Add percentage or proportion to dataset
Locked
Add percentage or proportion to dataset
 
|
Hi I have a problem to solve and I hope you can help me I have a big data set and I have shown below a small sample to get the spss syntax. Except the gender (numeric) other variables are stringing. School_no
Year
subject
gender
Male Female 1
2010
Maths
2
0.6
0 1
2010
Maths
2
0.6
0 1
2010
Maths
2
0.6
0 1
2010
Maths
1
0
0.4 1
2010
Maths
1
0
0.4 1
2010
english
2
0.5
0 1
2010
english
2
0.5
0 1
2010
english
2
0.5
0 1
2010
english
1
0
0.5 1
2010
english
1
0
0.5 1
2010
english
1
0
0.5 1
2011
Maths
2
0.66 0 1
2011
Maths
2
0.66 0 1
2011
Maths
1
0
0.33 1
2011
english
2
0.5
0 1
2011
english
2
0.5
0 1
2011
english
1
0
0.5 1
2011
english
1
0
0.5 2
2010
Maths
2
1
0 2
2010
Maths
2
1
0 2
2010
english
2
0.5
0 2
2010 english
1
0
0.5 I need to get proportion (if possible percentage too) of gender(male=1, female=2) for a particular school_no,per year, per subject per each line. So the proportion will be same for same (School_no , Year , subject, gender).
I need to create 2 columns(One with Female and one with male as shown above. I did this in excel to show the outcome ) I hope someone can help me with the script to get Male and fFmale columns from the dataset. Thanks Ananda |
|
|
See the AGGREGATE command. I'm confused by your example though, why in the first line is Male 0.6 and Female 0? (I would expect female would be 0.4 per your criteria.) See example below.
**********************************************. DATA LIST FREE / School_no (F1.0) Year (F4.0) subject (A10) gender (F1.0) Male Female (2F2.1). BEGIN DATA 1 2010 Maths 2 0.6 0 1 2010 Maths 2 0.6 0 1 2010 Maths 2 0.6 0 1 2010 Maths 1 0 0.4 1 2010 Maths 1 0 0.4 1 2010 english 2 0.5 0 1 2010 english 2 0.5 0 1 2010 english 2 0.5 0 1 2010 english 1 0 0.5 1 2010 english 1 0 0.5 1 2010 english 1 0 0.5 1 2011 Maths 2 0.66 0 1 2011 Maths 2 0.66 0 1 2011 Maths 1 0 0.33 1 2011 english 2 0.5 0 1 2011 english 2 0.5 0 1 2011 english 1 0 0.5 1 2011 english 1 0 0.5 2 2010 Maths 2 1 0 2 2010 Maths 2 1 0 2 2010 english 2 0.5 0 2 2010 english 1 0 0.5 END DATA. AGGREGATE OUTFILE=* MODE=ADDVARIABLES /BREAK School_no Year subject /Male_Per = FGT(gender,1) /Female_Per = FLT(gender,2). **********************************************. |
Re: Add percentage or proportion to dataset
|
|
Hi Andy Thanks for your help Andy. The reason for 0 in female column is in that, it has female name etc. So can’t put the male proportion
in the female column in the same row and need to put the proportion in Female column
same row. (Gender 2=male and 1= female). 1.
How can I put zero in the line if not male or
female in the row 2.
How can I have new variable called ‘Gender_2 and
put both proportion in the same line (Female and male)? Thanks for your help in advance Ananda > Date: Wed, 25 Nov 2015 05:47:00 -0700 > From: [hidden email] > Subject: Re: Add percentage or proportion to dataset > To: [hidden email] > > See the AGGREGATE command. I'm confused by your example though, why in the > first line is Male 0.6 and Female 0? (I would expect female would be 0.4 per > your criteria.) See example below. > > **********************************************. > DATA LIST FREE / School_no (F1.0) Year (F4.0) subject (A10) gender (F1.0) > Male Female (2F2.1). > BEGIN DATA > 1 2010 Maths 2 0.6 0 > 1 2010 Maths 2 0.6 0 > 1 2010 Maths 2 0.6 0 > 1 2010 Maths 1 0 0.4 > 1 2010 Maths 1 0 0.4 > 1 2010 english 2 0.5 0 > 1 2010 english 2 0.5 0 > 1 2010 english 2 0.5 0 > 1 2010 english 1 0 0.5 > 1 2010 english 1 0 0.5 > 1 2010 english 1 0 0.5 > 1 2011 Maths 2 0.66 0 > 1 2011 Maths 2 0.66 0 > 1 2011 Maths 1 0 0.33 > 1 2011 english 2 0.5 0 > 1 2011 english 2 0.5 0 > 1 2011 english 1 0 0.5 > 1 2011 english 1 0 0.5 > 2 2010 Maths 2 1 0 > 2 2010 Maths 2 1 0 > 2 2010 english 2 0.5 0 > 2 2010 english 1 0 0.5 > END DATA. > > AGGREGATE OUTFILE=* MODE=ADDVARIABLES > /BREAK School_no Year subject > /Male_Per = FGT(gender,1) > /Female_Per = FLT(gender,2). > **********************************************. > > > > ----- > Andy W > [hidden email] > http://andrewpwheeler.wordpress.com/ > -- > View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Add-percentage-or-proportion-to-dataset-tp5731050p5731052.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 |
|
Your presented data structure is confusing; Variable Subject is not further increasing the data resolution as it is the same for each combination of school and year. I miss a subject level variable or similar.
As one of Male and Female always is 0 a quick way to (almost) get a long data format is COMPUTE Proportion = Male + Female. EXECUTE. Variables Male and Female are now redundant, but then you cannot know which cases to combine to get male and female proportions. There are 5 cases of '1 2010 Maths' and 3 cases of '2 2010 Maths'. Another way, if every case is a separate case: IF (Male=0) M_prop = 1 - Female. IF (Female=0) F_prop = 1 - Male. EXECUTE. this makes variable gender redundant (and is equivalent to the solution by Andy W) HTH, PRogman
|
Re: Add percentage or proportion to dataset
|
|
Hi Andy and PRogman
=====================
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
Thank you both very much for your help. I managed to fix the problems using the below scripts with minor changes AGGREGATE OUTFILE=* MODE=ADDVARIABLES /BREAK School_no Year subject /Male_Per = FGT(gender,1) /Female_Per = FLT(gender,2). IF (Male=0) M_prop = 1 - Female. IF (Female=0) F_prop = 1 - Male. EXECUTE. IF (Male=1) T_prop = 1 - Female. IF (Female=2) T_prop = 1 - Male. EXECUTE. Thanks again and have a nice weekend Regards Ananda > Date: Thu, 26 Nov 2015 16:51:21 -0700 > From: [hidden email] > Subject: Re: Add percentage or proportion to dataset > To: [hidden email] > > Your presented data structure is confusing; Variable Subject is not further > increasing the data resolution as it is the same for each combination of > school and year. I miss a subject level variable or similar. > As one of Male and Female always is 0 a quick way to (almost) get a long > data format is > COMPUTE Proportion = Male + Female. > EXECUTE. > Variables Male and Female are now redundant, but then you cannot know which > cases to combine to get male and female proportions. There are 5 cases of '1 > 2010 Maths' and 3 cases of '2 2010 Maths'. > > Another way, if every case is a separate case: > IF (Male=0) M_prop = 1 - Female. > IF (Female=0) F_prop = 1 - Male. > EXECUTE. > this makes variable gender redundant (and is equivalent to the solution by > Andy W) > > HTH, PRogman > > > Aoba Victoria wrote > > Hi Andy > > > > Thanks for your help Andy. > > > > The reason for 0 in female column is in that, it has female name etc. So > > can’t put the male proportion > > in the female column in the same row and need to put the proportion in > > Female column > > same row. (Gender 2=male and 1= female). > > > > 1. > > How can I put zero in the line if not male or > > female in the row > > > > 2. > > How can I have new variable called ‘Gender_2 and > > put both proportion in the same line (Female and male)? > > > > Thanks for your help in advance > > Ananda > > > >> Date: Wed, 25 Nov 2015 05:47:00 -0700 > >> From: > > > apwheele@ > > >> Subject: Re: Add percentage or proportion to dataset > >> To: > > > SPSSX-L@.UGA > > >> > >> See the AGGREGATE command. I'm confused by your example though, why in > >> the > >> first line is Male 0.6 and Female 0? (I would expect female would be 0.4 > >> per > >> your criteria.) See example below. > >> > >> **********************************************. > >> DATA LIST FREE / School_no (F1.0) Year (F4.0) subject (A10) gender > >> (F1.0) > >> Male Female (2F2.1). > >> BEGIN DATA > >> 1 2010 Maths 2 0.6 0 > >> 1 2010 Maths 2 0.6 0 > >> 1 2010 Maths 2 0.6 0 > >> 1 2010 Maths 1 0 0.4 > >> 1 2010 Maths 1 0 0.4 > >> 1 2010 english 2 0.5 0 > >> 1 2010 english 2 0.5 0 > >> 1 2010 english 2 0.5 0 > >> 1 2010 english 1 0 0.5 > >> 1 2010 english 1 0 0.5 > >> 1 2010 english 1 0 0.5 > >> 1 2011 Maths 2 0.66 0 > >> 1 2011 Maths 2 0.66 0 > >> 1 2011 Maths 1 0 0.33 > >> 1 2011 english 2 0.5 0 > >> 1 2011 english 2 0.5 0 > >> 1 2011 english 1 0 0.5 > >> 1 2011 english 1 0 0.5 > >> 2 2010 Maths 2 1 0 > >> 2 2010 Maths 2 1 0 > >> 2 2010 english 2 0.5 0 > >> 2 2010 english 1 0 0.5 > >> END DATA. > >> > >> AGGREGATE OUTFILE=* MODE=ADDVARIABLES > >> /BREAK School_no Year subject > >> /Male_Per = FGT(gender,1) > >> /Female_Per = FLT(gender,2). > >> **********************************************. > >> > >> ----- > >> Andy W > >> > > > apwheele@ > > >> http://andrewpwheeler.wordpress.com/ > >> -- > > > > ===================== > > To manage your subscription to SPSSX-L, send a message to > > > LISTSERV@.UGA > > > (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 > > > > > > -- > View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Add-percentage-or-proportion-to-dataset-tp5731050p5731059.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 |
| Free forum by Nabble | Edit this page |