SPSSX Discussion

constructing an index of net income difference

Classic

List

Threaded

5 messages Options

Greg

constructing an index of net income difference

Hi everyone,

This might be a simple question, but I'd appreciate any suggestions.

I want to create an index of net income difference between central-cities and suburbs. My intention is to create the index in a way in which positive values mean that suburb incomes are higher than those of central-cities, whereas negative values mean that central-cities income is higher than suburbs.

In other words, net income difference: +: suburb incomes higher than central-cities
-: central-cities incomes lower than suburbs

Any input would be greatly appreciated!

Thanks,
Greg

Art Kendall

Re: constructing an index of net income difference

assuming you have info about the MSA that the suburbs and central city
are in (or other info that identifies which suburbs go with which
central city).
Something like this untested syntax will tell you for each MSA how much
larger/same/lower suburban income is than central city income.

if centralcity centralincome = income.
if suburb suburbincome = income.
Sort cases by MSA.
aggregate outfile= * /break=msa
/centralincome = mean(centralincome)
/suburbincome = mean(suburbincome).
numeric netincomediff (dollar10).
compute netincomediff = suburbincome-centralincome

Art Kendall
Social Research Consultants

grigoris wrote:

> Hi everyone,
>
> This might be a simple question, but I'd appreciate any suggestions.
>
> I want to create an index of net income difference between central-cities
> and suburbs. My intention is to create the index in a way in which positive
> values mean that suburb incomes are higher than those of central-cities,
> whereas negative values mean that central-cities income is higher than
> suburbs.
>
> In other words, net income difference: +: suburb incomes higher than
> central-cities
> -: central-cities
> incomes lower than suburbs
>
> Any input would be greatly appreciated!
>
> Thanks,
> Greg
> --
> View this message in context: http://www.nabble.com/constructing-an-index-of-net-income-difference-tp19641568p19641568.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

Art Kendall
Social Research Consultants

Greg

Re: constructing an index of net income difference

In reply to this post by Greg

Thank you for the input. However, I already did what you mentioned. However, and this is where I'm not sure how to figure it out yet, but would like to construct some sort of index in order to be able to tell which suburban areas had higher values than cities, aside from just reporting their difference in scale format.

For example, a positive & higher value would indicate that suburbs have higher income than cities. On the other hand, a negative & lower value would indicate that cities have higher incomes than suburbs.

Thanks,
Greg

grigoris wrote

Hi everyone,

This might be a simple question, but I'd appreciate any suggestions.

I want to create an index of net income difference between central-cities and suburbs. My intention is to create the index in a way in which positive values mean that suburb incomes are higher than those of central-cities, whereas negative values mean that central-cities income is higher than suburbs.

In other words, net income difference: +: suburb incomes higher than central-cities
-: central-cities incomes lower than suburbs

Any input would be greatly appreciated!

Thanks,
Greg

Art Kendall

Re: constructing an index of net income difference

do you mean you want separate indices for each suburban area around a
central city?
is something like this untested syntax what you are looking for?
*create new file with data for each area.
Sort cases by MSA area .
aggregate outfile= * /break= msa area
/areaincome = mean(income)
/sub_or_central = first(sub_or_central) .
*for every area in an msa create new variable with center city income.
* centralincome variable will be missing for suburban areas.
if sub_or_central eq 'central' centralincome=income.
aggregate outfile=* mode = addvariables /break=msa
/ centralincome = mean(centralincome)
/sub_or_central = first(sub_or_central) .
numeric netincomediff (dollar10).
*calculate the difference variable.
do if sub_or_central eq 'suburb'.
compute netincomediff = -centralincome-areaincome .
else.
compute netincomediff=-999999999.
end if.
missing values netincomediff(-999999999).
numeric diff_group(f1).
recode netincomediff (lo thru -.0001 = 1)(-.0001 thru .0001 = 2)(.0001
thru high=3)(missing=4)(else=5).
value labels diff_group
1 'suburbs lower'
2 'about the same'
3 'suburbs higher'
4 'not a suburb'
5 'oops should not have this'.

Art Kendall
Social Research Consultants

grigoris wrote:

> Thank you for the input. However, I already did what you mentioned. However,
> and this is where I'm not sure how to figure it out yet, but would like to
> construct some sort of index in order to be able to tell which suburban
> areas had higher values than cities, aside from just reporting their
> difference in scale format.
>
> For example, a positive & higher value would indicate that suburbs have
> higher income than cities. On the other hand, a negative & lower value would
> indicate that cities have higher incomes than suburbs.
>
> Thanks,
> Greg
>
>
> grigoris wrote:
>
>> Hi everyone,
>>
>> This might be a simple question, but I'd appreciate any suggestions.
>>
>> I want to create an index of net income difference between central-cities
>> and suburbs. My intention is to create the index in a way in which
>> positive values mean that suburb incomes are higher than those of
>> central-cities, whereas negative values mean that central-cities income is
>> higher than suburbs.
>>
>> In other words, net income difference: +: suburb incomes higher than
>> central-cities
>> -: central-cities
>> incomes lower than suburbs
>>
>> Any input would be greatly appreciated!
>>
>> Thanks,
>> Greg
>>
>>
>
> --
> View this message in context: http://www.nabble.com/constructing-an-index-of-net-income-difference-tp19641568p19658514.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
>
>
>

Art Kendall
Social Research Consultants

David Greenberg

Re: constructing an index of net income difference

In reply to this post by Greg

Take the ratio, and then take the natural logarithm of that ratio. It will behave as you want. David Greenberg, Sociology Department, New York University

----- Original Message -----
From: grigoris <[hidden email]>
Date: Wednesday, September 24, 2008 11:37 pm
Subject: Re: constructing an index of net income difference
To: [hidden email]

> Thank you for the input. However, I already did what you mentioned. However,
> and this is where I'm not sure how to figure it out yet, but would
> like to
> construct some sort of index in order to be able to tell which suburban
> areas had higher values than cities, aside from just reporting their
> difference in scale format.
>
> For example, a positive & higher value would indicate that suburbs have
> higher income than cities. On the other hand, a negative & lower value
> would
> indicate that cities have higher incomes than suburbs.
>
> Thanks,
> Greg
>
>
> grigoris wrote:
> >
> > Hi everyone,
> >
> > This might be a simple question, but I'd appreciate any suggestions.
> >
> > I want to create an index of net income difference between central-cities
> > and suburbs. My intention is to create the index in a way in which
> > positive values mean that suburb incomes are higher than those of
> > central-cities, whereas negative values mean that central-cities
> income is
> > higher than suburbs.
> >
> > In other words, net income difference: +: suburb incomes higher than
> > central-cities
> > -: central-cities
> > incomes lower than suburbs
> >
> > Any input would be greatly appreciated!
> >
> > Thanks,
> > Greg
> >
>
> --
> View this message in context: http://www.nabble.com/constructing-an-index-of-net-income-difference-tp19641568p19658514.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