C. Topi 1/4/2011

Like

1966 Views

I have a slightly unusual situation: I have a 3 levels hierarchy of geographical areas.

Level 1: Region
Level 2: City regions
Level 2 or 3: Local authorities

The region is made of several city regions + some local authorities
The city region is made of local authorities

I need to analyse the local authorities independently, then aggregate some of them at city region level, some others at region level (because some local authorities are not part of city regions, but directly of the region).

My doubt is: when I aggregate to region level can I put together city regions (which aggregate several local authorities) and some local authorities (which do not aggregate anything, the data are "native")?

I have reread the help but do not seem to understand how to do it.

Best

Tory Clark 1/4/2011
 

Like

1958 Views

Hi C. Topi,

You can create 2 levels of regions in LEAP using the tools under General\Regions and General\Region Groupings menus. However, it sounds like you will have some overlap between city regions and local authorities in addition to some local authorities at a 3rd level, which is not something that can be specified using those tools alone.

I think the best solution for you will be using a combination of the region tools and the LEAP tree to create a structure that includes all of the interactions that you are looking for.

Best,

Tory

C. Topi 1/5/2011
 

Like

1955 Views

I apologise Tory, but I do not understand what you are suggesting.

Do you have a similar example, or do you know anyone who had / has a similar set up I can contact? Thanks!

Best,

Tory Clark 1/5/2011
 

Like

1952 Views

Hi C. Topi,

Unfortunately I do not have an example on hand as this is a pretty unique situation. Perhaps another person on the forums has addressed this question in one of their datasets.

In the mean time, feel free to send me your dataset at leap@sei-us.org and one example of an interaction that you want to model and I can try to put together an example for you sometime this week.

Best,

Tory