On Mon, 14 Aug 2000, ELSA NICKL wrote:
> Hello ,
> I am doing a research about the application of spatial analysis
> oriented to relate educational and geographical variables. I started
> with the
> analysis of autocorrelation for some variables using lattice data.
> I would like to know in which case it is neccesary to standarize the
> contiguity matrix by rows. I obtained Moran Indexes for both cases
> (with and
> without standarization) and in both the results were very different.
In general, standardisation gives more influence to the neighbours of
polygons with few neighbours, and less to the neighbours of polygons with
many neighbours. If all your polygons have the same number of neighbours,
the coefficients will be similar, while if some have very few neighbours
and others have very many, the results may differ - as in your case, I
guess. A reference to the key article: Tiefelsdorf M, Griffith D, Boots B
1998 A variance stabilizing coding scheme for spatial link matrices,
Environment and Planning A, 31, pp. 165-180.
Have a look at the distribution of numbers of neighbours of your polygons,
and see if there are some with many, and some with very few neighbours.
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Breiviksveien 40, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 93 93
and: Department of Geography and Regional Development, University of
Gdansk, al. Mar. J. Pilsudskiego 46, PL-81 378 Gdynia, Poland.
*To post a message to the list, send it to ai-geostats@...
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@...
with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!