You are here: Home >> Urban fieldwork >> Urban rebranding >> Stage 4

Stage 4: Data analysis

Clone town surveys

Do the following when you have completed the survey.

for each type of shop counted (e.g. food retailer, newsagent, stationery, etc)	give 5 points
for each independently owned shop	give 50 points
for each chain store	give 5 points

Add up the scores from all three groups, then divide this sum by the total number of shops surveyed. NEF interpreted the figure as follows: any score of 25 or below indicates a clone town, a score of 40 or above indicates a home town, and a score between 25 and 40 indicates a border town.

Worked example

For example, a survey in the centre of a market town found a total number of 473 shops.

there were 25 different types of shop

there were 321 independent shops

there were 152 chain stores

(25 x 5) + (321 x 50) + (152 x 5) = 16935

16935 ÷ 473 = 35.8

We can conclude that the town is a border town, neither clone town nor home town.

Statistical analysis of cloning

In this investigation is the link between the degree of cloning in a town and the town's socio-economic characteristics. Field data on the degree of cloning was collected from 10 towns within the historic county of Cheshire. Data on deprivation have been taken from the National Statistics website.

Town	Index of Cloning*	Index of Multiple Deprivation**
Northwich	17.9	21.8
West Kirby	35.4	14.0
Congleton	38.0	14.6
Nantwich	39.1	18.2
Neston	39.2	24.5
Sandbach	39.6	12.3
Middlewich	41.1	12.8
Bollington	41.4	8.6
Alsager	43.1	10.1
Frodsham	45.7	14.3

* Data taken from NEF Clone Towns report

** Index of Multiple Deprivation 2000 for wards 2000. Where a town is made up of more than one ward, a mean figure has been calculated.

The link can be tested using the Spearman's Rank Correlation Coefficient test (need more information on this test?).

A null hypothesis is set up: there is no significant relationship between Index of Cloning and Index of Multiple Deprivation.

Town	Index of Cloning		Index of Multiple Deprivation		d	d2
			value	rank	value	rank
Northwich	17.9	1	21.8	9	8	64
West Kirby	35.4	2	14.0	5	3	9
Congleton	38.0	3	14.6	7	4	16
Nantwich	39.1	4	18.2	8	4	16
Neston	39.2	5	24.5	10	5	25
Sandbach	39.6	6	12.3	3	3	9
Middlewich	41.1	7	12.8	4	3	9
Bollington	41.4	8	8.6	1	7	49
Alsager	43.1	9	10.1	2	7	49
Frodsham	45.7	10	14.3	6	4	16

Σd2 = 262

1 - (6 x 232) ÷ 990

1 - (- 1.406)

= - 0.41

For 10 pairs of measurements at p=0.05 the Critical Value = 0.648.

Therefore the null hypothesis is accepted. There is no significant relationship between Index of Cloning and Index of Multiple Deprivation.

Retail change

Statistical tests can be used to test whether there has been a change before and after urban rebranding has taken place. In this investigation, fieldwork was used to classify the buildings in a town centre into 12 categories. An old GOAD map was used to classify the buildings in the same town centre in 1984. The focus of the investigation is the extent to which building usage has changed since 1984.

Building use	1984 total	2008 total
Finance, admin services	16	17
Clothing, footwear	34	19
Supermarkets	5	4
Electrical, computer, phones	8	4
DIY, hardware, furniture	12	9
Commercial services	27	18
Entertainment, catering	25	36
Department and specialist stores	31	49
Grocers, convenience stores	5	11
Charity shops	2	4
Vacant premises	13	13
Residential and other	23	17
TOTAL	201	201

The difference can be tested using the chi-squared test (need more information on this test?).

A null hypothesis is set up: there is no significant difference between the 1984 data and the 2008 data.

This calculation uses the chi-squared test to test the Goodness of Fit between the two sets of data. Here the Expected Values are the 1984 data, and the Observed Values are the 2008 data.

Building use	1984 total (E)	2008 total (O)	O - E	(O - E)2	(O - E)2 ÷ E
Finance, admin services	16	17	1	1	0.06
Clothing, footwear	34	19	-15	225	6.62
Supermarkets	5	4	-1	1	0.20
Electrical, computer, phones	8	4	-4	16	2.00
DIY, hardware, furniture	12	9	-3	9	0.75
Commercial services	27	18	-9	81	3.00
Entertainment, catering	25	36	11	121	4.84
Department and specialist stores	31	49	-18	324	10.45
Grocers, convenience stores	5	11	6	36	7.20
Charity shops	2	4	2	4	2.00
Vacant premises	13	13	0	0	0.00
Residential and other	23	17	-6	36	1.57
TOTAL	201	201			38.69

For Degrees of Freedom = 11, at p = 0.05 the Critical Value = 19.68.

The null hypothesis is rejected. There is a significant difference between the 1984 and the 2008 data. The largest increases were in entertainment, grocers and department/specialist stores; the largest decreases were in clothing, commercial and electrical. The conclusions will relate these findings to the overall aims of the rebranding scheme.

GO TO NEXT STAGE: Review