Groups of close neighbouring provinces found by drawing the first order graphs are presented
in Table 2. The second column of this table shows the average of the distances between the
immediate neighbours in the group. This figure may be regarded as an index of resemblance
within the group, the smaller this figure the higher the resemblance between all members of the
group. It may also be compared with the bench mark values provided by d(+) (=4.462) and d(-)
(=1.288).
Table 2. Homogenous groups of provinces
Neighbouring groups of the first |
Degree of |
Closest |
Distance of the |
Qom, Tehran. |
4.031 |
Fars |
4.854* |
Isfahan, Semnan, Yazd. |
2.395 |
Mazandaran |
3.348 |
Booshehr, Fars, Gilan, Hormozgan, |
2.690 |
Kerman |
2.715 |
Char Mahal B., East Azarbayjan, |
2.714 |
Ardebil |
2.533 |
Ilam, Kerman, Kermanshah, Lorestan |
2.559 |
Mazandaran |
2.715 |
Ardebil, Kohkiloyeh, Kordestan, West |
2.801 |
East Azarbayjan |
2.533 |
Sistan and Baloochestan. |
* |
West Azarbayjan |
5.549* |
* Distance larger than the critical minimum distance of d(+).
Columns 3 and 4 of Table 2 show the closest province to the group and its distance (second
order links) from a member of the group. These provinces are found by searching for the
minimum value amongst the next (second) minimum distances for all members of the group.
The degree of heterogeneity within the first group is rather large, once again indicating that
Tehran is by far the most developed province. It is interesting to note that the distance of the
closest province to this group is above the minimum critical distance making the closest
province to the group heterogeneous to the group by our criteria. The last group in Table 2
includes only one province, Sistan and Baluchestan indicating that this most backward