Excluding questions (with missing values) from a count

[Madeline Becker]

I am trying to score a questionnaire where 2 questions will be left blank
(no response) if the respondent doesn't live in a dorm (question 1) or
have a roommate (question 2).

When calculating a subscale, if no more than 2 responses are missing from
the subscale (not counting question 1 and question 2) then the subscale
can be calculated replacing the missing values with the subscale mean.
What would I use as the syntax in order to exclude question 1 and question
2 from the missing count.

Ex.  There are 20 variables in one of the subscales and there can only be
2 missing responses (not including a no response to question 1 and
question 2), to calculate the score for the subscale.  So I know that I
would write the following syntax to compute the mean and then compute the
subscale:

COMPUTE Subscale1Mean = Mean (v1, v2, v3, ....., v20).

COMPUTE Subscale1 = SUM.18 (v1, v2, v3, ....., v20).
  IF (missing (v1)) v1 = SubscaleMean.
  Etc.

But what syntax would I use to exclude question 1 and question 2 as being
counted as the 2 responses that are missing?

[Melissa Ives]

What is the point of the scale?  Should you be calculating it just for v3 to v20 then looking at responses by dorm living and roommate status?
If you calculate it as sum.16(v3 to v20) then add in v1 and 2--would that give you what you want or something odd?
Melissa

________________________________

From: SPSSX(r) Discussion on behalf of Madeline Becker
Sent: Thu 1/25/2007 7:43 PM
To: [hidden email]
Subject: [SPSSX-L] Excluding questions (with missing values) from a count



I am trying to score a questionnaire where 2 questions will be left blank
(no response) if the respondent doesn't live in a dorm (question 1) or
have a roommate (question 2).

When calculating a subscale, if no more than 2 responses are missing from
the subscale (not counting question 1 and question 2) then the subscale
can be calculated replacing the missing values with the subscale mean.
What would I use as the syntax in order to exclude question 1 and question
2 from the missing count.

Ex.  There are 20 variables in one of the subscales and there can only be
2 missing responses (not including a no response to question 1 and
question 2), to calculate the score for the subscale.  So I know that I
would write the following syntax to compute the mean and then compute the
subscale:

COMPUTE Subscale1Mean = Mean (v1, v2, v3, ....., v20).

COMPUTE Subscale1 = SUM.18 (v1, v2, v3, ....., v20).
  IF (missing (v1)) v1 = SubscaleMean.
  Etc.

But what syntax would I use to exclude question 1 and question 2 as being
counted as the 2 responses that are missing?





PRIVILEGED AND CONFIDENTIAL INFORMATION
This transmittal and any attachments may contain PRIVILEGED AND
CONFIDENTIAL information and is intended only for the use of the
addressee. If you are not the designated recipient, or an employee
or agent authorized to deliver such transmittals to the designated
recipient, you are hereby notified that any dissemination,
copying or publication of this transmittal is strictly prohibited. If
you have received this transmittal in error, please notify us
immediately by replying to the sender and delete this copy from your
system. You may also call us at [phone] for assistance.

[Maguin, Eugene]

Madelaine,

Ok, 20 items, v1 to v20, and v1 and v2 are the two items you describe. I'll
assume that if v1 is missing and v2 is not then you would include v2 in the
scale computation and vice versa.

Do if (not(missing(v1)) and not(missing(v2))).
+  compute newvar=mean.20(v1 thru v20).
Else if (not(missing(v1)) and missing(v2)).
+  compute newvar=sum.19(v1,v3 thru v20).
Else if (missing(v1) and not(missing(v2))).
+  compute newvar=sum.19(v2,v3 thru v20).
Else.
+  compute newvar=sum.18(v3 thru v20).
End if.


Gene Maguin

[Madeline Becker]

It is an adjustment scale and the students aren't supposed to respond to
the questions about roommates or dorm living if they don't have a roommate
or are not living in a dorm.  I think what you suggested will work.  I
will give it a try.
Thanks.

On Thu, 25 Jan 2007 21:00:46 -0600, Melissa Ives <[hidden email]>
wrote:

>What is the point of the scale?  Should you be calculating it just for v3
to v20 then looking at responses by dorm living and roommate status?
>If you calculate it as sum.16(v3 to v20) then add in v1 and 2--would that
give you what you want or something odd?

>Melissa
>
>________________________________
>
>From: SPSSX(r) Discussion on behalf of Madeline Becker
>Sent: Thu 1/25/2007 7:43 PM
>To: [hidden email]
>Subject: [SPSSX-L] Excluding questions (with missing values) from a count
>
>
>
>I am trying to score a questionnaire where 2 questions will be left blank
>(no response) if the respondent doesn't live in a dorm (question 1) or
>have a roommate (question 2).
>
>When calculating a subscale, if no more than 2 responses are missing from
>the subscale (not counting question 1 and question 2) then the subscale
>can be calculated replacing the missing values with the subscale mean.
>What would I use as the syntax in order to exclude question 1 and question
>2 from the missing count.
>
>Ex.  There are 20 variables in one of the subscales and there can only be
>2 missing responses (not including a no response to question 1 and
>question 2), to calculate the score for the subscale.  So I know that I
>would write the following syntax to compute the mean and then compute the
>subscale:
>
>COMPUTE Subscale1Mean = Mean (v1, v2, v3, ....., v20).
>
>COMPUTE Subscale1 = SUM.18 (v1, v2, v3, ....., v20).
>  IF (missing (v1)) v1 = SubscaleMean.
>  Etc.
>
>But what syntax would I use to exclude question 1 and question 2 as being
>counted as the 2 responses that are missing?
>
>
>
>
>
>PRIVILEGED AND CONFIDENTIAL INFORMATION
>This transmittal and any attachments may contain PRIVILEGED AND
>CONFIDENTIAL information and is intended only for the use of the
>addressee. If you are not the designated recipient, or an employee
>or agent authorized to deliver such transmittals to the designated
>recipient, you are hereby notified that any dissemination,
>copying or publication of this transmittal is strictly prohibited. If
>you have received this transmittal in error, please notify us
>immediately by replying to the sender and delete this copy from your
>system. You may also call us at [phone] for assistance.