cal. mean minus the lowest and highest scores

[msherman]

Dear List: I have searching the internet for a compute statement that will allow me to calculate the mean without the highest and lowest scores. That is, I have five raters and I want to get their mean score per proposal and not include the highest rating and the lowest rating (only the three middle ratings). I haven’t found anything yet.  Is there a command in spss that will allow me to do this? Thanks,  martin

 

Martin F. Sherman, Ph.D.

Professor of Psychology

Director of  Masters Education in Psychology: Thesis Track

 

Loyola University Maryland

Department of Psychology

222 B Beatty Hall

4501 North Charles Street

Baltimore, MD 21210

 

410-617-2417

[hidden email]

 

[Richard Ristow]

At 04:05 PM 3/10/2014, Martin Sherman wrote:

>Dear List: I have searching the internet for a compute statement
>that will allow me to calculate the mean without the highest and
>lowest scores. That is, I have five raters and I want to get their
>mean score per proposal and not include the highest rating and the
>lowest rating (only the three middle ratings).

This is crude; but, if your five values are in variables RATER1 to
RATER5, which are contiguous in the file, maybe this will work:

IF NVALID(RATER1 TO RATER5) GE 3
         ChopMean = (SUM(RATER1 TO RATER5)
                   - MAX(RATER1 TO RATER5)
                   - MIN(RATER1 TO RATER5))/(NVALID(RATER1 TO RATER5) - 2).

=====================
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]

Martin, assuming the 5 ratings are in 5 variables (with one row per subject) you might consider using the following functions in a COMPUTE statement:

SUM
MIN
MAX
NVALID

HTH.

msherman wrote

Dear List: I have searching the internet for a compute statement that will allow me to calculate the mean without the highest and lowest scores. That is, I have five raters and I want to get their mean score per proposal and not include the highest rating and the lowest rating (only the three middle ratings). I haven't found anything yet.  Is there a command in spss that will allow me to do this? Thanks,  martin

Martin F. Sherman, Ph.D.
Professor of Psychology
Director of  Masters Education in Psychology: Thesis Track

Loyola University Maryland
Department of Psychology
222 B Beatty Hall
4501 North Charles Street
Baltimore, MD 21210

410-617-2417
[hidden email]<mailto:[hidden email]>

[David Marso]

Martin should be more explicit about what should be done if there are NOT 5 ratings.
It would be ODD in my opinion to use a single middle rating in the case where there are only 3 valid ratings.  If only four then do we include the MIN or the MAX value?
I would be very cautious in proposing code without this clarification.
If there are complete data Richard's code can be simplified to:

COMPUTE chopmean = (SUM.5(v1 TO v5)-MIN.5(v1 TO v5)-MAX.5(min1 TO min5) )/3.

If there are missing data in a case the function will evaluate to SYSMIS.

Richard Ristow wrote

At 04:05 PM 3/10/2014, Martin Sherman wrote:

>Dear List: I have searching the internet for a compute statement
>that will allow me to calculate the mean without the highest and
>lowest scores. That is, I have five raters and I want to get their
>mean score per proposal and not include the highest rating and the
>lowest rating (only the three middle ratings).

This is crude; but, if your five values are in variables RATER1 to
RATER5, which are contiguous in the file, maybe this will work:

IF NVALID(RATER1 TO RATER5) GE 3
         ChopMean = (SUM(RATER1 TO RATER5)
                   - MAX(RATER1 TO RATER5)
                   - MIN(RATER1 TO RATER5))/(NVALID(RATER1 TO RATER5) - 2).

=====================
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]

Which raises the question of the underlying purpose.  If the goal is robustness/ outlier influence reduction, then this scheme may throw away data unnecessarily, and something like an M-estimator (available in EXAMINE) might be a better choice, although that would require the infamous VARSTOCASES and SPLIT FILES and a little other footwork.


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




From:        David Marso <[hidden email]>
To:        [hidden email],
Date:        03/10/2014 04:42 PM
Subject:        Re: [SPSSX-L] cal. mean minus the lowest and highest scores
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

Martin should be more explicit about what should be done if there are NOT 5
ratings.
It would be ODD in my opinion to use a single middle rating in the case
where there are only 3 valid ratings.  If only four then do we include the
MIN or the MAX value?
I would be very cautious in proposing code without this clarification.
If there are complete data Richard's code can be simplified to:

COMPUTE chopmean = (SUM.5(v1 TO v5)-MIN.5(v1 TO v5)-MAX.5(min1 TO min5) )/3.

If there are missing data in a case the function will evaluate to SYSMIS.


Richard Ristow wrote
> At 04:05 PM 3/10/2014, Martin Sherman wrote:
>
>>Dear List: I have searching the internet for a compute statement
>>that will allow me to calculate the mean without the highest and
>>lowest scores. That is, I have five raters and I want to get their
>>mean score per proposal and not include the highest rating and the
>>lowest rating (only the three middle ratings).
>
> This is crude; but, if your five values are in variables RATER1 to
> RATER5, which are contiguous in the file, maybe this will work:
>
> IF NVALID(RATER1 TO RATER5) GE 3
>          ChopMean = (SUM(RATER1 TO RATER5)
>                    - MAX(RATER1 TO RATER5)
>                    - MIN(RATER1 TO RATER5))/(NVALID(RATER1 TO RATER5) -
> 2).
>
> =====================
> 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





-----
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me.
---
"Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis."
Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?"
--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/cal-mean-minus-the-lowest-and-highest-scores-tp5724811p5724814.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