Using SPSS for descriptive comparative stats between a respondent and the sample population

classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

Using SPSS for descriptive comparative stats between a respondent and the sample population

ericanicoledsouza
This post was updated on .
I have conducted a survey with 800 respondents.

I am looking to give each respondent comparative feedback of how they have done compared to the average of the whole sample population. 

Eg: Each respondent will be given a score on the healthiness of their overall menu. I need to provide them with descriptive statistics of how their overall menu score has compared with the average score for overall menus of the rest of the sample population.

I want to find out the method of doing this using the SPSS statistical analysis software and if so if anyone has any online tutorials/ guides on how I can do this.
Any examples will also be valuable.

Many thanks!!
Reply | Threaded
Open this post in threaded view
|

Re: Using SPSS for descriptive comparative stats between a respondent and the sample population

Maguin, Eugene
I think this is a problem for the aggregate command. The idea is to add a new variable to each record that is the average of all the records, i.e., the sample mean. So for some person you want to say, your average is 4.35; the average of the whole group is 5.01. I think that satisfies what you wanted. So. Do this.

Aggregate outfile=* mode=addvariables/groupmean=mean(score).

Now then, the person's score is included in the computation of the group mean. You may not want that but if so, something can be done but you can ask.
Gene Maguin



-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ericanicoledsouza
Sent: Tuesday, October 18, 2016 6:57 PM
To: [hidden email]
Subject: Using SPSS for descriptive comparative stats between a respondent and the sample population

I have conducted a survey with 800 respondents.

I am looking to give each respondent comparative feedback of how they have done compared to the average of the whole sample population. 

Eg: Each respondent will be given a score on the healthiness of their menu.
I need to provide them with descriptive statistics of how their menu score has compared with the average score of the rest of the sample population.

I want to find out the method of doing this using the SPSS statistical analysis software and if so if anyone has any online tutorials/ guides on how I can do this.
Any examples will also be valuable.

Many thanks!!




--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Using-SPSS-for-descriptive-comparative-stats-between-a-respondent-and-the-sample-population-tp5733301.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

=====================
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
Reply | Threaded
Open this post in threaded view
|

Re: Using SPSS for descriptive comparative stats between a respondent and the sample population

Rich Ulrich
In reply to this post by ericanicoledsouza

Feedback to a general population?  IQ does that.  Percentiles are more intelligible to most people.


Since "healthiness" is a fairly fuzzy concept, I expect that telling them which decile they belong in

would suffice, unless you want to be even less fine-grain.


For percentile or decile, you can check the RANK procedure.  If you want to use just a few categories,

I think that you would want categories like (should be better, okay, better than most) -- which should

be based on an intelligent decision based on the meaning of the health scores.


--

Rich Ulrich



From: SPSSX(r) Discussion <[hidden email]> on behalf of ericanicoledsouza <[hidden email]>
Sent: Tuesday, October 18, 2016 6:57 PM
To: [hidden email]
Subject: Using SPSS for descriptive comparative stats between a respondent and the sample population
 
I have conducted a survey with 800 respondents.

I am looking to give each respondent comparative feedback of how they have
done compared to the average of the whole sample population. 

Eg: Each respondent will be given a score on the healthiness of their menu.
I need to provide them with descriptive statistics of how their menu score
has compared with the average score of the rest of the sample population.

I want to find out the method of doing this using the SPSS statistical
analysis software and if so if anyone has any online tutorials/ guides on
how I can do this.
Any examples will also be valuable.

Many thanks!!

===================== 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
Reply | Threaded
Open this post in threaded view
|

Re: Using SPSS for descriptive comparative stats between a respondent and the sample population

Bruce Weaver
Administrator
In reply to this post by Maguin, Eugene
I think what Gene is suggesting makes sense.  

And then, given that you want 800 separate reports (one per participant), you could use SPLIT FILE, splitting on an ID code.  Something like this:

AGGREGATE outfile=* mode=addvariables overwrite=yes
  /MeanScore=mean(score).

VARIABLE LABELS
 score "Your score on X"
 MeanScore "Average X-score for all participants".

SORT CASES by ID.
SPLIT FILE SEPARATE BY ID.
SUMMARIZE
  /TABLES= ID Score MeanScore
  /FORMAT= NOCASENUM
  /TITLE="Whatever title you want here"
  /FOOTNOTE="A footnote if you want one"
  /MISSING=VARIABLE
  /CELLS=NONE.
SPLIT FILE OFF.

You could use a simple LIST command where I have SUMMARIZE.  But it would display variable names only, whereas SUMMARIZE can display the variable labels (if there are any).  This should make for a nicer looking table that participants can understand more readily, assuming good informative variable labels.

Finally, if you actually have 2 or more scores per participant, this approach scales up pretty easily.  E.g., for two scores (score1 and score2) and two means:

AGGREGATE outfile=* mode=addvariables overwrite=yes
  /MeanScore1=mean(score1)
  /MeanScore2=mean(scor2).

VARIABLE LABELS
 score1 "Your score on X1"
 score2 "Your score on X2"
 MeanScore1 "Average X1-score for all participants"
 MeanScore2 "Average X1-score for all participants".

SORT CASES by ID.
SPLIT FILE SEPARATE BY ID.
SUMMARIZE
  /TABLES= ID Score1 MeanScore1 Score2 MeanScore2
  /FORMAT= NOCASENUM
  /TITLE="Whatever title you want here"
  /FOOTNOTE="A footnote if you want one"
  /MISSING=VARIABLE
  /CELLS=NONE.
SPLIT FILE OFF.

Provided you don't have too many scores to report, this should generate a pretty decent looking table for each participant.  If you have a large number of scores, the summary table will get to wide, and some other approach will be needed (e.g., perhaps restructuring the data file from wide to long such that you have one row per each participant x score combination).  

HTH.

Maguin, Eugene wrote
I think this is a problem for the aggregate command. The idea is to add a new variable to each record that is the average of all the records, i.e., the sample mean. So for some person you want to say, your average is 4.35; the average of the whole group is 5.01. I think that satisfies what you wanted. So. Do this.

Aggregate outfile=* mode=addvariables/groupmean=mean(score).

Now then, the person's score is included in the computation of the group mean. You may not want that but if so, something can be done but you can ask.
Gene Maguin



-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ericanicoledsouza
Sent: Tuesday, October 18, 2016 6:57 PM
To: [hidden email]
Subject: Using SPSS for descriptive comparative stats between a respondent and the sample population

I have conducted a survey with 800 respondents.

I am looking to give each respondent comparative feedback of how they have done compared to the average of the whole sample population. 

Eg: Each respondent will be given a score on the healthiness of their menu.
I need to provide them with descriptive statistics of how their menu score has compared with the average score of the rest of the sample population.

I want to find out the method of doing this using the SPSS statistical analysis software and if so if anyone has any online tutorials/ guides on how I can do this.
Any examples will also be valuable.

Many thanks!!




--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Using-SPSS-for-descriptive-comparative-stats-between-a-respondent-and-the-sample-population-tp5733301.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

=====================
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
bweaver@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.
Reply | Threaded
Open this post in threaded view
|

Re: Using SPSS for descriptive comparative stats between a respondent and the sample population

ericanicoledsouza
In reply to this post by ericanicoledsouza
Dear all,

Thank you so much for your help and guidance.

Cheers!