doubly-entry data structure

classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

doubly-entry data structure

Empi
Dear all

I would like to compute the double-entry Intraclass correlation, but can’t
develop  the syntax necessary to restructure the data - as described e.g. by
Robert R. McCrae ((2008) A Note on Some Measures of Profile Agreement,
Journal of Personality Assessment, 90:2, 105-109), computation of this
correlation requires to enter each element (with variables in the rows and
cases in the columns) in the paired profile to be entered “twice but in
reversed order across rows; the entry in the first column is entered in the
second column on a new row, and the entry that had been in the second column
is placed in the first column on that new row.”  Here is the example data
used by McCrae:

Single entry:

Variable Person 1 Person 2

X1 50 25
X2 55 30
X3 60 35
X4 65 40
X5 70 45

Data  structure needed for the double entry ICC:
Variable Column 1 Column 2

X1   50 25
       25 50

X2   55 30
       30 55

X3   60 35
       35 60

X4   65 40
       40 65

X5   70 45
       45 70

Any thoughts on how to go about this would be most appreciated…
Best,
Empi




--
Sent from: http://spssx-discussion.1045642.n5.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
Reply | Threaded
Open this post in threaded view
|

Re: doubly-entry data structure

Bruce Weaver
Administrator
Try this:

NEW FILE.
DATASET CLOSE ALL.
DATA LIST LIST / Element  Obs1a Obs2a (3F5.0).
BEGIN DATA
1 50 25
2 55 30
3 60 35
4 65 40
5 70 45
END DATA.

CORRELATIONS Obs1a WITH Obs2a.
* This matches the r_SR = 1.00 in Table 1.

* Reverse the order of the scores for Obs1b and Obs1b.
COMPUTE Obs1b = Obs2a.
COMPUTE Obs2b = Obs1a.
FORMATS Obs1b Obs2b (F5.0).
LIST.

* Restructure data into two columns as shown in Table 1.
VARSTOCASES
  /MAKE Obs1 FROM Obs1a Obs1b
  /MAKE Obs2 FROM Obs2a Obs2b
  /INDEX=Index(2)
  /KEEP=Element
  /NULL=KEEP.
LIST.

CORRELATIONS Obs1 WITH Obs2.
* This matches the r_12 value in Table 1.






Empi wrote

> Dear all
>
> I would like to compute the double-entry Intraclass correlation, but can’t
> develop  the syntax necessary to restructure the data - as described e.g.
> by
> Robert R. McCrae ((2008) A Note on Some Measures of Profile Agreement,
> Journal of Personality Assessment, 90:2, 105-109), computation of this
> correlation requires to enter each element (with variables in the rows and
> cases in the columns) in the paired profile to be entered “twice but in
> reversed order across rows; the entry in the first column is entered in the
> second column on a new row, and the entry that had been in the second
> column
> is placed in the first column on that new row.”  Here is the example data
> used by McCrae:
>
> Single entry:
>
> Variable Person 1 Person 2
>
> X1 50 25
> X2 55 30
> X3 60 35
> X4 65 40
> X5 70 45
>
> Data  structure needed for the double entry ICC:
> Variable Column 1 Column 2
>
> X1   50 25
>        25 50
>
> X2   55 30
>        30 55
>
> X3   60 35
>        35 60
>
> X4   65 40
>        40 65
>
> X5   70 45
>        45 70
>
> Any thoughts on how to go about this would be most appreciated…
> Best,
> Empi
>
>
>
>
> --
> Sent from: http://spssx-discussion.1045642.n5.nabble.com/
>
> =====================
> To manage your subscription to SPSSX-L, send a message to

> LISTSERV@.UGA

>  (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





-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
Sent from: http://spssx-discussion.1045642.n5.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
--
Bruce Weaver
bweaver@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.
Reply | Threaded
Open this post in threaded view
|

Re: doubly-entry data structure

Kirill Orlov
In reply to this post by Empi
Reply in view you want a ready function...
There are many versions of one-factor Intraclass correlation. I understand (reading your request) you that you may be after the inbiased initial version formulated by Pearson and called Combinatorial pairing ICC. (PLEASE say if that is not what you want.)

My matrix functions !iccow and !iccowb compute this and some other popular one-factor versions of ICC.
(Find MATRIX functions on http://www.spsstools.net/en/KO-spssmacros.)

Example.

*Situation "balanced classes" when each class has the same number of observations.
*(You might have 2+ classes in general) Input is "wide format".

data list list /class obs1 obs2.
begin data
1     50       25
2     55       30
3     60       35
4     65       40
5     70       45
end data.

matrix.
get data /variables= obs1 obs2.
!iccowb(data%PAIR%icc).
print icc.
end matrix.


*More general situation when classes may yield different number of observations.
*In this more general case, more different versions of the ICC can be computed.
*Input is usual "long format", like in ANOVA.

data list list /value class.
begin data
50        1
55        2
60        3
65        4
70        5
25        1
30        2
35        3
40        4
45        5
end data.

matrix.
get data /variable= value.
get class /variable= class.
compute class= design(class). /*turn the categorical var into dummies
!iccow(data%class%PAIR1%icc).
print icc.
end matrix.



19.07.2019 18:00, Empi пишет:
Dear all

I would like to compute the double-entry Intraclass correlation, but can’t
develop  the syntax necessary to restructure the data - as described e.g. by
Robert R. McCrae ((2008) A Note on Some Measures of Profile Agreement,
Journal of Personality Assessment, 90:2, 105-109), computation of this
correlation requires to enter each element (with variables in the rows and
cases in the columns) in the paired profile to be entered “twice but in
reversed order across rows; the entry in the first column is entered in the
second column on a new row, and the entry that had been in the second column
is placed in the first column on that new row.”  Here is the example data
used by McCrae:

Single entry:

Variable	Person 1	Person 2

X1	50	25
X2	55	30
X3	60	35
X4	65	40
X5	70	45

Data  structure needed for the double entry ICC:
Variable	Column 1	Column 2

X1   50 25	
       25 50

X2   55 30	
       30 55

X3   60 35	
       35 60

X4   65 40	
       40 65

X5   70 45	
       45 70

Any thoughts on how to go about this would be most appreciated…
Best,
Empi




--
Sent from: http://spssx-discussion.1045642.n5.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
Reply | Threaded
Open this post in threaded view
|

Re: doubly-entry data structure

Empi
In reply to this post by Bruce Weaver
Bruce - exactly the code I needed, thank you very much !!!



--
Sent from: http://spssx-discussion.1045642.n5.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
Reply | Threaded
Open this post in threaded view
|

Re: doubly-entry data structure

Empi
In reply to this post by Kirill Orlov
...Kirill, thank you for your support - and your macro-compilation is a
treasure trove, didn't know it before...
Best
E.



--
Sent from: http://spssx-discussion.1045642.n5.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