Life in Two Enclave Areas. Chapter 1: The sample and population in the areas surveyed

The examination of Gobnascale and the Fountain began by attempting to establish basic population figures for each area. At the time this study began, enumeration district data, where subsets of census data for 400 households could be obtained, was not yet available, - but became available within the first year of the study. However, since the Fountain fell below the size of an enumeration district, and since enumeration district data was only available on the 1991 census, that source of data was unsuitable. In order to examine migration in and out of the two areas over a period of time, it is necessary to be able to examine and compare data from more than one year, in order to examine shifts in population over time. Small area statistics from the 1971, 1981 and 1991 Census of Population for Northern Ireland were used for this purpose. In order to establish the boundaries of the two areas as accurately as possible, local community workers and residents were consulted and a grid reference which corresponded as closely as possible with the area was agreed. Data was obtained for each of these two areas for the three census years. Further details of this process, together with the grid references used are contained in Appendix 1.

The following tables summarise the findings on total population, religion and migration for each of the two areas:

From this exercise, we established a figure for the total population for each area, which also reflected the impressions of the total population figure held by local community organisations in both areas. The establishment of this figure allowed decisions to be made about the sample size for the survey. Details of the selection of the sample, together with data on the confidence limits and sampling procedures, are contained in Appendix 2.

This baseline data also afforded a perspective, - the limitations of the 1991 Census notwithstanding,- on migration in the two areas over the twenty year period examined. From the baseline data, it appeared important to examine in the survey the stability of the population in each area, and their intention to move or stay in the area. These issues were included in the questionnaire (see Appendix 3) and the findings are discussed in Section 1: Migration and the Stability of the Populations.

The sample selected in the survey in both areas were also asked about their length of residence, religion, identity, gender, age, housing tenure, housing type, number in household, their main activity in the previous week, annual income and source of income, and experience of unemployment in the last 5 years.

How long have lived where you now live?

Responses to question 1 are shown in the table below and in the following diagram.

Differences in responses between the two areas were significant at the .0 level. An examination of the data reveals that a "bulge" in the Gobnascale figures occurs in the over 10 and under 20 years and in the over 20 but under 40 year category, whereas the Fountain figures show no such trend. This "bulge" can be explained by the settlement pattern in Gobnascale, which was largely influenced by the building of a new housing area in the early 1970's, thus explaining the inflated numbers of people who have lived in the area since that housing became available. Conversely, 11% of Fountain respondents fell into the "60 years or more" category, whereas no Gobnascale respondents fell into this category. It would appear that Gobnascale has a relatively sudden pattern of settlement, with an influx of population in the 1970's, whereas no such sharp influx seems to have occurred in the Fountain.

We know from qualitative data that there was an influx of Protestants from other parts of the city into the Fountain in the early 1970, and this could explain the slight rise in the percentage of respondents in the 10-20 and 20-40 year categories. There is a fall in numbers, however, in the 5-10 year category in the Fountain, which would tend to suggest a marked loss of population in the period 1984-1989. Conversely, 6% of Fountain respondents, (compared with only 1% of Gobnascale respondents) indicated that they had lived in the area for less than 6 months. This would suggest a recent influx of new residents. A Housing Executive survey in the Fountain, also conducted in February 1996, suggests that there may also be some movement of residents within the area, and point out that the highest housing turnover is in the flats and maisonettes. This is also true to a much lesser extent in Gobnascale, where flats have been used as single person housing, although it takes place within a much more stable population.)

Nonetheless, the figures for the Fountain suggest a high level of new residents, whether they move into the area from outside, or move within the area. However, we know from qualitative data, that this may not necessarily be a positive development for an area with a population decline, since some of that influx may be accounted for by the unofficial use of some housing units as emergency or short-term housing, as housing for single people, often those being discharged from residential facilities, or, subversively, as giro-drops. Over 70% of these new residents - those living in the area for less than a year- are male, and qualitative evidence suggests that many of these males have alcohol problems. Certainly, from the qualitative data, we know that the influx of newcomers is a matter of some anxiety for Fountain residents, who fear that the culture of the area is endangered by an influx of people without previous connection to the area. Some common anxiety about the "high turnover of newcomers to the area" and about the "area [being] used as a 'dumping ground' for people with problems" surfaced in the interviews, and were included in the questionnaire (see question 13), which is discussed in section 4.

Graph

The next characteristic of the population and of the sample which was examined was the religious make up of the population, and of the sample. Tables 1 and 2 show the religious breakdown of the two areas over a twenty year period. Table 4 shows the comparative religious breakdown of the sample in both areas.

Predictably, the difference between the two areas is significant at the .0 level, and the religious composition in the two areas is markedly different. In Gobnascale 91% of the population is Catholic, with a further 6% indicating that they had either no religion or no denomination. Two (1%) Presbyterians appeared in the Gobnascale sample, and were the only other denomination present in the area.

In the Fountain, the situation was only slightly less religiously homogenous. Church of Ireland, Presbyterian and Methodists accounted for 73% of the Fountain sample, with a further 15% claiming no religion or no denomination. The sample reflected the trends in religious composition which were observable from our interrogation of small area census data for the two areas.

GRAPH

Question 25 of the questionnaire asked about the community identity of respondents. The results are shown in Table 5, and in the following diagram. Again, the responses are significantly different between the two communities at the .0 level.

GRAPH

Table 5 and the diagram above illustrate the slight difference in polarisation between the two samples, with the Fountain sample containing 3% of respondents who identify with the Catholic community, whereas the Gobnascale sample contained only 1% of respondents who identified with the Protestant community.

Question 29 established the gender of respondents. The results are shown in Table 6 and the following diagram. In the Gobnascale sample, only 33% of respondents were male, whereas 46% of Fountain respondents were male. The gender difference between the two areas is significant at the .01 level.

When we compare the gender make-up of the sample to data on the gender make-up of the population in both areas from the extracted census data, we find that the sample is only 33% male in Gobnascale, compared with 48% male in the population according to the census data. The Fountain sample matches the proportions of males in the census of population data for the Fountain (46% male)and the 52% female in the Fountain sample matches the population data also, if we allow the 2% no response rate in the Fountain to be female.

It would appear that our Gobnascale sample contains a disproportionate representation of females, and in future analysis, the data could be weighted to take this into account. This skew in favour of females possibly arose in the self-completion nature of the questionnaire. We were aware in carrying out the fieldwork, that, although we specified which family member we wished to complete the questionnaire, the internal division of labour within households meant that another family member- now apparently female - sometimes completed the questionnaire on behalf of the selected person. This could account for the skew.

GRAPH

We will return to the issue of gender when we look at the next set of data on age.

We established the age groupings of our sample in question 30 of the questionnaire. The results are shown in Table 7 and in the following diagram. There were no significant differences between the two samples in relation to age. However, examination of the age distribution shows that 31% of the sample
Table 7: Age

GRAPH

is over the age of sixty, whereas only 17% of the Gobnascale sample falls into the same category. However, this does not entirely reflect the age data which we obtained from extracts of the census for the two areas. These can be summarised as follows:

From an examination of the census data, it appears that 25% of the Fountain population was over the age of 60 in 1991, whereas only 7% of Gobnascale residents fell into this category. A further gender breakdown of these data reveals that 82 residents or 17.5% of the Fountain population are females over the age of 60, while only 63 (4.8%) of a total population of 1312 in Gobnascale fell into this category.

Conversely, at the lower end of the age range, the largest single proportion of the population in Gobnascale (34%) was under the age of 16. The equivalent proportion in the Fountain was 22%. Thus it appears that, whilst our sample may not reflect it, the Fountain contains a larger proportion of older people, whilst Gobnascale contains a larger proportion of young people.

Questions 31 and 32 in the questionnaire establish the tenure and nature of housing among respondents from each of the two areas. The following two tables and diagrams illustrate the responses to these questions.

GRAPH

Predictably, the vast majority (80%) of Gobnascale respondents live in public housing, as do a smaller majority (69%) of Fountain respondents. The differences between the two areas were significant at the .0001 level, due to a larger privately rented sector (10%) in the Fountain. There were similar levels of owner occupation in each area (Fountain 17% and Gobnascale 18%) from which the impact of the sale of public housing can be notices, especially in Gobnascale, which was an area predominated by public housing until relatively recently. The Fountain, on the other had, always contained a proportion of owner occupiers, albeit many of them in sub-standard housing. Therefore, we conclude that significant difference exist between the two samples in terms of housing tenure.

Again, in terms accommodation type, significant differences (at the .0 level) exist between the two areas. Terraced housing is the predominant type of accommodation in both the Fountain (48%) and the Gobnascale sample (67%), although self-contained purpose built flats or maisonettes (40%) account for a substantial amount of the housing in the Fountain. Semi-detached houses or bungalows accounted for 17% of Gobnascale housing in the sample, whilst it accounted for only 4% in the Fountain, reflecting the more suburban setting of Gobnascale. Similarly, 8% of Gobnascale housing was detached houses or bungalows, which accounted for only 1% of Fountain housing. The environment in which people live in these two areas is significantly different, both in terms of housing mix and housing tenure. Table 10 and the following diagram show the results of question 32 in the questionnaire.

GRAPH

Two questions in the questionnaire elicited information about household size. Question 33 asked for the number of people over the age of 18 in the household, and question 34 asked for the number under 18. Tables 11-13 show the results for these questions. Table 11 shows the number of persons over 18 in households in both areas. There were no significant differences between the two areas on this variable.

Table 12 shows the numbers of children under 18 in households in both areas. A significant difference at the .00005 level emerged on this question. 66% of households in the Fountain and only 38% of households in Gobnascale contained no children under 18. In all other categories there was a lower proportion of Fountain households, indicating an overall lower population of children in the Fountain.

GRAPH

When the data from questions 33 and 34 was aggregated, it was possible to compile data about household size in the two areas. Table 13, and the diagram following, shows the results of this exercise. We can then compare this with the data extracted from the 1991 census, shown in Table 14. No significant difference was found between the two areas in the number of people over 18 in the household. However, significant differences exist between the two areas at the .00005 level on the number of people under the age of 18 in households.

GRAPH

There is some variation between the respondents in our sample and the extracts from census data. According to the census data 47% of households in the Fountain and 22% of Gobnascale households are single person households, whereas only 41% of Fountain households and 14% of Gobnascale households in our sample fell into this category. Differences also occurred between our sample and the census data in the proportion of two and three person households in each area, and in the numbers of households with seven or more people in each area. Our sample contained proportionally less of these households than the census data would indicate is present in both areas. However, the degree of difference between the two areas seems to be similar between our sample and the census data. It is also worth noting that some changes may have occurred to the population structure in the five years since the census material was collected, which may explain some of the difference between our sample and the census data.

Question 35 asked respondents to indicate from a list of choices what they were doing in the previous week. Table 15 and the two diagrams following show their responses. The responses from the two areas were significantly different at the .002 level.

There are differences between the proportion in the two areas who are registered unemployed. 13% of Gobnascale respondents and 17% of Fountain respondents selected this category. Conversely, 20% of Fountain respondents and only 16% of Gobnascale respondents said that they were in full-time work, and 7% of Fountain respondents and 5% of Gobnascale respondents said that they were in part-time work. When we compare this with data extracted from the 1991 census for the two areas, (see tables 16-18) we find that the census data tends to confirm a higher economic activity level in the Fountain,

2 GRAPHS

Gobnascale. A much larger proportion of Gobnascale respondents (21%) described their activity in the previous week as "looking after the home": only 7% of Fountain respondents selected this option. Finally, a higher proportion of Gobnascale respondents (5%) reported "full or part time education" as their activity in the previous week, compared with only 1% of Fountain residents.

In the survey, we asked respondents, in questions 38 and 39 about their experiences of unemployment. Tables 19 and 20 and the accompanying diagrams show their responses. Our survey showed no significant differences in the employment and unemployment rates in the sample. However, again, caution is necessary because of the high no response rate to both questions. Between 50%-60% of respondents made no reply on these questions. However, on both questions, employment and unemployment in the last five years, and the number of months unemployed in the last five years, there were no significant differences between the two areas. This tends to contradict the data extracted from the 1991 census, which indicates a difference in favour of the Fountain between the two areas in employment and unemployment rates. The low response rate in this survey makes it difficult to effectively challenge the census data. Although it is possible that the economic changes which have taken place since the 1991 census was taken have impacted on the employment rates in both areas, it is impossible to be conclusive here, for the reasons outlined.

GRAPH

It is perhaps noteworthy, that in table 20, of those who did reply to the question, the biggest cohort were those who had been employed for all of the previous five years. (19% in Gobnascale and 12% in the Fountain)

Generally, from both the census data, and to some extent from the survey data, the evidence suggests that there is an overall lower economic activity level in Gobnascale in comparison to the Fountain, and according to census data, this is accompanied by higher unemployment levels, and lower employment rates in Gobnascale. The survey data were more difficult to interpret because of the low response rate on some questions.

We then went on to examine income levels in both areas. However, the data on income must be treated cautiously, due to the very high non-response level in both areas. Question 37 on the questionnaire asked respondents to indicate their income level on a range of income bands. Responses from both areas are shown in table 21 and in the following diagram. If the "no response" and "don't know" categories are aggregated, then 70% of respondents in Gobnascale and 63% in the Fountain did not make a response to this question. For those who did reply, there was a peak of responses in both areas in the £2,000-£4,999 response category and in the £5,000- £9,999 category, with less than 10% of respondents in both communities claiming to earn more than £9,999. There were no significant differences between the two ares in terms of income, although, as was stated earlier, these findings must be treated with caution, due to the poor response rate. However, in spite of differences in economic activity and employment rates, that there should be no significant difference in income levels between the two areas, would suggest that those in employment are in low wage employment, thus preventing increased employment from altering income levels in the Fountain area.

GRAPH

We also asked respondents about their sources of income in question 40, and their responses are shown in table 22. Again, there was a very poor response rate to this question, with between 50% and 84% of respondent making no response on questions. However, from the responses we did collect, it is perhaps noteworthy that a significant difference between the two areas was observed on all of the options listed. Again, a higher proportion (22%) of Fountain respondents indicated that they had income from employment (the Gobnascale rate was 13%). 40% of Gobnascale respondents and only 28% of Fountain respondents said that they received income support; similarly, more Gobnascale respondents than Fountain respondents were in receipt of child benefit (Gobnascale 31%: Fountain 14%); one parent benefit (Gobnascale 12%: Fountain 4%); sickness benefit (Gobnascale 7%: Fountain 4%); old age pension(Gobnascale 33 %: Fountain 26%). Fountain rates were higher for unemployment benefit (Gobnascale 6 %: Fountain 7 %); family credit (Gobnascale 3%: Fountain 4 %):;housing benefit (Gobnascale 11%: Fountain 15%); invalidity benefit (Gobnascale10%: Fountain 12%); pension from employment (Gobnascale 2%: Fountain 9%); and income from other sources (Gobnascale 5 %: Fountain 6%).

Generally, as with the other economic data, it is difficult to be conclusive because of the poor response rates. One would expect economic differences to exist between the two areas, even if only due to their different age structure or location. These differences are likely to be matters of pattern rather than absolute differences in income levels. Both areas experience high levels of economic deprivation and marginalisation, benefit dependence in both communities is high, and conditions, although they may be improving, have plenty of scope for more improvement.

Return to list of contents