1. Top of page
  2. Abstract
  3. The Decline in Public Confidence and Advocacy Democracy Reforms
  4. The Importance of Process
  5. Petitioner Expectations and Evaluations
  6. Modelling Process, Outcomes and System Trust
  7. Conclusion
  8. About the Author
  9. References

To (re)connect and (re-)engage citizens with their governing institutions, many advanced industrial democracies have adopted innovative reforms designed to ‘transform’ the way citizens interact with public institutions. This article assesses the extent to which the Scottish Parliament's petitioning system, a reform designed to connect the Scottish public with its parliament, influences broader trust in the political system. Using structural equations to model data from a survey of Scottish Parliament petitioners, the article finds that process evaluations far exceed outcome evaluations in influencing petitioner trust in political institutions. Hence simply adopting reforms that allow for ‘civic engagement’ is not enough to improve public support for the political system. Great care must be taken in adopting transformative reforms to ensure that those members of the public who do choose to ‘engage’ with political institutions see procedures as fair and politically neutral.

Recognising the need to confront a marked decline in public confidence in governing systems, many advanced industrial democracies have adopted transformative reforms designed to expand the opportunities for citizens to connect with their governments (Cain et al., 2003; Dalton, 2004) and thereby increase public support for political institutions. This was certainly the case during the consideration of the institutional arrangements that would form the Scottish Parliament. In keeping with the goal of establishing a ‘new politics’ in Scotland, a deliberate decision was taken to adopt innovative reforms that would forge and support a strong link between the Scottish public and the Scottish Parliament.1

According to the rhetoric of the day this set of civic engagement policies was designed to further the objective of spanning the ‘gap’ between the public and the parliament and create an open, transparent, accountable and accessible ‘people's parliament’. Policies were to encourage participatory democracy and allow for the direct proposal (and due consideration) of petitions to the parliament, unprecedented access to information on the parliament's operations, access to committee and chamber debates and the ability of members of the public to join with MSPs on cross-party groups.

Although highly innovative in the UK context, the Scottish Parliament's civic engagement policies and programmes reflect a concern across most established Western democracies with ‘engaging’ the public in governing. These programmes stem from beliefs reflected in the media as well as the findings of public opinion scholars that publics are increasingly disengaged from political affairs and distrustful of their political leaders and institutions (Dalton, 2004; Newton, 2006; Norris, 1999a; 2002).2 Responding to public calls for more participatory democracy, European governments have moved to expand the opportunities of citizens to be actively involved in policy development and formulation (Dalton et al., 2003, p. 2). Despite the increased use of ‘participatory democracy’ reforms, however, we know surprisingly little about the effectiveness of these programmes and whether they help to meet the larger goal of increasing public institutional support.

This article models petitioner evaluations of the Scottish Parliament's petitions system and their influence on trust in political institutions. As one of the more innovative petitions systems, housed within a developing parliament, Scotland's petitions provide a good case to begin building our knowledge of how advocacy reforms (Dalton et al., 2003) encourage participatory democracy and foster institutional support. Building on the procedural justice literature pioneered by Tom Tyler (1994; 1997; 2000), I propose that perceptions of the procedural ‘fairness’ of the petitions system are an important intervening variable in the relative effectiveness of the petitions system in bolstering public support. Advocacy reforms are distinct from direct democracy reforms in that the final decision-taking powers are reserved to the parliamentary institution (Dalton et al., 2003); hence perceptions of how these institutions ‘handle’ petitions may be a vital component in influencing public institutional support. Using data from a survey of petitioners (submitting petitions between 1999 and 2006) linked to data on actual parliamentary processes, I develop structural models linking petitioner perceptions of procedural fairness to broader political support. Before presenting this analysis I first provide a short review of the literature on transformative democracy reforms followed by a description of the Scottish Parliament's petitions system. The article concludes with a brief discussion of the importance of the choices politicians make while implementing democratic reforms for public confidence in political systems.

The Decline in Public Confidence and Advocacy Democracy Reforms

  1. Top of page
  2. Abstract
  3. The Decline in Public Confidence and Advocacy Democracy Reforms
  4. The Importance of Process
  5. Petitioner Expectations and Evaluations
  6. Modelling Process, Outcomes and System Trust
  7. Conclusion
  8. About the Author
  9. References

Public confidence and support for democratic institutions is declining across established Western democracies (Dalton, 1999; 2004; Newton, 2006; Norris, 1999a). Documenting this decline across several countries, Russell Dalton (2004) notes that in Britain there has been general erosion in support of the party system and a decline in confidence expressed in the House of Commons. John Curtice and Roger Jowell (1995; 1997) demonstrate that in the UK voters are increasingly ‘sceptical’ about government and ‘the ability of the system to respond to the demands of the citizenry’ (1995, p. 167), concluding that ‘Britain still appears to have low levels of confidence in both its political system and its politicians’ (1997, p. 107). Looking more specifically at Scotland, Curtice (2005) similarly documents the fact that confidence in governance has declined since the establishment of the Scottish Parliament.3 Of course, one possibility in the Scottish case is that during the devolution campaign in the late 1990s public expectations were raised to an unusually high level and have since returned to reflect a ‘normal’ level of dissatisfaction with governing institutions. Nonetheless, it is clear that general confidence in public institutions in Scotland broadly reflects that in other established industrial democracies.

The decline in public trust and support for leaders and institutions across democratic countries has been attributed to intensive coverage of scandals, the personal failings of politicians, partisan arguments and divisions in lawmaking bodies and the persistent criticism that governmental policies are inefficient and ineffective (see Dalton, 1999; see also Bromley et al., 2004). Some analysts have proposed that as a result of changing patterns of social connections within societies there has been a general decline in social capital and a concurrent increase in the disjuncture between citizens and political institutions (Putnam, 2001; but see Newton, 2006). Still other scholars have argued that given the changing nature of issues addressed by governments (Inglehart, 1997), popular expectations grew to include increased participation in governance (Dalton, 2004). As governing institutions persistently failed to meet these expectations, public disillusionment with democratic institutions has increased (Dalton, 1999; 2004). Pippa Norris (1999b), however, proposing that in addition to the previously discussed explanations we must also take into account more institutionally focused explanations, concludes that ‘citizens who live in democracies with a strong tradition of civil liberties expressed considerable confidence in their political system’ (p. 233).

Whatever the causal explanations, political disengagement and declining support for institutions have been tied to a general decline in the traditional forms of public political participation. While we must recognise that ‘there is little agreement about [the] consequences’ of institutional distrust (Norris, 1999a, p. 4), due, in part, to the difficulty in sorting out the causal direction, declining election turnout rates have been correlated with declining public political support. Other related symptoms of the public malaise are declining party affiliation (Whiteley and Seyd, 2002), campaign participation and general compliance with (some) law (Dalton, 2004).

However, Charles Pattie, Patrick Seyd and Paul Whiteley (2003a; 2003b; 2004) have found that public participation in the political sphere is multidimensional. People disinclined toward one form of civic participation may, instead, be more inclined to use another form of participation. The 2000 and 2001 Citizen Audit surveys found that alternative, non-traditional forms of participation and civic engagement, such as product boycotts, ethical shopping, fund-raising for causes, participating in public demonstrations and signing petitions are much in use across Britain (Pattie et al., 2003a; 2003b; 2004).

The evidence on participation in the UK – whether ‘traditional’ participation or not – is somewhat ambiguous. The Hansard Society's 2008‘Audit of Political Engagement’ series offers a less sanguine assessment of the state of public political engagement, stating:

Around 12% of people are politically active, according to our definition (i.e. in the last two or three years they have done at least three political activities from a list of eight). Almost half of the public (48%) report not having [signed a petition, boycotted products for political reasons, urged someone to contact an MP, contacted an MP, been to a political meeting, taken part in a demonstration, donated money or paid a membership fee to a political party or taken part in a political campaign]’ (Hansard Society, 2008, p. 4).

In short, there are a large number of potential explanations for the general decline in public trust in political – especially parliamentary – institutions.4 Further, depending on the survey sample and measures, there are different interpretations of the evidence on public political participation. For example, the just-mentioned Hansard study found that only ‘12% of people are politically active’ (Hansard Society, 2008, p. 4), while Pattie, Seyd and Whiteley, using a more extensive list of participatory options, report: ‘Contrary to the claims of political apathy, people frequently participate in activities designed to influence political outcomes’ (Pattie et al., 2003b, p. 621). Despite these conflicts in the literature, it is clear that there is a positive correlation between distrust and disengagement. Perhaps more importantly, the popular perception among politicians and policy makers is that they must ‘do something’ about public disengagement and create opportunities for the public to become more involved in the policy-making process (Dewar, 1998; Scottish Office, 1999).

To create these ‘new’ opportunities for participation, many advanced industrial democracies have adopted reforms designed to ‘transform’ established representative democracies into more participatory democracies. Dalton, Scarrow and Cain (2003) identify three sets of transformative reforms: representative reforms, direct democracy reforms and advocacy democracy reforms. The first of these reforms seeks to improve representative democracy. These reforms aim to alter candidate selection or fundamentally alter the electoral systems used to elect all levels of public officials (e.g. the use of the ‘Additional Member System’ for the Scottish Parliament or the adoption of the single transferable vote system for local Scottish councils).

A second set of reforms attempt to bypass the representative process altogether and establish direct democracy mechanisms whereby the citizenry is expected to be involved in all stages of the policy process, from making initial proposals, to deliberation and finally ratification. Direct democracy reforms have received a great deal of academic attention, in part owing to the prominence that public initiatives have taken in American state politics – especially California (see Gerber et al., 2001; Lupia and Matsusaka, 2004 for reviews). While many analysts have focused on the California tree, others have taken a broader view of the forest and examined public support for initiatives and referenda in a comparative context. Looking across ‘sixteen affluent democracies’, Bowler, Donovan and Karp (2007) found that support for direct democracy reforms is greatest among people who express high levels of distrust in government. Generally the politically suspicious wish to ‘better control discretion delegated to their representatives’ (Bowler et al., 2007, p. 14). Specifically looking at their findings for the UK, the authors report that people who do not think democracy is working well want more opportunities to participate. Focusing on the German context, Dalton, Burklin and Drummond (2001) note that direct democracy receives relatively greater support among those at the margins of politics: the less interested, the less educated and those who support protest parties.

A third type of reform, the type most relevant for this article, is designed to foster ‘advocacy democracy’. Advocacy democracy is marked by citizen participation in the formulation of policies, though citizens are not the final decision makers. As opposed to direct democracy reforms that transfer the decision-taking process directly to the public, advocacy reforms, according to Dalton, Scarrow and Cain, ‘seek to influence the process rather than make outright decisions, as is done with referendums and initiatives’ (Dalton et al., 2003, p. 11). According to this definition, it is clear that formalised petitioning systems should be categorised as advocacy reforms. Petitions systems allow for members of the public to raise issues or propose policies to the parliament while the scrutiny and decision-taking functions are reserved to the parliament itself. Advocacy reforms are oriented to involve members of the public in the political process. They are designed to allow the public to gather information, raise and move issues from the public to the government – or even decision – agenda5 and seek involvement in the policy process without formally having a role in the final outcome of any inquiry.6

Despite their use in many established political systems (e.g. European Union and Germany), petitions have been the subject of less academic scrutiny than have higher-profile direct democracy reforms (though see Hill, 1974; Judge, 1978). While many scholars have delved into public support for initiatives and referenda, shedding a great deal of light on public support for direct engagement in the legislative process as well as on the types of group that tend to use direct democracy mechanisms and the strategies they tend to employ (Bowler et al., 1998; 2007; Dalton et al., 2001; Gerber et al., 2001; LeDuc, 2003; Lupia and Matsusaka, 2004; Matsusaka, 2004), far fewer have given ‘advocacy’ reforms the same level of attention.

What can Scotland tell us about advocacy reforms and institutional support? As noted above, the Scottish petitions system was specifically incorporated into the institutional design of the Scottish Parliament (established in 1999) in part, to encourage a ‘new politics’ in Scotland marked by participatory democracy and an open, transparent and accountable parliament (Dewar, 1998; McGarvey and Cairney, 2008; Scottish Office, 1999). Petitions are initiated by individuals, groups or organisations and submitted to the Public Petitions Committee. For petitions to be considered they should be limited to policy-relevant concerns that fall within the purview of the parliament referred to as ‘devolved’ matters. These features distinguish the Scottish petitions system from others that may have either explicit ‘ombudsman’ functions (e.g. the petitions system in the German Bundestag) or require petitions to be submitted by chamber members (as is the case in the Westminster parliament). Once a petition is submitted, the parliament, and especially the Public Petitions Committee (PPC), is obliged to assess each petition, take evidence and finally take a decision on each petition. Further, the Scottish Parliament is setting new standards in openness and information availability, making it possible to track all petitions submitted to the parliament and all petitioners via the parliament's website.7 Hence, the petitions data needed to begin studying the system are readily available and span the entire life of the parliament.

In addition to the petitions system, Scotland presents a good case to assess these sorts of programmes. Although not racially diverse, Scotland is socio-economically diverse, having some areas of extreme wealth with some of the most expensive housing in the UK (outside London) as well as areas with extreme deprivation, poor health and the lowest life expectancy in Europe. It has commercial hubs, with a retail district that is generally recognised as being only second to London within the UK, but also has great expanses of extremely rural and isolated areas.

Most studies that focus on process preferences and procedural justice issues tend to be conducted at the local, city or community level (in the US case) where individuals can be expected to have more direct contact and personal experience with governing officials (Tyler, 1994). Tyler states (1994, p. 811): ‘National-level institutions differ from those usually studied in the procedural justice literature in their relative remoteness from citizens and in the more abstract nature of decision making’. Scotland, however, offers a balance: as a ‘small nation’ with a relatively open parliament it does not suffer from this remoteness; however, as a national (though ‘devolved’) parliament, it has primary legislative authority over devolved matters.8 Petitions, then, have the very real potential to influence legislative statutes.

The Importance of Process

  1. Top of page
  2. Abstract
  3. The Decline in Public Confidence and Advocacy Democracy Reforms
  4. The Importance of Process
  5. Petitioner Expectations and Evaluations
  6. Modelling Process, Outcomes and System Trust
  7. Conclusion
  8. About the Author
  9. References

The argument underlying this project is that transformative reforms, if they are to be successful, must be more than symbolic. It is not enough for transformative reforms, and especially advocacy reforms, simply to ‘allow for’ public engagement. Instead, allowing for engagement is simply the starting point of these reforms. If advocacy reforms are to foster, in any meaningful way, a participatory political culture and connect the public with governing institutions, then those individuals who do engage with the advocacy reforms must see the process by which these reforms work as politically neutral and potentially influential. This is due to the nature of advocacy reforms, which reserve the final decision-taking authority to the parliament. This leaves open the very real possibility that people who make use of the opportunity to ‘engage’ will not be happy with the final result and can place blame for their petition's ‘failure’ on the parliamentarians or parliamentary officials. Petitioners may level charges of bias or even malfeasance in the consideration of petitions. Under these conditions, the process by which petitions are considered is vitally important.

The research on ‘procedural justice’ developed by Tyler (1994; 1997; 2000), as well as Hibbing and Theiss-Morse's (2001) work emphasising public perceptions of political processes, provides unmitigated evidence that individual-level evaluations of how ‘fair’ (or ‘unfair’) a political process is have a very strong influence on the willingness to accept the outcomes of those processes. If individuals perceive political processes as ‘fair’ they tend to be more likely to accept the results of a process even if they did not gain their preferred outcome. On the other hand, if individuals believe that a process was not fair, then they tend to be much less enthusiastic about the outcome even if they achieved their preferred result. In other words, individuals are often willing to accept outcomes they do not prefer if they believe the outcomes were derived through a fair process.

As petitions systems are process oriented, the success or failure of these systems, as measured by petitioner satisfaction with outcomes, should largely be determined by the extent to which petitioners see the petitions consideration process as procedurally neutral. We can hypothesise, then, that petitioners who evaluate the overall petitioning process as fair are more likely to accept the outcomes of the petitioning process.

Beyond the acceptance of outcomes, it is also likely that petitioner procedural judgements will influence their relative levels of support for political institutions. Marcia Grimes argues that ‘personal ability to influence decision outcomes has less bearing on institutional legitimacy than assessments of the fairness of decision processes per se’ (Grimes, 2006, p. 306). Grimes continues: ‘Dissatisfaction with one's ability to exert influence in an issue may exact a much larger toll on the perceived legitimacy of political authorities if those authorities have promised opportunities for such influence but not fulfilled that promise’ (Grimes, 2006, p. 307). Following this, I argue that petitioners will use their judgements of the petitioning system, as an institution within the parliament, to make broader inferences about the political system. As most individuals do not interact with their parliaments on a regular basis, it is likely that individuals will have a relatively small set of interactions on which to base institutional evaluations. Interaction with the petitions system may serve as a cue, providing information on which to update prior beliefs about the parliament. If petitioners perceive that their petition was handled in a manner that they consider to be ‘unfair’, they may well use this assessment of parliamentary procedure to update any beliefs about the parliament, thus resulting in relatively low levels of support for the institution. And if the landmark institution that is to engage the public is not seen as ‘fair’ then this may impugn diffuse support for the entire system. Hence my second hypothesis: individual-level process evaluations will have a positive relationship with diffuse, systemic support at the Scottish (national) level. A corollary to this hypothesis is that the process evaluations will have a stronger relationship with system support than will outcome evaluations.

A third hypothesis posed in this article deals with the extent to which we might find ‘spillover’ effects on UK-level institutional trust. While it is possible that there may be some degree of crossover influence of process evaluations on trust in London-based political institutions, it is not likely that this influence will be substantial. The petitions system, conceived and implemented in Scotland, is a uniquely Scottish institution with no parallel at Westminster.9 It is likely that petitioners – the group under study in this project – would be aware of this distinction. On one hand it is possible that disgruntlement among petitioners may lead to the conviction that politicians, in general, are not open to public input through petitions. This negative affect may spill over to attitudes toward other UK political institutions. However, any spillover effects are likely to be limited.

Petitioner Expectations and Evaluations

  1. Top of page
  2. Abstract
  3. The Decline in Public Confidence and Advocacy Democracy Reforms
  4. The Importance of Process
  5. Petitioner Expectations and Evaluations
  6. Modelling Process, Outcomes and System Trust
  7. Conclusion
  8. About the Author
  9. References

To gauge petitioners' evaluations of the petitions system, the parliament and institutional trust, a survey was conducted during the summer of 2006. All primary petitioners were sent a questionnaire by post using the contact information they provided when petitioning the parliament.10 Repeat petitioners11 were sent one copy of the survey. In all, 722 questionnaires were sent by post to petitioners.12 Petitioners were provided with a postage-paid, addressed return envelope. A follow-up letter was sent to non-responders approximately three weeks after the first questionnaire packet was sent. Two weeks later, all petitioners who had not yet responded to the questionnaire were sent another copy of the questionnaire and a postage-paid, addressed envelope.13 The usable response rate, or the percentage of petitioners who returned a completed (or mostly completed) questionnaire, was 52 per cent.

Determining the extent to which the survey respondents are representative of the entire population of petitioners is difficult. As there are no comparable data on the entire population of petitioners, it is not possible to say exactly how well responses reflect all petitioners. That said, one factor that can be determined from the address file of all petitioners is the proportion of the population of petitioners from the different electoral regions across Scotland. Table 1 displays the distribution of all petitioners (from 1999 to June 2006) across Scotland's regions as well as the distribution of respondents to the survey of petitioners across the same regions. While there are some differences (e.g. Glasgow and Lothians petitioners are somewhat under-represented in the survey and South of Scotland petitioners are somewhat over-represented), generally the respondents to the survey reflect the overall geographic distribution of the population of respondents.

Table 1.  Regional Distribution of Petitioner Population and Survey Respondents
  Percentage of population Percentage of respondents
  1. Source:Survey of petitioners, June 2006.

Central Scotland88
Highlands and Islands1112
Mid Scotland and Fife1215
North East Scotland910
South of Scotland1720
West of Scotland78
Outside Scotland32

Further, 63 per cent of responses were received from petitioners who submitted their (first) petition during the first session of the parliament. This is roughly the same percentage of petitions submitted during the first session of the parliament. Responses to the petitioner survey largely reflect the relative proportions of first and second session petitioners in the general population of petitioners.

Inevitably there will be self-selection effects in petitioner responses to the survey. I cannot definitely say therefore how well my survey respondents ‘represent’ all petitioners. However, using the distribution of respondents across regions and parliamentary sessions we can have some degree of confidence that survey respondents will broadly correspond to the general population of petitioners.

The survey data were merged with a data file containing information on all petitions submitted to the Scottish Parliament's Public Petitions Committee up to June 2006. This provides us with a data set that combines survey data with contextual information covering the procedures and processes of the PPC in handling each petition. Combining these data sources I am able to compare the actual process through which each petition was considered with petitioner evaluations of those processes.

Before presenting the structural equations modelling the relationship between petitioner perceptions of procedural fairness and institutional trust, this section first summarises petitioners' expectations when they first submitted petitions, their evaluations of the petitions system and their assessments of how their petitions were handled by the PPC. These variables will then be used as the basis of the measurement equations in the structural model that follows. Of course, one concern with these survey data relates to the validity of first session petitioners' recall of their expectations and evaluations of the petitions process. In general, asking petitioners retrospectively to describe expectations they held when they first submitted a petition is obviously a less than perfect measure of petitioners' initial expectations. It is possible that the time gap (potentially of several years) from when first session petitioners submitted a petition to the time of the survey further undermines the validity of their responses. Additionally, these measures may be contaminated by their actual experiences. Difference of means tests between first and second session petitioners, however, reveal no significant differences in reported respondent expectations between these groups. While I cannot be absolutely certain that respondents' expectations and evaluations were not influenced by the passage of time, the statistical evidence would seem to indicate that there is little cause for concern.14

Looking at petitioner expectations, Table 2 shows that petitioners report they went into the process with rather high expectations. Nearly 90 per cent of respondents indicate that they expected that their petition would be handled fairly when they first submitted a petition to the parliament. Additionally, 86 per cent thought that petitioning the parliament would give them a voice in the parliament – that their concerns would be listened to. Overall, then, we find skewed expectations of the petitioning process with vast majorities of petitioners reporting they had strikingly high expectations the first time they petitioned the parliament.

Table 2.  Petitioner Expectations of Petitioning Process
  ‘When I first submitted a petition I expected that:
It would be handled fairly’ My concerns would be listened to by the parliament’
  1. Source:Survey of petitioners, June 2006; cell entries are percentages.

Agree strongly33.333.6
Neither agree nor disagree7.89.8
Disagree strongly2.02.2

Process-oriented questions asked respondents to what extent they agreed that the petitions system is ‘easy to understand’, that the PPC kept them informed of their petition's progress and handled their petition ‘fairly’. Table 3 presents the question wording and the distribution of petitioners' responses. Petitioners report that the system is generally easy to understand with an astonishing 89 per cent of respondents agreeing with the statement. However, petitioners felt much less in touch with their petition throughout the process, with 35 per cent agreeing that they had been kept informed of their petition's progress and almost 56 per cent reporting they had not been kept informed. Turning to whether petitioners agreed that their petition was handled fairly by the PPC, we find that almost 36 per cent thought that their petition was handled fairly, while almost 13 per cent did not think that their petition received fair treatment from the PPC. Looking back at Table 2, there is a drop of about 54 per cent between the percentage of petitioners who expected their petition to be handled fairly and the percentage that actually thought their petition was handled fairly.15,16

Table 3.  Petitioner Process Evaluations
  Agree strongly Agree Neither agree nor disagree Disagree Disagree Strongly
  1. Source:Survey of petitioners, June 2006; cell entries are percentages.

I found the petitions process easy to understand13.3676.345.073.381.84
The Petitions Committee kept me informed12.9223.546.9252.154.46
My petition was handled fairly by the Petitions Committee14.3122.0051.085.387.23

Considering the three ‘process’ measures together, the average inter-item correlation (Cronbach's alpha) is 0.80. Structurally, a factor analysis of the three items reveals a single factor solution with an eigenvalue of 1.44.

Table 4 presents petitioner marginal response rates to outcome-oriented questions asking them to evaluate the extent to which they thought their petition received full consideration, their petition was a success and whether they were satisfied with the outcome. Approximately 76 per cent of respondents disagreed with the statement, ‘I consider my petition to be a success’, 16 per cent rated their petition a success and 8 per cent neither agreed nor disagreed. Asking respondents to what extent they agreed with the statement that ‘My petition did not receive full consideration’, we find that roughly 70 per cent of respondents agreed that their petition did not receive full consideration while 23 per cent were satisfied with the level of consideration the PPC gave to their petition.17 Finally, we look to petitioners' general evaluations of the outcome of the petitioning process. Approximately 29 per cent of petitioners were not satisfied with the outcome of the petitioning process. Conversely, nearly 17 per cent of petitioners were satisfied with the outcome. Factor analysis reveals that these three items load on the same factor (eigenvalue of 2.53). The three measures are highly correlated with a Cronbach's alpha of 0.91.

Table 4.  Petitioner Outcome Evaluations
  Agree strongly Agree Neither agree nor disagree Disagree Disagree strongly
  1. Source:Survey of petitioners, June 2006; cell entries are percentages.

My petition received full consideration8.8213.767.7658.9110.76
I consider my petition to be a success6.898.667.7761.8414.84
I am satisfied with the outcome of the petitioning process4.9311.9754.7511.416.90

A confirmatory analysis revealed that the process and outcome factors are correlated, with the process ‘fairness’ item showing a slight tendency to cross-load on the ‘outcomes’ factor. Hence, while the concepts being studied here are statistically distinct, as one would expect (Verba, 2006) there is some degree of conceptual blurring in respondents' perceptions and evaluations.18

What does all this tell us about petitioner assessments of the petitioning system in Scotland? Of those people who do participate in the petitions system, they indicate that they first petitioned the parliament with relatively high expectations. They expect that their petitions will be handled in a fair manner. And many believe that, after they have completed the petitioning process, this is indeed the case. However evaluations of outcomes are more mixed. Petitioners report relatively high levels of belief that their petitions did not receive due consideration.

Modelling Process, Outcomes and System Trust

  1. Top of page
  2. Abstract
  3. The Decline in Public Confidence and Advocacy Democracy Reforms
  4. The Importance of Process
  5. Petitioner Expectations and Evaluations
  6. Modelling Process, Outcomes and System Trust
  7. Conclusion
  8. About the Author
  9. References

What do these evaluations and expectations mean for attitudes toward the larger governing system? Recall that I hypothesised that individuals' evaluations of the parliament's petitions system would influence their overall level of support for the political system. To test these hypotheses I estimate a structural equations model, linking petitioner initial expectations, procedural evaluations, outcome evaluations and systemic trust. The model simultaneously estimates a series of equations designed to differentiate the influence of petitioner evaluations of process from outcome evaluations on support for the broader system. Given the hypotheses discussed above, I expect that process evaluations will predict outcome evaluations. More importantly, I expect that process evaluations will better predict support for the political system than will outcome evaluations.

In this system of equations the previously discussed measures of process evaluations (see Table 3) are used to measure the latent factor ‘Process Evaluations’. Another latent factor measures petitioner outcome evaluations using the measures presented in Table 4. As petitioner expectations, whether unduly high or low, may influence their perceptions of the petitioning process, the factor petitioner ‘Expectations’ (from Table 2) is included in the structural model as a control. A second exogenous factor, ‘Efficacy’,19 is included to control for individuals' prior beliefs that they are able to influence political decisions at the local and national levels.

Trust in the governing institutions and political system is measured with two different factors: one indicating respondent ‘trust’ in Scottish institutions, the other reflecting respondent ‘trust’ in UK institutions. The measured variables contributing to each factor come from responses to a question beginning: ‘Now thinking of political institutions like the Scottish Parliament, please use a scale of 0–10 to indicate how much trust you have for each of the following, where 0 means ‘no trust’ and 10 means ‘a great deal of trust’. Respondents were asked to rate their relative trust for: ‘the Scottish Parliament’ (TRSTPARL), ‘the Scottish Executive (McConnell's government)’20 (TRSTEXEC) and ‘Scottish politicians, generally’ (TRSTPOLS), as well as ‘the parliament at Westminster’ (TRSTWEST), ‘the British government (Blair's government)’ (TRSTPM) and ‘British politicians, generally’ (TRSTUKP). The expectation is that petitioners' evaluations of the petitions system will have a far greater influence on the Scottish political system than on the UK system as a whole.

Applying maximum likelihood estimation with robust standard errors to these data, four main equations of interest are simultaneously estimated:

  • image(1)

The variable ‘Days Open’ measures the (logged) number of days the first petition submitted by each petitioner was ‘open’ for consideration by the parliament's Public Petitions Committee. In other words, it is the gap, in number of days, between the date that a petition was submitted and the date that the petition was finally closed by the PPC. One obvious possibility is that petitions closed quickly might not be seen as having been fully considered by the PPC.21

The second equation models petitioner outcome evaluations as a direct result of procedural evaluations:

  • image(2)

Finally the third and fourth equations model respondent trust in the Scottish and UK political systems while allowing the disturbance terms for the two dependent variables to be correlated.

  • image(3)
  • image(4)

In addition to the Process and Outcomes factors, the two equations include controls for respondents' activity in political organisations and groups22 as well as identification with either of the two parties making up the Scottish Executive (now called the Scottish Government) in 2006 (i.e. Labour or Liberal Democrats)23 or, in the UK equation, the prime minister's Labour party.24 Given the literature on social capital, I would expect that individuals more tied to their communities through political groups or organisations may express higher levels of institutional trust (Putnam, 2001). Additionally, people who identify with either of the two parties making up the Scottish Executive or the UK government may be expected to express higher levels of ‘trust’ in the government of the day.

The standardised solution to the structural model presented in Figure 1 supports the hypothesis that procedural evaluations will influence petitioner assessments of petition outcomes. Clearly, the more ‘fair’ that individuals saw the process of petitioning the higher they rated the outcomes of the petitioning process (β = 0.85, p < 0.01). The pseudo R-square for the Outcome equation is a strong 0.71.


Figure 1. Structural Model of Process Evaluations on System Trust Notes:Model estimated in EQS 6 using maximum likelihood estimation with robust standard errors. Coefficients displayed are the standardised solution, CFI = 0.90; N-NFI = 0.88; Chi Square = 593.66 (193 df). Source:Survey of petitioners, June 2006.

Download figure to PowerPoint

More interestingly, we find very clear support for the hypothesis that it is individual-level assessments of procedural fairness that have the greatest influence on trust in the political system. When individuals saw the petitioning process as being fairer, they expressed greater trust in Scottish and UK political systems. The effect, however, is substantially greater on trust in the Scottish political system, where a 1 standard deviation increase in process evaluations predicts a 0.72 standard deviation shift in favour of greater system trust (p < 0.01). When looking to trust in the UK system the coefficient is almost half that predicting trust in the Scottish system and, more importantly, with a Z-score of just 1.6, the relationship fails to achieve statistical significance. Likewise, outcome evaluations fail to achieve statistical significance in either of the two equations predicting system trust. Overall, while the equation predicting system trust (Scotland) can be said to ‘explain’ approximately 58 per cent of the variance in the Scottish factor, the UK-level equation reports a pseudo-R2 of just 0.22. As a whole, the system of equations seems to provide a solid model of institutional trust as a function of process and outcome evaluations.25 The comparative fit index (CFI) equals 0.90 and the Bentler-Bonett non-normed fit index equals 0.88 with an overall model Chi-square of 593.66 (193 df).

One concern about this system of equations, however, may be that it includes only a limited set of controls. To test whether the general model holds with better-specified equations, we turn to two OLS regression equations (estimated using robust standard errors) separately predicting trust in the Scottish and UK political systems.26 The dependent variables in these equations are simply additive indices, combining the three Scottish trust variables into a variable ranging from 0 to 30 (high values indicating a greater degree of institutional trust) and the same for the three UK variables.

In addition to additive indices measuring process and outcome evaluations, the OLS equations contain a wide swathe of control variables. To account for the influence of partisan ties I include measures that asked respondents ‘how likely is it that you would ever vote for each of the following parties?’ The likely vote measures (using 0–10 scales) were asked for the four main political parties (Labour, SNP, Conservatives and Liberal Democrats) as well as for the Greens. Additionally, I include dummies for respondent self-identification with the main party groups (with Labour as the excluded category) as well as for those respondents who indicated that they did not identify with any of the political parties and those who responded ‘don't know’. While including both sets of partisan measures may seem a bit unwieldy, recall that the goal here is to control for partisan influences as thoroughly as possible.

The OLS models also control for political efficacy,27 social trust,28 respondent's degree of political activity29 and frequency of political discussion with friends and neighbours.30 Finally, both equations include controls for the demographic variables age (in years), sex (women = 1), university education (university = 1, otherwise 0) and self-identification as working class (working class = 1, otherwise 0).

The results presented in Table 5 (Model 1) clearly indicate that evaluations of how ‘fair’ the petitions system is influences trust in the Scottish political system. A full range change in process evaluations (holding all other variables in the model at their mean) would predict moving from a score of about 4 on the system trust scale for a petitioner who thought the process was completely unfair to a system trust score of approximately 16.5 for a petitioner who thought the petitioning process was very fair. In other words, after controlling for partisanship, efficacy, social capital and demographic measures, process perceptions can move petitioners almost halfway along the Scottish system trust scale while outcome evaluations exhibit a negligible, statistically insignificant influence.

Table 5.  OLS Regression of Scottish and UK System Trust on Process and Outcome Evaluations and Controls
  Model 1 Model 2
Trust Scottish system Trust UK system
b r.s.e. b r.s.e.
  1. Note:Equations estimated with robust standard errors

  2. Source:Survey of petitioners, June 2006.

  3. * p < 0.05; ** p < 0.01;***p < 0.001.

Petitions Process (0–12)0.39**
Petitions Outcomes (0–12)
Likely vote for: Labour (0–10)0.77**0.161.22**0.17
Likely vote for: Conservatives (0–10)**0.16
Likely vote for: Liberal Democrats (0–10)0.45*
Likely vote for: SNP (0–10)−0.040.13−0.080.15
Likely vote for: Greens (0–10)0.020.13−0.050.15
PID: Conservative (0,1)−0.241.63−4.36*2.03
PID: Liberal Democrat (0,1)0.581.640.401.81
PID: SNP (0,1)3.171.75−0.691.87
PID: Green (0,1)
PID: Socialist (0,1)−0.751.68−1.191.61
PID: None (0,1)0.641.49−0.161.66
PID: ‘Don't Know’ (0,1)−0.021.73−3.211.89
Efficacy (0–8)0.230.18−0.000.17
Social trust (0–8)0.39*
Active in political organisations (0–5)1.44**0.340.99**0.38
Talk politics (0–3)−1.13**0.41−0.400.40
Age (in years)
University (0,1)0.110.681.390.76
Women (0,1)1.050.67−0.130.67
Working class (0,1)−0.320.700.450.78
 R2adj = 0.63 N = 234R2adj = 0.48 N = 235

The other variables in the model generally perform as one would expect. Not surprisingly, Labour and Liberal Democrat voters tended to express a greater degree of trust in the Scottish political system at a time when Labour and Liberal Democrats formed the Scottish Executive (Government). Individuals with higher levels of social trust and those more active in political organisations also tended to express higher degrees of support for the Scottish system. On the other hand, individuals reporting that they spoke with friends and neighbours about politics a great deal tended to exhibit lower levels of trust in the Scottish political system. Given the nature of the political topics of the day – the excessive costs of the parliament building, expenses scandals and the controversy over the reform of Section 28 – the negative relation between political talk and lower trust may make sense. The adjusted R2 for the model is a healthy 0.63.

Turning to trust in the broader UK system centred on Westminster and Downing Street, there is no evidence whatsoever that, in the face of extensive controls, evaluations of the uniquely Scottish petitions system influences trust in the UK political system. While, again, the other variables in the model perform as one would expect, evaluations of the petitions system bear no relation to support for UK political institutions and politicians.

The statistical findings in this article are fairly straightforward. The structural and OLS models show that process evaluations related to a key institutional reform designed to ‘engage’ members of the public with policy making have a clear and significant influence on support for the political system among the citizens who make use of the opportunity to become involved. This link is limited, however, to the immediate system that adopts and implements the reforms – there is no evidence that sentiments of institutional trust spill over to the broader level. As the petitions system is limited to matters devolved to the Scottish Parliament and is in itself a uniquely Scottish institution, the ‘positive’ effects are, not surprisingly, limited to the Scottish political system.


  1. Top of page
  2. Abstract
  3. The Decline in Public Confidence and Advocacy Democracy Reforms
  4. The Importance of Process
  5. Petitioner Expectations and Evaluations
  6. Modelling Process, Outcomes and System Trust
  7. Conclusion
  8. About the Author
  9. References

The old chestnut in politics is that perception is everything. Not only must policies, procedures and programmes be ‘fair’, the public also must perceive them as fair. Indeed, in terms of institutional support the latter is almost more important than the former. Many politicians have seen their careers ruined because they were perceived to be involved in scandalous behaviour that was, in fact, fully legal and legitimate. It appears that in adopting transformative democracy reforms designed to (re-)engage the public and foster a political culture that values participatory democracy, great care and attention must be devoted to public perceptions of these programmes. If the programmes are actually going to fulfil their promise of reconnecting publics with political institutions and increasing relative levels of political support, it is important that people who actively engage with these programmes see them as being ‘fair’ and politically neutral.

This research further serves to highlight the fact that ‘process’ matters for public evaluations of political institutions. Procedural justice scholars have demonstrated that in several contexts individuals' assessments of how ‘fair’ they believe a process to be will influence their willingness to accept outcomes. This research demonstrates that this is certainly the case with petitioning systems. As opposed to other variants of civic engagement reforms that rely on deliberative or direct democracy mechanisms, in petitioning systems the control of procedures and the final decision-taking authority remain in the hands of the parliament. This likely heightens the importance of individuals' evaluations of procedural fairness.

Of course, we must acknowledge that almost ten years after it began, the popular uptake on the petitions system is still somewhat limited. However, it is less than clear what the appropriate institutional comparison might be. As the Scottish system specifically rules out an ombudsman function, enforces the rule that petitions must be relevant to broad policy concerns, and will not allow ‘frivolous’ or ‘failed’ petitions to be (re)submitted, the overall numbers of petitions submitted should obviously not be compared with the German, European or Westminster systems. All the same, members of the public have not been beating down the doors to submit petitions to the Scottish Parliament. This would seem to imply that the overall influence of the Scottish petitions system on broad institutional support is likely limited to petitioners.

One clear implication of this research, however, is that the choices made by elected parliamentarians and unelected committee staff in considering petitions may have a substantial influence on the effectiveness of policies designed to (re-)engage the public with governing institutions. As is often the case in implementing political reforms, the devil is in the detail. Here, seemingly pedantic decisions about procedure may have much wider consequences than individual decision makers may realise.

About the Author

  1. Top of page
  2. Abstract
  3. The Decline in Public Confidence and Advocacy Democracy Reforms
  4. The Importance of Process
  5. Petitioner Expectations and Evaluations
  6. Modelling Process, Outcomes and System Trust
  7. Conclusion
  8. About the Author
  9. References

Christopher J. Carman is the John Anderson Research Senior Lecturer in Policy Change in the Government Department at the University of Strathclyde. He has previously taught at the Universities of Glasgow and Pittsburgh. In addition to projects examining political representation in the UK and the US, he has an ongoing project examining public engagement with the Scottish petitions system. He has published in a variety of journals as well as authored reports on the Scottish petitions system for the Scottish Parliament.

Christopher J. Carman, Department of Government, University of Strathclyde, Glasgow, Scotland; email:

  • The author would like to thank the Political Studies editor and anonymous reviewers for their very helpful and insightful comments. Robert Johns, John Curtice, Laura Miller and EPOP panel participants also provided most helpful comments on this project. A previous version of this article was presented at the Annual Meeting of the Elections, Public Opinion and Parties study group of the Political Studies Association, Bristol, UK, 7–9 September 2007. The Economic and Social Research Council supported research for this article under grant RES-000-22-1820-A.

  • 1

    In the foreword to the report of the Consultative Steering Group (CSG) that drafted the initial framework for the Scottish Parliament, the CSG chairman, Henry McLeish, MP (who would go on to be the second First Minister of Scotland) stated: ‘the establishment of the Scottish Parliament offers the opportunity to put in place a new sort of democracy in Scotland, closer to the Scottish people and more in tune with Scottish needs ... In particular our recommendations envisage an open, accessible Parliament; a Parliament where power is shared with the people; where people are encouraged to participate in the policy making process’ (Scottish Office, 1999; see also Arter, 2004; Winetrobe, 2001).

  • 2

    Pattie, Seyd and Whiteley (2003a; 2003b; 2004), as discussed below, demonstrate that ‘traditional’ forms of participation are on the decline, while newer, alternative forms of participation are remaining level, if not increasing in use in the UK.

  • 3

    For a somewhat contrary view, the 2007 Scottish Social Attitudes survey reports a significant increase in trust in the Scottish Government, up around 20 percentage points (to 71 per cent) from the 2006 survey. This ‘represents the highest level of trust in devolved government since 1999, where results arguably reflected aspirations for the new institutions rather than assessments of their performance in practice’ (Ormston, 2008, p. 5). It is, of course, highly likely that this significant surge in ‘trust [in] the Scottish Executive to work in Scotland's best interests’ reflects a honeymoon period with the new (at the time of the survey), minority SNP government. The expectation is that these levels would slip over time. Further, according to the same survey less than half of the Scottish public believes that they can ‘trust’ the Scottish Government to make ‘fair decisions’. Public concern about the processes of governing in Scotland are also reflected in the fact that only 43 per cent of survey respondents think that the Scottish Government is ‘good’ at listening to people's views before taking decisions.

  • 4

    See Norris (1999a) for a very cogent treatment of the distinctions between diffuse and specific institutional support (or ‘trust’).

  • 5

    See Cobb and Elder (1972) for the classic work on levels of agenda.

  • 6

    We must recognise that involving members of the public in ‘the political process’ could be problematic. Hibbing and Theiss-Morse (2001) have clearly demonstrated that people tend to have a rather limited understanding of political processes and the inherent bargaining that they involve. Thinking of the classic analogy, problems may arise when individuals who do not regularly participate in politics see sausage being made.

  • 7
  • 8

    According to McGarvey and Cairney (2008, p. 2), the Scotland Act 1998 sets out a ‘clear list of reserved powers which UK State institutions Westminster and Whitehall retain with the rest falling under the jurisdiction of the Scottish Parliament and Government’. For example, a few of the policy areas devolved to the Scottish Parliament and Government are: health, education, economic development, criminal and civil law and home affairs, transport, environment, agriculture, fisheries and local government. See McGarvey and Cairney (2008) for a concise introduction to the topic of devolution.

  • 9

    It is true, as previously mentioned, that petitions may be submitted to the House of Commons. However, this requires that an MP submit the petition to the Speaker on behalf of the petitioning citizen. After submission, House of Commons rules stipulate that petitions be forwarded to the relevant minister or public official. Ministers are under no obligation (beyond that of political expediency) to respond to petitions. Petitions to the House of Commons (and to the prime minister through the internet-based Number 10 system) are strictly advisory.

  • 10

    This information is publicly available on the Scottish Parliament's Public Petitions Committee website, available from:

  • 11

    In total, there have been 94 repeat petitioners submitting a total of 342 petitions. Repeat petitioners submitted approximately 35 per cent of all petitions up to June 2006. On average, repeat petitioners submitted 3.6 petitions with the median number of petitions submitted by repeaters being 2.

  • 12

    This equates to the number of unique primary petitioners.

  • 13

    This general methodology is suggested by Mangione (1995).

  • 14

    Additional statistical tests found no difference between the reported expectations of individuals with ‘open’ petitions and those with ‘closed’ petitions. To clarify, ‘open’ petitions are those on which the PPC has yet to take a final decision, while closed petitions are those on which the PPC has taken a final decision and ‘closed’ the petition to further consideration.

  • 15

    The correlation between fairness expectations and evaluations is only approximately 0.18 (significant at p < 0.001). Petitioners who expected their petitions to be handled fairly generally were inclined to say they were handled fairly, while petitioners who expected unfair treatment reported that they were treated unfairly. However, the relatively low correlation coefficient (fairness expectations may only be said to ‘explain’ about 3 per cent of the variance in fairness evaluations) indicates that the overall relationship between fairness expectations and evaluations is weak.

  • 16

    Difference of means tests comparing petitioners from the first and second sessions of parliament revealed no significant differences in the means of the expectation and evaluation measures between these two groups of petitioners.

  • 17

    Table 4 presents the ‘flipped’ version of this variable, recoded to reflect ‘fair’ as the high value. This was done as consistent variable coding is required for the structural model and indices computed in the article.

  • 18

    The factor analyses reported here represent each ‘factor’ analysed separately. However, to test the overall structure I conducted a confirmatory factor analysis containing all process and outcome items. Here a two-factor structure was revealed, with 68 per cent of variance ‘explained’, the process items loading on factor one (‘process’) while the outcome items loaded on factor two (‘outcomes’). The item assessing ‘process fairness’ revealed a slight tendency to load on both factors (0.71 on the ‘process’ factor and 0.53 on the ‘outcomes’ factor).

  • 19

    The measure of ‘efficacy’ is an additive index of two 5-point agree–disagree scales: ‘People like me can influence political decisions affecting my local area’ and ‘People like me can influence political decisions affecting Scotland’.

  • 20

    The decision was taken to specifiy ‘McConnell's government’ following ‘Scottish Executive’ based on survey experiments conducted by the Scottish Social Attitudes series indicating that, as of 2004, the Scottish public did not clearly distinguish between the terms ‘Scottish Executive’ and ‘Scottish Parliament’ (see Bromley and Given, 2005). Therefore, First Minister McConnell was specified in the question to help respondents distinguish between the parliament and the government. This, of course, does increase the likelihood that naming the first minister may have lowered reported trust (i.e. respondents may have evaluated McConnell's performance rather than the generic institution of the Scottish Executive). However, as the models presented control for whether respondents identified with the parties in government at the time, this may help to control for response bias induced by specifically mentioning the first minister by name. One further point of clarification: to balance the questions, the prime minister was also identified by name.

  • 21

    I considered the possibility that the relationship between the gap variable and process evaluations might be curvilinear: people whose petitions were closed very quickly and those whose petitions were open for extended periods of time might have lower procedural evaluations. Models including a curvilinear term did not perform any better than the ‘days open’ variable.

  • 22

    Using a series of yes/no tick boxes, respondents were asked to indicate if they were a member of: ‘a political organisation other than a political party’, ‘a community group’, ‘a social organisation or sports club’, ‘a religious group or organisation’ or ‘a voluntary organisation’. Dichotomised into 1 (for ‘yes’) and 0 (‘no’), these five variables were added to compute a 0–5 measure of ‘activism’ (mean = 1.16).

  • 23

    Measured as a 1, 0 dummy variable indicating whether the respondent identified with Labour or Liberal Democrats.

  • 24

    Measured as a 1, 0 dummy variable indicating whether the respondent identified with Labour.

  • 25

    One possibility is that instead of estimating trust in the Scottish and UK systems separately, the model fit would improve if a single factor of overall system trust (combining the Scottish and UK measures) was estimated. Testing this possibility, the system was re-estimated using one factor for system trust. This resulted in a substantially poorer overall fit (CFI = 0.69, X2 = 1,004.80). Despite the poorer fit, however, the general result held, with process evaluations (β = 0.57, p < 0.01) trumping outcome evaluations (β = 0.04, p > 0.05).

  • 26

    Seemingly unrelated regression (SUR) models were considered for this portion of the analysis to allow for the correlation of the error terms across the equations predicting Scottish and UK institutional trust. However, according to Greene (1990), SUR models using the same set of regressors in the two equations provide less efficient estimates. Greene suggests estimating two distinct OLS equations in this case.

  • 27

    Additive index of measures as previously described.

  • 28

    Additive index of two 5-point agree–disagree scales, asking respondents: ‘Most ordinary people you meet can be trusted’ and ‘You can't be too careful dealing with people’.

  • 29

    Additive index of measures as previously described.

  • 30

    Responses to 4-point scale asking if respondents ‘frequently’, ‘sometimes’, ‘rarely’ or ‘never’‘talk to friends and neighbours about politics’.


  1. Top of page
  2. Abstract
  3. The Decline in Public Confidence and Advocacy Democracy Reforms
  4. The Importance of Process
  5. Petitioner Expectations and Evaluations
  6. Modelling Process, Outcomes and System Trust
  7. Conclusion
  8. About the Author
  9. References
  • Arter, D. (2004) The Scottish Parliament: A Scandinavian-Style Assembly? London: Frank Cass.
  • Bowler, S., Donovan, T. and Tolbert, C. (eds) (1998) Citizens as Legislators: Direct Democracy in the United States. Columbus OH: Ohio State University Press.
  • Bowler, S., Donovan, T. and Karp, J. (2007) ‘Enraged or Engaged? Preferences for Direct Citizen Participation in Affluent Democracies’, Political Research Quarterly, 60 (3), 35162.
  • Bromley, C., Curtice, J. and Seyd, B. (2004) ‘Is Britain facing a crisis of democracy?’, London: CREST/ESRC briefing [online]. Available from:[Accessed 4 March 2010.
  • Bromley, C. and Given, L. (2005) Public Perceptions of Scotland after Devolution. Edinburgh: Scottish Executive Social Research.
  • Cain, B. E., Dalton, R. J. and Scarrow, S. E. (eds) (2003) Democracy Transformed? Expanding Political Opportunities in Advanced Industrial Democracies. Oxford: Oxford University Press.
  • Cobb, R. and Elder, C. (1972) Participation in American Politics: The Dynamics of Agenda-Building. Boston: Allyn and Bacon.
  • Curtice, J. (2005) ‘Restoring Confidence and Legitimacy? Devolution and Public Opinion’, in A. Trench (ed.), Has Devolution Made a Difference? London: Constitution Unit, pp. 21735.
  • Curtice, J. and Jowell, R. (1995) ‘The Sceptical Electorate’, in R. Jowell, J. Curtice, A. Park, L. Brook, D. Ahrendt and K. Thomson (eds), British Social Attitudes, the 12th Report. Aldershot: Ashgate, pp. 14171.
  • Curtice, J. and Jowell, R. (1997) ‘Trust in the Political System’, in R. Jowell, J. Curtice, A. Park, L. Brook, K. Thomson and C. Bryson (eds), British Social Attitudes, the 14th Report. Aldershot: Ashgate, pp. 89109.
  • Dalton, R. J. (1999) ‘Political Support in Advanced Industrial Democracies’, in P. Norris (ed.), Critical Citizens: Global Support for Democratic Governance. Oxford: Oxford University Press, pp. 5776.
  • Dalton, R. J. (2004) Democratic Challenges, Democratic Choices. Oxford: Oxford University Press.
  • Dalton, R. J., Burklin, W. and Drummond, A. (2001) ‘Public Opinion and Direct Democracy’, Journal of Democracy, 12 (4), 14153.
  • Dalton, R. J., Scarrow, S. E. and Cain, B. E. (2003) ‘New Forms of Democracy? Reform and Transformation of Democratic Institutions’, in B. E. Cain, R. J. Dalton and S. E. Scarrow (eds), Democracy Transformed? Expanding Political Opportunities in Advanced Industrial Democracies. Oxford: Oxford University Press, pp. 122.
  • Dewar, D. (1998) ‘The Scottish Parliament’, Scottish Affairs: Special Issue on Understanding Constitutional Change, 412.
  • Gerber, E., Lupia, A., McCubbins, M. and Kiewiet, D. R. (2001) Stealing the Initiative. Upper Saddle River NJ: Prentice-Hall.
  • Greene, W. H. (1990) Econometric Analysis. New York: Macmillan.
  • Grimes, M. (2006) ‘Organizing Consent: The Role of Procedural Fairness in Political Trust and Compliance’, European Journal of Political Research, 45 (2), 285315.
  • Hansard Society (2008) The Audit of Political Engagement 5, the 2008 Report. London: Hansard Society.
  • Hibbing, J. and Theiss-Morse, E. (2001) ‘Process Preferences and American Politics: What the People Want Government to Be’, American Political Science Review, 95 (1), 14553.
  • Hill, L. B. (1974) ‘Parliamentary Petition, the Ombudsman and Political Change in New Zealand’, Political Studies, 22 (3), 33746.
  • Inglehart, R. (1997) Modernization and Postmodernization: Cultural, Economic and Political Change in 43 Societies. Princeton NJ: Princeton University Press.
  • Judge, D. (1978) ‘Public Petitions and the House of Commons’, Parliamentary Affairs, 31 (4), 391405.
  • LeDuc, L. (2003) The Politics of Direct Democracy: Referendums in a Global Perspective. Toronto: Broadview Press.
  • Lupia, A. and Matsusaka, J. G. (2004) ‘Direct Democracy: New Approaches to Old Questions’, Annual Review of Political Science, 7, 46382.
  • McGarvey, N. and Cairney, P. (2008) Scottish Politics: An Introduction. Basingstoke: Palgrave.
  • Mangione, T. W. (1995) Mail Surveys: Improving the Quality. Thousand Oaks CA: Sage.
  • Matsusaka, J. (2004) For the Many or the Few: The Initiative, Public Policy and American Democracy. Chicago IL: Chicago University Press.
  • Newton, K. (2006) ‘Political Support: Social Capital, Civil Society and Political and Economic Performance’, Political Studies, 54 (4), 84664.
  • Norris, P. (1999a) ‘Introduction: The Growth of Critical Citizens?’, in P. Norris (ed.), Critical Citizens Global Support for Democratic Governance. Oxford: Oxford University Press, pp. 130.
  • Norris, P. (1999b) ‘Institutional Explanations for Political Support’, in P. Norris (ed.), Critical Citizens Global Support for Democratic Governance. Oxford: Oxford University Press, pp. 21734.
  • Norris, P. (2002) Democratic Phoenix. New York: Cambridge University Press.
  • Ormston, R. (2008) Scottish Social Attitudes Survey 2007 Core Module Report 1: Attitudes to Government in Scotland. Edinburgh: Scottish Government Social Research.
  • Pattie, C., Seyd, P. and Whiteley, P. (2003a) ‘Citizenship and Civic Engagement: Attitudes and Behaviour in Britain’, Political Studies, 51 (4), 61633.
  • Pattie, C., Seyd, P. and Whiteley, P. (2003b) ‘Civic Attitudes and Engagement in Modern Britain’, Parliamentary Affairs, 56 (4), 61633.
  • Pattie, C., Seyd, P. and Whiteley, P. (2004) Citizenship in Britain: Values, Participation and Democracy. Cambridge: Cambridge University Press.
  • Putnam, R. (2001) Bowling Alone: The Collapse and Revival of American Community. New York: Simon & Schuster.
  • Scottish Office (1999) Shaping Scotland's Parliament: Report of the Consultative Steering Group. Edinburgh: Stationery Office.
  • Tyler, T. (1994) ‘Governing amid Diversity: The Effect of Fair Decisionmaking Procedures on the Legitimacy of Government’, Law and Society Review, 28 (4), 80931.
  • Tyler, T. (1997) ‘The Psychology of Legitimacy: A Relational Perspective on Voluntary Deference to Authorities’, Personality and Social Psychology Review, 1 (4), 32345.
  • Tyler, T. (2000) ‘Social Justice: Outcome and Procedure’, International Journal of Psychology, 35 (2), 11725.
  • Verba, S. (2006) ‘Fairness, Equality and Democracy: Three Big Words’, Social Research, 73 (2), 499540.
  • Whiteley, P. and Seyd, P. (2002) High Intensity Participation: The Dynamics of Party Activism in Britain. Ann Arbor MI: University of Michigan Press.
  • Winetrobe, B. (2001) Realising the Vision: A Parliament with a Purpose. London: The Constitution Unit.