Nelson through his classification removed the shortcomings of the earlier classifications by using a stated procedure that could be objectively checked by other workers. His paper ‘A Service Classification of American Cities’ was published in the journal Geography in 1955.
He decided to base his method of classification entirely upon major industry groups as listed in the 1950 Census of Population for standard metropolitan areas, urbanized areas and urban places of 10,000 or more population. He omitted the little significance groups like agriculture and construction, and finally, arrived at the nine activity groups (manufacturing; retail; professional services; wholesale; personal service; public administration; transport and communication; finance, insurance, real estate and mining).
The problem of city specialization, and also the degree of specialization above the
average was solved by giving margins of different degree to different size classes. He did find a definite tendency for the percentages employed in some activities vary with city size. The question – ‘When is a city specialized?’ was solved by using a statistical technique – the Standard Deviation (SD).
He reasoned that that SD was the simplest and most widely understood of all statistical measures of variation, and that the degree of variation could be compared by use of SD even if, in some cases, we are dealing with large numbers , as in the manufacturing, and in others with small numbers, as in mining as wholesale trade.

A city can be specialized in more than one activity and to varying degrees. Thus he showed for each city all activities that qualified for plus 1, plus 2, or plus 3 SDs above the
mean. According to Nelson’s Classification, a city can be specialized in more than one activity and to varying degrees.
Suppose, any city which is classified as Pf 2F, it means that it has 22.87 or more but
less than 28.76 per cent of its labour-force employed in professional service and 4.44 or more but less than 5.69 per cent employed in finance, insurance and real estate. In short, his table indicates, the number of SDs shows the degree to which the urban centre stands out for the activity in question. A city which does not fall even under 1 SD, average in any activity appears as diversified D, in Nelson’s classification.
With the help of the Standard Deviation (SD) method, Nelson was able to show the degree of dominance of other industries in comparison to the most dominant industry. To clarify, the degree of specialization is measured in multiples of standard deviation. We will learn to calculate the degree of specialization using following steps.
Further, we have to construct a matrix or table showing the national average, standard deviation (SD) and multiples of standard deviation.
Step I
Firstly, collect the industry wise data of workers for all the cities of the country. Secondly, calculate the proportion of workers in each industry by using a simple percentage formula as follows.
![]()
The above formula shows the calculation for arriving at the proportion of workers in the Manufacturing (Mf) sector. Thirdly, we will calculate the proportion of workers in all nine sectors. Finally, we will get the percentage of workers in each industry. Please note that this will be the national average or mean for each of the industries.
Step II
In this step, we have to calculate the percentage of workers in each industry for different cities of the country. The formula for calculating this is as follows.
![]()
The above formula shows the proportion of workers in the Manufacturing (Mf) sector for Mumbai. Similarly, calculate the proportion of workers in the manufacturing sector for all the cities.
Now, we have to repeat the same process for all the industries. Consequently, we will get the proportion of workers in each of the nine industries for all the cities. You will get data as shown below.
Earlier in Step I, we have already calculated national average or mean workers in each industry.
Step III
In this step, we will calculate the standard deviation of each industry across the cities by using the following formulae.

- In the above formulae X-Mean will be calculated for each city. Then all the values will be summed and divided by the total number of cities (n). By using this formula, we will get the SD for the manufacturing sector (Mf). Similarly, calculate the standard deviation of all the industries
Read More at THIS Source
Pingback: Functional Classification of Cities | Rashid's Blog: An Educational Portal
It is amazing.. It helps me a lot.. Thank you… Pls post the mckenzie’s classification of cities.. Please I need it..
LikeLike