(no subject)

[Åsa M]

I am trying to create a total score of a questionnaire. It includes 22 items and I would like cases to be included even if values are missing on up to 7 items (i.e. as long as there is data on 15 items, a total score should be calculated).
 
How do I get SPSS to include cases with missing values when using "compute" to create a new variable?
 
Thank you!!!!
_________________________________________________________________
Har karriären somnat av? Här hittar du lediga jobb & karriärtips!
http://msn.jobbguiden.se/SE/Custom/MSN/Default.aspx
====================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

[Judith Saebel]

Try this:
 
compute newvar = sum.15(var1,var2,var3,var4,var5,var6,var7,var8,var9,var10,var11,var12,
                                        var13,var14,var15,var16,var17,var18,var19,vat20,var21,var22).
 
 
HTH,
 
Judith Saebel

________________________________

From: SPSSX(r) Discussion on behalf of Åsa M
Sent: Wed 21/11/2007 20:41
To: [hidden email]
Subject:



I am trying to create a total score of a questionnaire. It includes 22 items and I would like cases to be included even if values are missing on up to 7 items (i.e. as long as there is data on 15 items, a total score should be calculated).

How do I get SPSS to include cases with missing values when using "compute" to create a new variable?

Thank you!!!!
_________________________________________________________________
Har karriären somnat av? Här hittar du lediga jobb & karriärtips!
http://msn.jobbguiden.se/SE/Custom/MSN/Default.aspx
=======
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

[Spousta Jan]

Hi Asa,

There are many ways, the simplest is

Compute  mymean = mean.15(var1 to var22).

Or you can first compute the number of missings and skip the "bad" cases:

Do if (nmiss(var1 to var22) <= 7).
* here the transformations.
End if.

Hope this helps

Jan


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Asa M
Sent: Wednesday, November 21, 2007 11:12 AM
To: [hidden email]
Subject:

I am trying to create a total score of a questionnaire. It includes 22 items and I would like cases to be included even if values are missing on up to 7 items (i.e. as long as there is data on 15 items, a total score should be calculated).

How do I get SPSS to include cases with missing values when using "compute" to create a new variable?

Thank you!!!!
_________________________________________________________________
Har karriären somnat av? Här hittar du lediga jobb & karriärtips!
http://msn.jobbguiden.se/SE/Custom/MSN/Default.aspx
=======
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



_____
�
Tato zpráva a všechny připojené soubory jsou důvěrné a určené výlučně adresátovi(-ům). Jestliže nejste oprávněným adresátem, je zakázáno jakékoliv zveřejňování, zprostředkování nebo jiné použití těchto informací. Jestliže jste tento mail dostali neoprávněně, prosím, uvědomte odesilatele a smažte zprávu i přiložené soubory. Odesilatel nezodpovídá za jakékoliv chyby nebo opomenutí způsobené tímto přenosem.
�
This message and any attached files are confidential and intended solely for the addressee(s). Any publication, transmission or other use of the information by a person or entity other than the intended addressee is prohibited. If you receive this in error please contact the sender and delete the message as well as all attached documents. The sender does not accept liability for any errors or omissions as a result of the transmission.

-.- --

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

[Richard Ristow]

At 05:11 AM 11/21/2007, Åsa M wrote:

>I am trying to create a total score of a
>questionnaire. It includes 22 items and I would
>like cases to be included even if values are
>missing on up to 7 items (i.e. as long as there
>is data on 15 items, a total score should be calculated).

Other respondents have told you, correctly, how
to do this; for example, Jan Spousta:

>There are many ways, the simplest is
>Compute  MySum = sum.15(var1 to var22).

A second question is, whether to do it;
particularly, whether to do it that way.

This computation effectively treats the missing
values as zeroes. If the true, unobserved values
have mean greater than zero, the presence of
missing values biases the sum down, perhaps seriously.

There's no universal solution. If all the items
are expected to have roughly equal expected
values in any one case, and you want to estimate
the *mean*, not the sum, Jan Spousta's

>Compute  mymean = mean.15(var1 to var22).

is a good choice. If you need an estimate of the
sum, for example to estimate whether the score
exceeds threshold defined in terms of the sum, try

COMPUTE  EstimSum  = mean.15(var1 to var22)*nvalid(var1 to var22).

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

[Richard Ristow]

Correcting a mistake:

On Nov 21, 2007 5:52 PM, I wrote,

>To: [hidden email]
>Subject: Re: compute if there is less than 7 missing values
>
>At 05:11 AM 11/21/2007, Åsa M wrote:
>
>>I am trying to create a total score of a
>>questionnaire. It includes 22 items and I would
>>like cases to be included even if values are
>>missing on up to 7 items

>If you need an estimate of the
>sum, for example to estimate whether the score
>exceeds threshold defined in terms of the sum, try
>
>COMPUTE  EstimSum  = mean.15(var1 to var22)*nvalid(var1 to var22).

That gives the sum of the NON-MISSING values (by a convoluted route you'd never use). The correct expression to estimate the sum of all 22 values is

COMPUTE  EstimSum  = 22*mean.15(var1 to var22).

Again, this is based on the assumption that all items have roughly equal means; and that the 'missing' items have actual values, typical of those items' distributions, that happen not to be observed.

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