Averaging and Computing Variables Based on Conditions
Locked
Averaging and Computing Variables Based on Conditions
 
|
Hi all...I apologize if this was sent once before. I think I sent it out before I subscribed. Hopefully this email goes out.
I have a dataset that looks something like this...
Code q1 q2 ReadingScore 12345 1 0 521
12345 1 1 530 5421 1 1 651 5421 0 0 402
My goal is to average the ReadingScore based on the questions and Code. So I want all of the people who answered 1 (Yes) from Code 12345 to have their ReadingScores averaged together and created into a new variable. I would also like to have all of the people who answered No (0) from 5421 to have their ReadingScore's averaged together and have that created into a new variable. Etc...
Is this possible? I would appreciate any help...or just putting me down the right path. |
Automatic reply: Averaging and Computing Variables Based on Conditions
|
|
I'm currently traveling on business and will return to the office on Tuesday, June 18th. If you need to reach someone sooner please contact Janice Pearcy at [hidden email].
Thank you, Kim Jinnett |
Re: Averaging and Computing Variables Based on Conditions
|
|
In reply to this post by Zachrey Santangelo
I think that you will want to use Aggregate with the break variables being code and q1 (Is it q1, q2 or both q1 and q2? You don’t specify). I’m guessing you want to create a new data set with variables of code q1 and average reading score. But, if you want to add the new variable to the existing dataset use the mode=addvariables subcommand. Gene Maguin From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Zachrey Santangelo Hi all...I apologize if this was sent once before. I think I sent it out before I subscribed. Hopefully this email goes out. I have a dataset that looks something like this... Code q1 q2 ReadingScore 12345 1 0 521 12345 1 1 530 5421 1 1 651 5421 0 0 402 My goal is to average the ReadingScore based on the questions and Code. So I want all of the people who answered 1 (Yes) from Code 12345 to have their ReadingScores averaged together and created into a new variable. I would also like to have all of the people who answered No (0) from 5421 to have their ReadingScore's averaged together and have that created into a new variable. Etc... Is this possible? I would appreciate any help...or just putting me down the right path. |
Re: Averaging and Computing Variables Based on Conditions
|
|
Thank you for the response Gene. It is q1, q2. I did not want to combine them. Sorry about not specifying. So this...? AGGREGATE /outfile = * Mode=addvariables
/break = Code q1 /RdgScoreMean = MEAN(ReadingScore). I am guessing I need to redo this statement for q2. So this gives me two scores under RdgScoreMean. One for Yes and one for No to q1. Is there any way to break that up? Have one column for Yes to Q1 and one for No to Q1?
On Mon, Jun 10, 2013 at 1:19 PM, Maguin, Eugene <[hidden email]> wrote:
|
Re: Averaging and Computing Variables Based on Conditions
|
In reply to this post by Zachrey Santangelo
Gene's solution may save the averages as a new variable, but my way gives
you a chance to eyeball the data first. Curiosity alone would draw me to investigate the combinations as well. From: Zachrey Santangelo [mailto:[hidden email]] Sent: 10 June 2013 20:09 To: John F Hall Cc: Bruce Weaver Subject: Re: Averaging and Computing Variables Based on Conditions Hi John, Thank you for the quick reply. I only have around 40 codes, so the first option is definitely viable. In the second option am I not combining q1 and q2? I was trying to keep q1 and q2 separate. Sorry I did not specify that in my initial question. On Mon, Jun 10, 2013 at 12:42 PM, John F Hall <[hidden email]> wrote: Quick reaction is (in syntax): File > new > syntax, then type in: means Readingscore by code by q1 q2. . . and run the job. Hope you haven't got thousands of codes! You could also try: compute comb = q1 * 10 + q2. format q1q2 (n2). means Readingscore by code by comb. John F Hall (Mr) [Retired academic survey researcher] Email: [hidden email] Website: www.surveyresearch.weebly.com SPSS start page: www.surveyresearch.weebly.com/spss-without-tears.html From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Zachrey Santangelo Sent: 10 June 2013 18:07 To: [hidden email] Subject: Averaging and Computing Variables Based on Conditions Hi all...I apologize if this was sent once before. I think I sent it out before I subscribed. Hopefully this email goes out. I have a dataset that looks something like this... Code q1 q2 ReadingScore 12345 1 0 521 12345 1 1 530 5421 1 1 651 5421 0 0 402 My goal is to average the ReadingScore based on the questions and Code. So I want all of the people who answered 1 (Yes) from Code 12345 to have their ReadingScores averaged together and created into a new variable. I would also like to have all of the people who answered No (0) from 5421 to have their ReadingScore's averaged together and have that created into a new variable. Etc... Is this possible? I would appreciate any help...or just putting me down the right path. ===================== 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 |
Re: Averaging and Computing Variables Based on Conditions
Administrator
|
In reply to this post by Zachrey Santangelo
"Is there any way to break that up? Have one column for Yes to Q1 and one for No to Q1? "
Follow the AGGREGATE(s) with a CASESTOVARS. OTOH: If you are simply after the MEANS you might want to drop the MODE=ADDVARIABLES from the AGGREGATE commands. When you combine aggregated data into a RAW file you end up with a mix of information at different levels of observation. I will stop now because we have NO idea what you need to do with the summary data.
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?" |
Re: Averaging and Computing Variables Based on Conditions
|
|
I just wanted to report the Reading Score Averages for people that answered Yes to q1 and the Reading Score averages for people that responded no to question 1. Same thing with question 2.
Also wanted them grouped by Code. On Mon, Jun 10, 2013 at 3:53 PM, David Marso <[hidden email]> wrote:
|
Re: Averaging and Computing Variables Based on Conditions
Administrator
|
If all you want to do is report results (i.e., there are no further computations involving those means), then go with John's one-liner:
MEANS Readingscore BY code BY q1 q2.
--
Bruce Weaver bweaver@lakeheadu.ca http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." PLEASE NOTE THE FOLLOWING: 1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. 2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/). |
«
Return to SPSSX Discussion
|
1 view|%1 views
| Free forum by Nabble | Edit this page |