The Scottish Government

« Previous | Contents | Next »

Listen

Living In Scotland: An Urban-Rural Analysis Of The Scottish Household Survey

CHAPTER ONE: INTRODUCTION

BACKGROUND

It is often assumed that there are significant differences between urban and rural life. Rurality is associated with a wide range of social and economic differences, and often assumed to involve major disadvantages. This report presents the results of an analysis of urban-rural differences in Scotland in relation to relative opportunities for employment, income levels, quality of health, and the local neighbourhood, at the end of the twentieth century, as revealed by the Scottish Household Survey (SHS).

THE SCOTTISH HOUSEHOLD SURVEY

The SHS was established in 1999 and is a major new source of information on many aspects of Scottish life. This survey involves interviews in around 15,000 households each year. Each interview gathers information about the household from the household member who earns the highest income, or his/her spouse. In addition, one adult member of the household (aged 16 or over) is selected at random and asked further questions about their employment and health, and their views on the local neighbourhood. The SHS also gathers information on topics such as transport, but these are not considered in detail in this report.

The need to monitor rural-urban change was carefully considered in the design of the sampling scheme for the SHS. Attention was also given to classifying urban and rural areas in a way that might prove useful in analysis. An eight-fold classification was developed based on the two key factors of settlement size, and (for smaller settlements), distance (drive-time) from a settlement of 10,000 or more people. This classification is shown in Table 1.1, and a map is provided in Appendix 1.

Table 1.1 Rural/Urban categories used in the Scottish Household Survey

In the SHS relatively few people are interviewed in some of the more remote areas. The consequence is that samples of some sub-groups in these areas are too small for statistically significant differences to be identified. For this reason categories 4 and 5 are often combined, as are categories 7 and 8, to form a six-fold classification. The present report adopts this simplified classification. In this scheme 'remote' areas are those 30 minutes or more drivetime from a settlement of 10,000 or more people.

For some analyses in the present report, even the six-fold classification yields small numbers of people in the remote small towns and remote rural areas. In such cases the classification is further simplified, resulting in just two categories: rural and urban areas, as illustrated in Table 1.2 below.

Table 1.2 Urban/ rural breakdowns used in this report

Data for the Scottish Household Survey are designed to be representative on a two-year cycle, and data from any single year are not representative of all Scottish local authority areas. SHS data is now available for 1999, 2000 and 2001. In this report the majority of analysis is based on a combined data set from all those interviewed in 1999 and 2000. Data from 2001 are not included. This sample provides us a total sample of 30,217 households, of which 6,359 live in rural areas and 3,947 in remote areas. Information for the randomly selected adult is derived from 28,330 interviews, of which 5,945 live in rural areas and 3,727 in remote areas ¹. Some data included in Chapter 6 is drawn from questions introduced into the SHS in January 2000, and so contain only half the sample population available in all other tables. These cases are indicated in the text.

The results of this analysis have been compared to those detailed in other SHS publications, particularly Scottish Executive/MORI/NFO-System 3 report Scottish Household Survey Bulletin #6 (2001) which focused on the characteristics of urban and rural areas (http://www.scotland.gov.uk/shs/docs/shsb-00.asp). As a means of establishing the context, this report reiterates some of the basic description of rural areas ², but presents more in-depth analysis in relation to the key issues of employment, income and health.

The SHS data set is weighted in a number of different ways to ensure representativeness. The two weighting devices of relevance in this analysis are the household weight which corrects for disproportionate sampling and non-response, and the individual/random adult weight which allows for unequal probability of selection in multi-adult households and differential non-response to that section of the interview.

More information about the Scottish Household Survey including questionnaires and response rates can be obtained from Scotland's People: Results from the 2001 Scottish Household Survey, Volume 6: Technical Report. (Nov. 2002).

Before going on to look at the topics on which this report is focused, Chapter 2 presents some background information showing the characteristics of people living in the various types of area, highlighting any notable differences between rural and urban areas. Chapter 3 then considers Employment, Chapter 4 Income, Chapter 5 Health, and Chapter 6 the Neighbourhood. At the end of each chapter key results and themes are summarised. The Appendices provide additional analysis looking at subgroups of the population.

Some of the data presented in this report was previously published in SHS Bulletin 6

Notes on the Standard Table Format used in this report:

• The source for all the tables in this report is the Scottish Household Survey 1999/2000. In order to keep the presentation as clear as possible individual tables do not have a note of this source.
• Most tables have a standard layout, with six columns containing data for the six types of area (large urban areas, other urban areas, accessible small towns, etc). In a small number of cases the six columns are aggregated to rural and urban subtotals. The final column in both these types of table gives the data for the full (Scottish) SHS sample.
• All data under the heading '%' is given in the form of a percentage of the total sample for the area indicated in the heading.
• All percentages shown in tables are based on weighted data, however unweighted bases are shown in the bottom row of the tables.
• '0' in a table signifies fewer than 0.5% of cases, '-' signifies no cases at all.
• Due to rounding the tables may not always add up to 100%.
• With one or two exceptions the data in all the tables has been tested using a Chi-square test to assess the statistical significance of the difference between each column (area type) from the total of the remaining columns within the Scotland sample. Where the data for a column is significantly different from that of the rest of Scotland (at the 95% confidence level), a ' *' is shown at the head of the column. Where the difference is not significant at 95%, the column is headed with a ' <'.Where the sample size is particularly small, so that one or more cells has an "expected" count of less than 5 the column is headed with a " #".

« Previous | Contents | Next »