set',#i) !CONCAT('rater3_set',#i) !CONCAT('rater4_set',#i) !CONCAT('rater5_set',#i)

  /SCALE('ALL VARIABLES') ALL

  /MODEL=ALPHA

  /ICC=MODEL(MIXED) TYPE(ABSOLUTE) CIN=95 TESTVAL=0.

END REPEAT.

But I'm getting the following error

>Error # 4097.  Nombre del comando: RELIABILITY 

>El programa de transformaciones contiene un comando DO REPEAT sin END REPEAT 

>de finalización o el bucle DO REPEAT contiene un comando que no es válido en 

>este tipo de bucle.

Apparently I can't use reliability with do repeat. I was looking in the manual and in the nabble list, but i didn't find any reference about this. Could any of you please suggest me an alternative?.

Thanks a lot guys

Norberto

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

[David Marso]

DEFINE something ()
!DO ...
!DO ...
stuff
!DOEND
!DOEND
!ENDDEFINE.

Norberto Hernandez wrote

Good afternoon. I'm working in a standarization of 20 sets of affective
images (each set has 60 images) and i'm trying to obtain an interjudge
reliability between 5 judges. My data looks like the following:

rater1_set1 rater2_set1 rater3_set1 rater4_set1 rater5_set1....
rater4_set20 rater5_set20
1 0 1 1 1 0 1
0 1 1 0 1 0 1
1 1 1 1 1 1 1

Im trying to use the following syntax:

DO REPEAT #i=1 to 20.
RELIABILITY  VARIABLES = !CONCAT('rater1_set',#i) !CONCAT('rater2_set',#i)
!CONCAT('rater3_set',#i) !CONCAT('rater4_set',#i) !CONCAT('rater5_set',#i)
  /SCALE('ALL VARIABLES') ALL
  /MODEL=ALPHA
  /ICC=MODEL(MIXED) TYPE(ABSOLUTE) CIN=95 TESTVAL=0.
END REPEAT.

But I'm getting the following error

>Error # 4097.  Nombre del comando: RELIABILITY
>El programa de transformaciones contiene un comando DO REPEAT sin END
REPEAT
>de finalización o el bucle DO REPEAT contiene un comando que no es válido
en
>este tipo de bucle.

Apparently I can't use reliability with do repeat. I was looking in the
manual and in the nabble list, but i didn't find any reference about this.
Could any of you please suggest me an alternative?.

Thanks a lot guys
Norberto

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

[Jon K Peck]

The general principle here is that DO REPEAT is a transformation command, and you can never put procedures such as RELIABILITY inside such a block.  You can do this easily with Python programmability, but a macro would also work in this case.


Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
phone: [phone]




From:        Norberto Hernandez <[hidden email]>
To:        [hidden email]
Date:        08/20/2014 12:58 PM
Subject:        [SPSSX-L] Error using RELIABILITY and DO REPEAT command
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

Good afternoon. I'm working in a standarization of 20 sets of affective images (each set has 60 images) and i'm trying to obtain an interjudge reliability between 5 judges. My data looks like the following:

rater1_set1 rater2_set1 rater3_set1 rater4_set1 rater5_set1.... rater4_set20 rater5_set20
1 0 1 1 1 0 1
0 1 1 0 1 0 1
1 1 1 1 1 1 1

Im trying to use the following syntax:

DO REPEAT #i=1 to 20.
RELIABILITY  VARIABLES = !CONCAT('rater1_set',#i) !CONCAT('rater2_set',#i) !CONCAT('rater3_set',#i) !CONCAT('rater4_set',#i) !CONCAT('rater5_set',#i)
  /SCALE('ALL VARIABLES') ALL
  /MODEL=ALPHA
  /ICC=MODEL(MIXED) TYPE(ABSOLUTE) CIN=95 TESTVAL=0.
END REPEAT.

But I'm getting the following error

>Error # 4097.  Nombre del comando: RELIABILITY 
>El programa de transformaciones contiene un comando DO REPEAT sin END REPEAT 
>de finalización o el bucle DO REPEAT contiene un comando que no es válido en 
>este tipo de bucle.

Apparently I can't use reliability with do repeat. I was looking in the manual and in the nabble list, but i didn't find any reference about this. Could any of you please suggest me an alternative?.

Thanks a lot guys
Norberto
===================== To manage your subscription to SPSSX-L, send a message to LISTSERV@... (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

[Art Kendall]

did all 5 judges rate all 1200?

Why do you have 20 sets?

What is the ultimate goal of your project?

Have you considered a design that takes all 3 things into account: Set of images, images within sets, and judges? (or more than 3 things if there is design within set)

[Norberto Hernandez]

Thanks to all the advices guys,, I'll try coding it into a macro and check the results. 

Art, this interrater validation has the purpouse to select, between the 20 sets of images, those who are better suited to use in mexican population (trying to avoid cultural bias, since the images used are from the International Affective Picture System from the USA). I'm using SPSS ICC for obtain absolute agreement between raters. And yes, the raters view and score all the 1200 images. Untill now, I didn't considere use a 3 factor design (sets x images x raters) but if you could advice me with a reference I'll appreciate a lot.

Regards.

Norberto

2014-08-21 11:42 GMT-05:00 Art Kendall <[hidden email]>:

did all 5 judges rate all 1200?

Why do you have 20 sets?

What is the ultimate goal of your project?

Have you considered a design that takes all *3 *things into account: Set of
images, images within sets, and judges? (or more than 3 things if there is
design within set)






-----
Art Kendall
Social Research Consultants
--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Error-using-RELIABILITY-and-DO-REPEAT-command-tp5727035p5727053.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