Determining GPA

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

Determining GPA

Bill Oglesby
Good Morning,

I have a student file (duplicated by SSN) that links students with courses and grades (e.g., 111-11-
1111, ENG 101, B), and I am trying to determine how many students have earned a GPA of 2.0 or
higher, where A = 4.0, B = 3.0, etc. Now, I am able to determine the GPA for each SSN; however, I am
dealing with a rather large file, and I don't have the patience to manually count thousands of entries.
Is it possible, instead, to produce a count, by GPA, such that I could easily determine how many
students have earned a 2.0 or higher, 3.0 or higher, etc? Thanks,

Bill

=====================
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: Determining GPA

ViAnn Beadle
Recode your GPA variable into the A, B, C, ... categories using the RECODE
command and run FREQUENCIES on it.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Bill Oglesby
Sent: Tuesday, October 14, 2008 7:31 AM
To: [hidden email]
Subject: Determining GPA

Good Morning,

I have a student file (duplicated by SSN) that links students with courses
and grades (e.g., 111-11-
1111, ENG 101, B), and I am trying to determine how many students have
earned a GPA of 2.0 or
higher, where A = 4.0, B = 3.0, etc. Now, I am able to determine the GPA for
each SSN; however, I am
dealing with a rather large file, and I don't have the patience to manually
count thousands of entries.
Is it possible, instead, to produce a count, by GPA, such that I could
easily determine how many
students have earned a 2.0 or higher, 3.0 or higher, etc? Thanks,

Bill

=====================
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: Determining GPA

DataMaestro
If I understand the problem, the solution is trickier than that.

Are all the courses 3 credits?  If not, don't you have to "weight" the grade by the number of credits first?  That might be tedious.

It sounds like there is a record for each grade for each student.  To be able to calculate the GPA by student, one household or parent record needs to be created for each student.  Then the calculation is simple.  There are examples of this data reduction technique in the documentation - I have never personally done this.  Maybe someone else in the List could help.

George



--- On Tue, 10/14/08, ViAnn Beadle <[hidden email]> wrote:

> From: ViAnn Beadle <[hidden email]>
> Subject: Re: Determining GPA
> To: [hidden email]
> Date: Tuesday, October 14, 2008, 9:49 AM
> Recode your GPA variable into the A, B, C, ... categories
> using the RECODE
> command and run FREQUENCIES on it.
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]]
> On Behalf Of
> Bill Oglesby
> Sent: Tuesday, October 14, 2008 7:31 AM
> To: [hidden email]
> Subject: Determining GPA
>
> Good Morning,
>
> I have a student file (duplicated by SSN) that links
> students with courses
> and grades (e.g., 111-11-
> 1111, ENG 101, B), and I am trying to determine how many
> students have
> earned a GPA of 2.0 or
> higher, where A = 4.0, B = 3.0, etc. Now, I am able to
> determine the GPA for
> each SSN; however, I am
> dealing with a rather large file, and I don't have the
> patience to manually
> count thousands of entries.
> Is it possible, instead, to produce a count, by GPA, such
> that I could
> easily determine how many
> students have earned a 2.0 or higher, 3.0 or higher, etc?
> Thanks,
>
> Bill
>
> =====================
> 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

=====================
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: Determining GPA

Bill Oglesby
In reply to this post by Bill Oglesby
George,

You are correct in that there is a separate record for each student, which consists of the course
enrolled in and the final grade assigned (originally coded as As, Bs, Cs, etc.). By the way, not all of
the courses are three credit hour classes, so the computation of GPA would be as follows: 4 pts for
an A, 3 for a B...0 pts for an F multiplied by the number of credit hours per course, divided by the
total number of hours taken that semester (this is probably information that I should have included
originally).

I guess, then, that there are two questions: 1) What is the least tedious way to compute GPAs, and 2)
how, then, do you determine how many students earned a 2.0 or higher?

Bill

=====================
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: Determining GPA

DataMaestro
Bill,

It appears as if you have a separate record for each student for each grade.  If so, you can use REPEATING DATA to create ONE parent record for each student which would hold separate fields for every course that each student took.  I haven't used that command before, but there are some good examples in the help menus.  Once you have the parent or master record created, with a new pass of the data then all is left is to create a new variable called GPA which is computed from all the separate course fields set up.  You use SPSS syntax to do the math.  Run a simple frequncy on GPA and you will have total number of respondents for each GPA.

George


--- On Wed, 10/15/08, Bill Oglesby <[hidden email]> wrote:

> From: Bill Oglesby <[hidden email]>
> Subject: Re: Determining GPA
> To: [hidden email]
> Date: Wednesday, October 15, 2008, 3:54 PM
> George,
>
> You are correct in that there is a separate record for each
> student, which consists of the course
> enrolled in and the final grade assigned (originally coded
> as As, Bs, Cs, etc.). By the way, not all of
> the courses are three credit hour classes, so the
> computation of GPA would be as follows: 4 pts for
> an A, 3 for a B...0 pts for an F multiplied by the number
> of credit hours per course, divided by the
> total number of hours taken that semester (this is probably
> information that I should have included
> originally).
>
> I guess, then, that there are two questions: 1) What is the
> least tedious way to compute GPAs, and 2)
> how, then, do you determine how many students earned a 2.0
> or higher?
>
> Bill
>
> =====================
> 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: Determining GPA

ViAnn Beadle
In reply to this post by Bill Oglesby
How about this?

First compute the product of hours and grade for each course taken.

Then using AGGREGATE, break on Student ID and semester and obtain the sum of
the product and the sum of hours.

You'll now have a record with student id, semester id, total product and
total hours. Then use compute again to calculate the GPA for that semester.
Now recode the GPA into a new variable with values 0 and 1 mapping to less
than 2.0 and greater equal 2.0. FREQUENCIES will give you the frequency of
the value across all students and all semesters. To get the breakdown by
semester, use CROSSTABS. Here's a simple example for students 1 and 2 in
semesters 1 and 2:

data list list / student grade hours semester.
begin data
1 4.0 3 1
1 2.0 4 1
1 1.0 3 1
1 0.0 4 2
1 3.0 3 2
1 4.0 3 2
1 1.0 3 2
2 4.0 4 1
2 3.0 3 1
2 1.0 3 1
2 3.0 3 1
2 4.0 3 2
2 3.0 4 2
2 1.0 3 2
2 2.0 4 2
end data.
compute gradprod = grade* hours.
AGGREGATE outfile=*
        /break = student semester
 /totprod=sum(gradprod)
 /tothour=sum(hours).
compute gpa=totprod/tothour.
recode gpa (2 thru highest=1)(else=0) into pass.
frequencies variables=pass.
crosstabs tables= pass by semester.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Bill Oglesby
Sent: Wednesday, October 15, 2008 1:55 PM
To: [hidden email]
Subject: Re: Determining GPA

George,

You are correct in that there is a separate record for each student, which
consists of the course
enrolled in and the final grade assigned (originally coded as As, Bs, Cs,
etc.). By the way, not all of
the courses are three credit hour classes, so the computation of GPA would
be as follows: 4 pts for
an A, 3 for a B...0 pts for an F multiplied by the number of credit hours
per course, divided by the
total number of hours taken that semester (this is probably information that
I should have included
originally).

I guess, then, that there are two questions: 1) What is the least tedious
way to compute GPAs, and 2)
how, then, do you determine how many students earned a 2.0 or higher?

Bill

=====================
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: Determining GPA

ViAnn Beadle
In reply to this post by Bill Oglesby
Oops, I forgot to specify cell percentages for the crosstabs so here's the
amended syntax:

crosstabs tables= pass by semester/ cells = all.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Bill Oglesby
Sent: Wednesday, October 15, 2008 1:55 PM
To: [hidden email]
Subject: Re: Determining GPA

George,

You are correct in that there is a separate record for each student, which
consists of the course
enrolled in and the final grade assigned (originally coded as As, Bs, Cs,
etc.). By the way, not all of
the courses are three credit hour classes, so the computation of GPA would
be as follows: 4 pts for
an A, 3 for a B...0 pts for an F multiplied by the number of credit hours
per course, divided by the
total number of hours taken that semester (this is probably information that
I should have included
originally).

I guess, then, that there are two questions: 1) What is the least tedious
way to compute GPAs, and 2)
how, then, do you determine how many students earned a 2.0 or higher?

Bill

=====================
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
|

Sizing a graph in GPL (and date labels)

Garry Gelade
In reply to this post by ViAnn Beadle
Dear List


I am having problems with getting SPSS (15.0) to size my GPL graphs.  I am
using the GRAPH: statement to do this as follows

BEGIN GPL
 GRAPH: begin(scale(30%, 80%))
 ...
 GRAPH end()
END GPL

Or

BEGIN GPL
 GRAPH: begin(scale(3,75in, 8in))
 ...
 GRAPH end()
END GPL

In both cases the graph is produced correctly except the size is always the
same whatever I put in the GRPAH statements or whether I include the GRAPH
statements or not.

Second problem: The x-axis is an SPSS date field. GPL gives me labels like
1-Aug-2007 o'clock. How can I get rid of the o'clock?

Thanks for any thoughts.

Garry Gelade

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