Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: programming question for instrument scoring with weights
From
Stas Kolenikov <[email protected]>
To
[email protected]
Subject
Re: st: programming question for instrument scoring with weights
Date
Thu, 9 Jun 2011 11:14:18 -0500
Right. Did not RTFM carefully enough. Marguerite may have to follow
the -egen = rowtotal()- with something like
egen miss_items = rowmiss( [all contributing items] )
replace total_score = . if miss_items
if the missing values in the items are to produce missing values in
the total score.
On Thu, Jun 9, 2011 at 11:06 AM, Austin Nichols <[email protected]> wrote:
> Stas--
>
> Not well, though:
>
> . egen m=rowtotal(rep78 price), missing
>
> . su m rep78 price
>
> Variable | Obs Mean Std. Dev. Min Max
> -------------+--------------------------------------------------------
> m | 74 6168.432 2949.481 3294 15909
> rep78 | 69 3.405797 .9899323 1 5
> price | 74 6165.257 2949.496 3291 15906
>
>
>
> On Thu, Jun 9, 2011 at 12:02 PM, Stas Kolenikov <[email protected]> wrote:
>> Oops. Thanks, Austin! There's -missing- option of -egen = rowtotal()-
>> controlling for the missing value issue, too.
>>
>> On Thu, Jun 9, 2011 at 10:48 AM, Austin Nichols <[email protected]> wrote:
>>> Stas; Marguerite <[email protected]>:
>>> I think the -egen- "function" below is meant to be rowtotal(), not
>>> total(). But beware of missing values counting as zeros in such a
>>> formulation.
>>>
>>> On Thu, Jun 9, 2011 at 10:52 AM, Stas Kolenikov <[email protected]> wrote:
>>> ...
>>>> I am assuming that your items are coded 1 for yes, 0 for no. Or you
>>>> can do this sequentially:
>>>>
>>>> egen int total_score = total( item1 item2 [fill in the dots] item19 )
>>>> replace total_score = total_score + 4 if item2 == 1
>>>> ...
>>>> replace total_score = total_score - 3 if item3a == 1
>>>>
>>>> All of these ways are cumbersome, and the more lines of code you have,
>>>> the more likely it is you will have an error somewhere. The choice
>>>> between different ways of coding your problem is a matter of
>>>> aesthetics, and how readable the code will be to you and other users
>>>> of it.
>
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
>
--
Stas Kolenikov, also found at http://stas.kolenikov.name
Small print: I use this email account for mailing lists only.
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/