syntax for stanine

[E. Bernardo]

Hi all,

I want to transform a variable into stanine.  Do you have syntax for this?

 

Thank you.

Eins

---

Have a new Yahoo! Mail account?
Kick start your journey by importing all your contacts!

[Daciuk, Tim]

I would think that the syntax for a variable called OldVar would look something like this:

 

RANK VARIABLES= OldVar (A)

  /PERCENT

  /PRINT=YES

  /TIES=MEAN.

RECODE OldVar (Lowest thru 4=1) (4 thru 11=2) (11 thru 23=3) (23 thru 40=4) (40 thru 60=5) (60

    thru 77=6) (77 thru 89=7) (89 thru 96=8) (96 thru Highest=9) INTO Stanine.

EXECUTE.

 

The only issue might be that for exact precision, you would need to know how the ranking apriori of the stanine calculation deals with ties.  I have assumed ties=mean, but, am really not sure of the ‘actual’ way this is handled.

 

Tim Daciuk

Director, Worldwide Demo Resources

SPSS Inc.

 

 

---

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Eins Bernardo
Sent: Monday, February 23, 2009 7:26 PM
To: [hidden email]
Subject: syntax for stanine

 

Hi all,

I want to transform a variable into stanine.  Do you have syntax for this?

 

Thank you.

Eins

 

---

Have a new Yahoo! Mail account?
Kick start your journey by importing all your contacts!

[Luca Meyer-3]

What kind of data do you have? Can you provide an example of your dataset?

 

Meanwhile this might provide you with some clue: http://dictionary.reference.com/browse/stanine

 

Cheers,

Luca

 

Luca MEYER

www.lucameyer.com

---

Da: SPSSX(r) Discussion [mailto:[hidden email]] Per conto di Eins Bernardo
Inviato: martedì 24 febbraio 2009 1.26
A: [hidden email]
Oggetto: syntax for stanine

Hi all,

I want to transform a variable into stanine.  Do you have syntax for this?

 

Thank you.

Eins

---

Have a new Yahoo! Mail account?
Kick start your journey by importing all your contacts!

[Art Kendall]

I have not used stanines since about 1975, so I did this to check my understanding.
using PERCENT and NORMAL in RANK gives the same results.

Open a new instance of SPSS.  Copy the syntax below to a syntax file. Click <run>. Click <all>.
programming style: Note the use of fixed width font and to create the vertical alignment of the first RECODE.
This helps with eye-checking syntax and with communicating the logic to others.


* make up some data.
INPUT PROGRAM.
LOOP id=1 TO 1000.
COMPUTE x = rnd(rv.uniform(0=-.5,100.499)).
COMPUTE y = rv.normal(5,2).
compute z = rv.t(30).
END CASE.
END LOOP.
END FILE.
END INPUT PROGRAM.
FORMATS id (F3.0).
* create stanine i.e., mean=5, sd =2.
* each value interval is 1/2 SD wide except the end ones.
RANK VARIABLES=  x y z
   /NORMAL
    INTO z_x z_y z_z.
recode  z_x z_y z_z
 (low   thru -1.75 =1)
 (-1.75 thru -1.25 =2)
 (-1.25 thru -0.75 =3)
 (-0.75 thru -0.25 =4)
 (-0.25 thru  0.25 =5)
 ( 0.25 thru  0.75 =6)
 ( 0.75 thru  1.25 =7)
 ( 1.25 thru  1.75 =8)
 ( 1.75 thru    hi =9)
 into via_norm_x via_norm_y via_norm_z.
formats via_norm_x via_norm_y via_norm_z (f1).
RANK VARIABLES= x y z (A)
  /PERCENT
  into pct_x pct_y pct_z.
RECODE  pct_x pct_y pct_z(Lowest thru 4=1) (4 thru 11=2) (11 thru 23=3) (23 thru 40=4) (40 thru 60=5) (60
    thru 77=6) (77 thru 89=7) (89 thru 96=8) (96 thru Highest=9) INTO via_pct_x via_pct_y via_pct_z.

formats via_pct_x via_pct_y via_pct_z (f1).
* compare results of 2 methods.
CROSSTABS tables = via_norm_x by via_pct_x/via_norm_y by via_pct_y/via_norm_z by via_pct_z.

Art Kendall
Social Research Consultants
Eins Bernardo wrote:

Hi all,

I want to transform a variable into stanine.  Do you have syntax for this?

 

Thank you.

Eins

---

Have a new Yahoo! Mail account?
Kick start your journey by importing all your contacts!