Background to the Research

• An earlier NI study (Hayes and McAllister 1996) on the workings of the political process and people's trust in it was published in Social Attitudes in NI fifth report (1996). The current study asked the same module of questions as were asked in the 1994 NISA, and is intended to complement the earlier analysis of the 1994 data.
  

Research Approach

• This book chapter involves secondary analysis of data from the 1996 NISA survey.
  

Main Findings

• NI differs from Britain in many social, economic and political aspects, which introduce different perspectives on the question of trust in politics and politicians. The author states that NI attitude surveys are often misleading, as respondents may state moderate views in surveys that subsequent events tend to disprove. As an example, the 1996 NISA survey reported only 2% support for Sinn Fein, whereas that party obtained 15.5% of the vote in the 1996 Forum Election. The political strength of the moderate Alliance Party tends to be overestimated in surveys.
• Religion is a key classification variable in surveys in NI. The 1996 NISA survey found that 14% of those surveyed stated that they were of "no religion". For the purposes of this study, the religion variable is based on the denomination in which the respondent was raised, i.e. their family's religion. The author justifies this classification on the basis that respondents attitudes are formed when they were part of a particular community, and many aspects of their circumstances e.g. education and employment, are determined by the community from which they come, even if they no longer actively practice their religion. The result of this reclassification for the present study was to reduce the number in the "no religion" category to 2%.
• The 1996 NISA survey contained a series of ten items designed to assess respondents' levels of political efficacy and political trust. Political efficacy was defined as whether or not individuals believe that they have, or can have, an impact on the political process. It also seeks to measure an individual's ability to understand and participate in political life, and the level of their belief that political representatives are responsive and responsible to the electorate. Political trust measures the extent to which individuals believe that political representatives act in the best interests of society, rather than in their own best interests or the interests of their political party.
• As the same questions were asked in the 1994 and 1996 NISA surveys, comparisons can be made. In the case of nine out of the ten items, there had been a decrease in the levels of political efficacy and trust. The exception was an increase in the number of people disagreeing with the statement "people like me have no say in what the government does". The author links this particular result to the fact that the 1996 NISA fieldwork coincided with the Forum Election, which had a high (65%) turnout. For the other nine items, levels of efficacy and trust declined by 2%-3% between 1994 and 1996. The study was also able to compare 1996 results in NI with data for Britain as a whole. While there were regional differences in the Britain figures between England, Scotland and Wales, NI results were lower, sometimes by a large margin, for seven out of the ten items. In another two the figures were the same.
• Within NI, there were differences in levels of political efficacy and trust between respondents raised in the Catholic community and those raised in the Protestant community. In eight of the ten items, Catholics indicated lower levels of efficacy and trust than Protestants. Seven of these differences were statistically significant. The study also considered other variables; gender, age, residential area, education, social class, political identification and party partisanship. Overall, in terms of political trust, the most significant differences were found in the variables of political identification, and party partisanship, with unionists reporting higher levels of trust than nationalists. In the case of efficacy there were significant differences depending on the level of education, with those with a higher level of education significantly more likely to report an understanding of the political process.
• An Index of political efficacy and trust was determined by assigning scores to the replies to the ten items, either 1 for disagreeing with a statement, or 0 for agreeing with it. The results showed that 58% of respondents in NI scored 0 for all five questions on political efficacy, and only 17% responded favourably to two or more of the questions. There were statistically significant differences in Catholic and Protestant responses, with 63% of Catholics reporting 0 over all five statements, compared to 55% of Protestants. In terms of political trust, 35% of all NI respondents scored 0. Again, the differences between the two religions were statistically significant (47% Catholics, 25% Protestants). In contrast, 37% of all NI respondents reported a "high" level of trust by responding favourably to two or more of the five statements.
• Further analysis of the Index figures by other independent variables showed significant differences in terms of gender, education and social class, but no significant differences in terms of age or area of residence. Significant differences were detected between Catholic and Protestant responses, after controlling for the independent variables. Overall, the results showed that in every category of every variable, there was a larger percentage of Protestants than Catholics who scored "high" on political efficacy and trust. This demonstrates a religious difference regardless of differences in other personal or socio-economic characteristics.
• The study then classifies respondents into four categories dependant on their efficacy and trust scores. Those classified as "weak democrats" score low on both efficacy and trust, believing that they cannot influence politics, and having no trust in the political system. By contrast, "strong democrats" score high in both efficacy and trust, believing in the political process and trusting those involved in it. The term "optimists" is used to define those who score low in political efficacy, but have trust in politicians, while "sceptics" score highly in political efficacy, but distrust those who are involved in politics. Comparisons with British figures show that "weak democrats" account for approximately half the respondents in both regions, while "optimists" account for another quarter. There are, however, fewer "strong democrats" in NI (9%) compared to Britain (16%), and the totals of "weak democrats" are higher (54% in NI, 48% in Britain).
• There are significantly fewer "strong democrats" among NI Catholics (5%) than Protestants (13%), and a higher number of "weak democrats" (65% of Catholics, 45% of Protestants). Similarly, the percentage of "optimists" in the Catholic population (23%) is lower than that of Protestants (33%).
• The study found significant differences in the levels of political efficacy and trust when investigating Catholic and Protestant respondents in terms of their interest in politics, political identification and party partisanship. Catholics were found to have lower levels of political efficacy and trust than their Protestant counterparts.

Conclusions

• The author concludes that NI's particular political system of direct rule is reflected in lower levels of political efficacy and trust than that found in respondents in Britain. The levels of regional difference are, however, not as great as those found between the two main communities in NI. Catholic and Protestant views on political efficacy and trust are widely divergent.
• The results also show a decline in democratic values between 1994 and 1996, with lower levels of political efficacy and trust reported in 1996 than in the 1994 NISA survey. The author points to some factors which may have influenced the 1996 results. Media stories of corruption in British politics, and the end of the Irish Republican Army's 1994 ceasefire on 9 February 1996 may have been factors influencing respondents' attitudes during the time the 1996 fieldwork was carried out.