Does a population hierarchy exist on the Ards Peninsula?

Have a little read: ... HYPOTHESIS 1 Does a population hierarchy exist on the Ards Peninsula? AIMS: The aim of this hypothesis is to see if the Rank Size Rule is true of the Ards Peninsula. In other words, to verify if there is a definite order to the spacing of settlements on the Ards Peninsula that could be said to be close to that stated by the Rank Size Rule. METHODOLOGY: In order to achieve the aims I have outlined above I used several different methods of collating and analysing the data. However, I must first describe how the data was collected. Much of the data was collected on the trip to the Ards Peninsula, although a good deal of the information regarding this hypothesis in particular was collected using census data. The data regarding population was gathered mostly by driving very slowly through the settlements and counting the houses as we went, then multiplying the number of houses by the Ards peninsula average population per house which is approx. 2.5. For many of the larger settlements we used 1991 census data. Using this data I will make a table showing the population of each of the twenty-five settlements we visited and their rank. The settlements will be ranked from one to twenty-five, with one having the largest population. Having done this I will then plot a line graph of population against population rank. If I draw on another line showing the theoretical result if the Ards Peninsula conformed exactly to Rank Size Rule, I will be able to compare these two lines and should be able to see whether or not my hypothesis is correct, and if so to what extent. For the theoretical basis regarding Rank Size Rule see following page. PROBLEMS: The main problem we faced regarded calculating population of the smaller settlements. The method of counting the houses and then multiplying by the average number of people per house which theoretically should be living there, is very inaccurate. Not only is it extremely unlikely that the total would level itself off, but if the number of houses was miscounted by even one or two it could make a considerable difference to our data on the population of the smaller settlements. Also, the census data we used was nine years old and so slightly out of date. While the population of the Ards Peninsula changes only marginally every ten years between censi, it is worth noting. However, I do not believe that these points invalidate any of my hypotheses, as the figures are likely to even themselves out. In the case of the smaller settlements on the Ards Peninsula, the figures are so small to begin with. For example the Nun's Quarter which has a count of 12, therefore, if any mistake were made the deficit would be so small it would be negligible, and highly unlikely to make any difference on a graph. HYPOTHESIS 1 THEORETICAL BASIS * The Rank Size Rule In any large area of land there are usually few large cities, several medium sized towns and then many small towns or villages. This applies to virtually all regions of the world, regardless of its stage of economic development. The settlements in a certain region can be ranked from largest to smallest according to the size of their populations, with the largest ranked number one. Using a line graph and plotting the settlement's rank against its population we see that there is generally a direct correlation between the two. This was first noticed in 1913, by F. Auerbach. However, it was not until 1949 that G.K.Ziph devised the Rank Size Rule based on the findings of Auerbach. This rule tries to express the relationship between a settlement's rank and its population in more precise mathematical terms. "If all the urban settlements in an area are ranked in descending order of population, the population of the nth town will be that of the largest town." In simpler terms this basically means that if for example the population of the largest settlement in an area is 4 million then the population of the settlement ranked second (n=2), then using the Rank Size Rule ( ) the population of the settlement ranked second should be 2 million, one half the size of the largest settlement. Another example would be for the fourth largest settlement. Using the Rank Size Rule ( ) where n=4 we see that the population of the fourth largest settlement, according to this rule would be 1 million, one quarter the population of the largest settlement. The graph below shows the theoretical result of the Rank Size Rule when a settlement's rank is plotted against its population. HYPOTHESIS 1 TABLE SETTLEMENT POP. (ACTUAL) POP. (EXPECTED) POP. RANK Portaferry 2324 2324 1 Millilse 1531 1162 2 Portavogie 1482 775 3 Ballywalter 1072 581 4 Kirkubbin 987 465 5 Greyabbey 697 387 6 Cloughy 647 332 7 Carrowdore 444 291 8 Ballyhalbert 332 258 9 Loughreice 166 232 10 Rubane 131 211 11 Ballygalget 98 194 12 Ballycrambeg 81 179 13 Glastry 76 166 14 Ballyboley 59 155 15 Ebenezer 58 145 16 Cardy 56 137 17 Ballyeasborough 54 129 18 Peeltown 52 122 19 Ballyigan 44 116 20 Ardkeen 32 111 21 Cookstown 26 106 22 Ballyblack 24 101 23 Nuns Bridge 20 97 24 Nuns Quarter 12 93 25 With the information in this table I can plot a graph to show the relationship between population, rank and theoretical population. HYPOTHESIS 1 INTERPRETATION The graph (1) showing population against population rank shows us that there is not a very strong link between Rank Size Rule and the settlements on the Ards Peninsula. I believe that the discrepancies between the theoretical result and the actual result are too large to say that there is a link. As I expected, the actual results follow the same curve shape as the theoretical results. This was to be expected because if virtually any set of figures are ranked in descending order and plotted on a graph then it is inevitable that a shape similar to that of the one we have achieved in this graph will present its self. Another thing that has to be said about this graph is that the results from 1 to 9 are above the theoretical results, whereas the results ranked 10 to 25 are below the theoretical results.The reasons for this are uncertain, but one answer that is definitely worth noting is the fact that the top nine settlements are at coastal locations with the exception of Carrowdore. Coastal locations provide scenery for the tourist industry and ports for the fishing industry, to name only a few of the advantages. Coastal locations attract more people to go and live there and encourage industries and services to grow. This would explain why many of the larger settlements have coastal locations. This idea would seem to be confirmed by the fact that the majority of the smaller settlements all have inland locations. As well as showing us that there is no outstanding correlation between the Rank Size Rule and the settlements on the Ards Peninsula the graph has other uses. The following graph shows the settlements' population against their population rank just as in graph 1. However this time I have used the graph to separate the settlements into a hierarchy. We can now see more clearly the settlements grouped together according to the size of their populations and what significance this has to my main hypothesis. This graph shows a step by step arrangement, which the settlements follow. It shows that the top ten settlements by population are spaced out and have two or three in each level whereas the settlements ranked eleven to twenty five are close together on the graph. The reasons for this are again due to the fact that the larger settlements are at coastal locations. Although it is also worth noting that the smaller settlements populations are more similar due to the fact that more people go to live in larger settlements, as there are usually more services there. HYPOTHESIS 1 CONCLUSION From this sub-hypothesis we have discovered that the connection between the Rank Size Rule and the settlements on the Ards Peninsula which we visited is a very loose one. There was of course a certain