Virtual Town Meeting Dissertation

Overview: Virtual Town Meeting

Dr. Peter Frederick Sielman

Dr. Peter Frederick Sielman is credited as the brainchild behind the [Salem, Connecticut] Virtual Town Meeting and is the author of the following text and supplemental media. The following is a draft of Dr. Sielman’s thesis for his Doctoral degree in Political Science at the University of Connecticut, and the final copy of his dissertation can be viewed and purchased from the University of Connecticut virtual library here.

Abstract

The concept of virtual town meeting permits eligible voters to participate in the legislative function of town meetings without the necessity of being physically present. Remote participation is via electronic means, using live cablecasting and webcasting of the physical town meeting and live emailing to permit remote participants to be part of the deliberations and to vote on motions. ^ This dissertation reports on tests of the virtual town meeting concept carried out in Salem, Connecticut in the period 2008–2010. It describes the electronic implementation hardware and software employed in the tests and the processes used to assure the smooth and secure functioning of the virtual town meetings. ^ This dissertation also reports on a case study survey of eligible town meeting participants to ascertain the distribution of feelings of political efficacy amongst various groups, age cohorts and genders. The results of the survey are compared to NES data cited by Niemi, Craig and Mattei. Conclusions are drawn with regard to how to measure and index political efficacy data, the correlation between the availability of remote town meeting participation and feelings of political efficacy and the potential role of virtual town meetings in future evaluations of participatory and deliberative democracy.

Citation

Sielman, Peter Frederick, “Virtual town meeting” (2010). Doctoral Dissertations. AAI3420185.
https://opencommons.uconn.edu/dissertations/AAI3420185

Chapter 1: Introduction

This is a study of the relationship between participatory and deliberative democracy and political efficacy as applied to the town meeting form of government. It is unique because it documents and examines the first example of a “virtual” town meeting in which citizens can participate remotely in a legally binding municipal town meeting. Given this uniqueness, I review the relevant literature in the context of the tests and associated case study that I have conducted that were focused on two groups of participants. The two groups are defined by the nature of their participation. One group attends town meeting in person; the other group is able to participate remotely in the same town meeting via electronic means. Both groups can listen to and see the deliberation. Under the control of the moderator, both groups can participate in the deliberation and propose motions consonant with the call of the meeting. Both groups can vote on motions raised at the town meeting. This form of town meeting is called a virtual town meeting and is legal within the statutes of the State of Connecticut (Sielman 2006).

Virtual town meeting provides remote participation through the application of simple, standard, widely used electronics. The meeting is viewable in real time[1] and, with the utilization of 2 cameras, the remote participant may actually have a better view of attending speakers than the remainder of the attendees. Remote participants contribute to the deliberations and vote on motions through the use of email. The moderator controls who “speaks” (both in person and remotely) by having a laptop computer that displays the incoming email traffic.

The primary objective of the virtual town meeting tests was to determine whether or not providing the possibility of remote participation would increase the number of people who do participate. Virtual town meeting is not aimed at getting all eligible voters to participate in the town meeting’s legislative function, but rather its aim is to eliminate some of the perceived barriers to participation. The tests were aimed at verifying the results of a previous survey (Sielman, 2006) that predicted that 5 % of the eligible voters who had not participated in town meetings might do so if it could be done remotely. For security purposes, the virtual town meeting process requires potential remote participants to register their unique email address. Of those who have signed up at this writing, 12% participated in the last virtual town meeting.

A secondary objective of the virtual town meeting tests was to measure the political efficacy feelings of participants, both town meeting attendees and remote participants. The measurement mechanism was a 19-question survey modeled after the NES political efficacy surveys. Analysis of the survey data compared respondents on the basis of their age, gender and the circumstances under which they signed up for potential virtual participation. Those who remotely participated in the last virtual town meeting and had filled out earlier survey questionnaires were asked to do so again after their remote participation to enable a PrePost test comparison of whether or not their remote participation altered their feelings of political efficacy.

The details of the tests, a description of the different hardware implementations and the processes used to conduct virtual town meetings are presented in chapter 3. This chapter also provides the numerical participation numbers achieved in the tests that have been run to date. Chapter 4 presents and analyzes the survey structure, compares the responses to those measured by Niemi, Craig and Mattei (1991) and plots the variations of responses with age, gender and survey venue. This chapter also presents and analyses the PrePost test data and draws conclusions about the relationship between participation and feelings of political efficacy.

Chapters 3 and 4 represent the meat of the study. They are preceded in Chapter 2 by a review of the scholarly literature germane to participatory democracy, deliberative democracy, the town meeting form of government, political efficacy and the application of technology to these fields. In Chapter 5, I summarize the conclusions that I believe can be fairly drawn from the test and survey, and their relationship to the literary background. Virtual town meeting is in its infancy. In the context of the ever increasing use of the Internet, it is my conviction that this concept will evolve, not only in its application to towns that have the town meeting form of government, but that we will also probably see its application to other implementations of participatory and deliberative democracy.

Chapter 2: Background

Town meeting is a form of deliberative and participatory democracy. There has been a large outpouring of scholarly literature on deliberative democracy and participatory democracy in the past three decades. Scholars disagree on both the merits of deliberative democracy and participatory democracy and the potential tensions associated with them, as well as about the methods that are most suitable for scholarly research in these fields. The aim, here, is to identify and evaluate the portions of this corpus that are relevant to a virtual town meeting and to concentrate on those sections of the literature that address political efficacy effects on the participants.

The premier authors who have studied and reported on the town meeting form of government[2] are Mansbridge (1980), Zimmerman (1999 & 2002) and Bryan (2003). Mansbridge reported in detail on the annual town meeting of a Vermont town with a population of 350, emphasizing the correlations between attendance and speaking at town meeting with measurable characteristics such as age, gender, length of residence in town and socioeconomic status. Her correlations were based on interviews/questionnaires with 69 residents (1980, 99). Mansbridge selected a Vermont town rather than a Massachusetts or Connecticut town for her study because the latter: “…were run by an obvious professional and business elite” (1980, viii). By contrast, this study is based on the Town of Salem, Connecticut with a population of approximately 4,000. After more than 20 years of attendance at Salem town meetings (of which there are several every year) it is not clear to me that the town meetings are run by “an obvious professional and business elite.”

Zimmerman (2002) concentrates on the current (and growing) problems associated with reduced participation in town meetings and offers up several strategies that have been used to increase participation in order to enhance democratic legitimacy. These strategies include child and elder care provision, transportation assistance, mailing meeting warrants to all households, using a consent calendar to speed up meetings, seeking citizen input in warrant generation, dividing a single meeting into two or more meetings and allowing proxy votes (1999, 175-179). While Zimmerman examined many strategies, he did not include remote participation via electronic means, the approach used in a virtual town meeting.

Bryan, who claims the inspiration for his work came from a meeting with Mansbridge (2003, ix), reported on data collected by his students at town meetings all over Vermont for over 30 years covering more than 1435 separate sessions of town meetings. Bryan, who is very enthusiastic about the institution of town meetings as the purest form of ”real democracy,” attempts to correlate attendance at Vermont’s annual town meetings with town population, affluence, growth rates, relationship to school districts and contention level of the items on the meeting call. He concludes that no single variable is a sure surrogate for attendance although town size affects overall attendance numbers but not percentage of eligible participant attendance. However, he states that: “beyond town size, issues are the single most important variable that draws citizens to town meeting” (2003, 233).[3]

While these three authors have done the most sustained research on town meeting, there are many others whose work is relevant to this study. For my purposes, the four most important areas of research examine: participatory and deliberative democracy, implementation schemes for increasing opportunities for participation and deliberation, the impact of modern communications technology,[4] and political efficacy. I address the literature associated with participatory democracy combined with the literature associated with deliberative democracy because town meetings represent a combined form of participatory and deliberative democracy. While one can have participation without deliberation—for example many of the traditional measures of political participation such as voting, fund-raising, contacting a representative, and campaign activity do not necessarily incorporate deliberation—the obverse is not generally true. Some of the literature addresses only participation and some addresses both participation and deliberation, but I extract from both the issues relevant to my research.

Participatory and Deliberative Democracy

Although some authors deal separately with deliberation and participation, in the context of a virtual town meeting, participation and deliberation are inextricably intertwined. To the extent possible, in reviewing the germane literature, I have organized the material to sequentially address:

1. The perceived advantages and disadvantages of participatory and deliberative democracy [Values].
2. Evaluation of practices that enhance participatory and deliberative democracy [Practices].
3. Proposals for effective political science studies of participatory and deliberative democracy [Studies].
4. Schemes that have been proposed to widen the use of participatory and deliberative democracy [Dissemination].

I am fortunate that two distinguished scholars, Diana Mutz and Dennis Thompson, have recently published broadly applicable survey articles on participatory and deliberative democracy (Mutz 2008; Thompson 2008). Each of these articles addresses the other and provides references to most of the relevant scholarship. Both of these articles reference the review of the relevant literature produced by Delli Carpini, Cook and Jacobs (2004), which covers “Public Deliberation, Discursive Participation and Citizen Engagement.” With these major guides to the corpus, I am able to extract the theoretical and empirical highlights of deliberation and participation that are germane in the context of a virtual town meeting.

Values

Delli Carpini, Cook and Jacobs (2004) organize their review of deliberative and participatory democracy by defining public deliberation, listing the assumptions that underlie its purported benefits, and listing what we know and don’t know about the impact of public deliberation on civic engagement (divided into: political talk, social psychology, case studies and survey-based research, experimental research on public deliberation and the potential role of online deliberation). They accept Chambers’ definition of deliberation:

[D]eliberation is debate and discussion aimed at producing reasonable, well-informed opinions in which participants are willing to revise preferences in light of discussion, new information, and claims made by fellow participants…an overarching interest in the legitimacy of outcomes (understood as justification to all affected) ideally characterizes deliberation (2003, 316).

Unfortunately, they do not reference current town meetings and refer to the concept only at the beginning of the paper as “town hall meetings of colonial New England” as if they had died out with powdered wigs (2004, 315-316). This position is unpersuasive because virtual town meeting clearly fits within the confines of Chambers’ definition. The virtual town meeting is the legislative body of the town that discharges legislative business after engaging in debate and discussion aimed at producing reasonable, well-informed opinions. The legitimacy portion of the definition is especially cogent, since virtual town meeting also aims at arriving at justified decisions.

Delli Carpini, Cook and Jacobs go on to list five necessary characteristics of effective discursive participation (2004, 318):

1. Discourse with other citizens.
2. Discourse as a form of participation’
3. Related to formal institutions of civic and political life.
4. Can occur through a variety of media.
5. Focused on issues (local, national or international) of public concern.

Virtual town meeting meets all of these criteria. Citizens participate by engaging in discourse within a formal political institution, and in the unique case of the virtual town meeting, they do so through a variety of media. In the virtual town meeting characteristic (5) generally focuses on local concerns, but citizens occasionally pass “sense of the meeting” resolutions on topics of wider concern.

Beyond these definitional issues, Delli Carpini, Cook and Jacobs, referencing many other authors, summarize the benefits of public participation in deliberation. They argue that public participation can lead to empathy, to an expanded sense of interest in community benefits, and to exposure to “reasoned argumentation” (2004, 320). They also point out that participants learn deliberation skills, are more likely to become engaged in politics, and will likely have greater faith in democracy and the legitimacy of their form of government. Quoting Chambers (2003, 318) on whether or not deliberation changes minds, they conclude that under the right conditions it does and that it has the side benefit of promoting tolerance and encouraging a public-spirited attitude (2004, 320). Berry, Portney and Thomson have an even broader view than most authors of the benefits of participatory democracy. Their view is particularly applicable in community settings:

In a participatory democracy, individual action in the affairs of the community is a critical part of the way people mature… participation is a key to transforming individuals so that they develop bonds with their neighbors, come to understand shared values and take on new identities… Competence in political participation is something people learn through practice and experience (1993, 258).

The tests conducted as part of this study specifically examine whether the act of participating in virtual town meeting enhances the participants’ sense of political efficacy.

Surowiecki addresses decision-making by expanded groups by pointing out that whatever new members do know is not totally redundant with what everyone else knows and hence the diversity improves the decision making (2004, 31). Delli Carpini, Cook and Jacobs echo Surowiecki’s claim about collective decisions being superior because more information can be brought to bear (2004,327-8). They further claim that when people feel greater accountability for their decisions, they are more likely to be objective and unbiased. In virtual town meeting, because the participants are acting in an important official capacity that can have direct and significant effects upon their lives, they feel the accountability to which Delli Carpini, Cook and Jacobs refer. Delli Carpini, Cook and Jacobs also report on some research that aims to test the hypothesis that collective political decisions are superior to individual ones. The artificial tests provided groups of respondents with information about, and then asked them to discuss, three fictitious candidates before choosing for whom to vote. Their results were that information was not aggregated (2004, 332). The reality and the absence of artificiality inherent in virtual town meeting makes it a much better testing ground for the value of collective political decisions.

Virtual town meeting builds on the theories of authors like Barabas who claim that deliberation increases knowledge and alters opinions if participants hear diverse messages and are willing to keep an open mind (2004, 687). Cohen echoes and amplifies this when he says that: “We have a strong case for political legitimacy when the exercise of political power has sufficient justification” (1998, 222). Beyond increases in knowledge, Morrell addresses the attitudes that may result from deliberation:

Establishing processes that encourage all citizens to empathize with one another will result in deliberations characterized more by mutual respect, civic magnanimity, reciprocity and a commitment to continued deliberation. (2008)

Gambetta extends this idea in a different direction when he points out that: “…deliberation… spurs the imagination indirectly if it reveals that, on all known options, no compromise is possible, for this provides an incentive to think of new ones” (1998, 22). [5]

Verba, Schlozman and Brady have a more nuanced perspective on the relationship between participation and a sense of the common good. They caution that even though they were surprised by the extent to which respondents believed their actions were motivated by common good:

[W]e would not presume to assess the extent to which the participatory system …achieves something that might be characterized as the common good. Even when the public good is conceived as something more than the summation of individual preferences, there are competing conceptions of the good of all… This does not mean that they do not also use their participation as a vehicle for furthering their own narrow interests or … that they do not sometimes construe what is good for themselves as being good for the country. (1995, 528-9)

Throughout his book, Bryan (2003) is very positive about the benefits of participation for the participants in town meetings. He includes civics education, tolerance, forbearance and acquaintance with the rules of group deliberation. Gargarella (paraphrasing Pateman) agrees with Bryan in the sense that deliberation may also be important in “educating” people to act impartially and help people live with others (1998, 261).

Not all authors subscribe to these benefits. Delli Carpini, Cook and Jacobs relate that many citizens do not wish for, and indeed might react negatively toward, efforts to engage them more directly in political decision making through deliberation. Hibbing and Theiss-Morse (2003) are perhaps the most forceful detractors of the value of participation in decision-making, arguing that the paramount drive of most voters is conflict avoidance. They point to the conflicts that can evolve from face-to-face deliberation, selectively quoting Mansbridge (1983, 68, 69, 273, 276-7, 293 and 295) and arguing that some [emphasis added] voters prefer to place their governance in the hands of empathetic non-self-interested decision-makers (ENSIDs). They believe that some citizens take the view that “debate is not necessary or helpful” (Hibbing and Theiss-Morse 2003, 143). They express doubt about the notion that participation educates participants to practice greater tolerance[6]. Mutz agrees partially with Hibbing and Theiss-Morse when she writes:

My own observations generally concur with Hibbing and Theiss-Morse in some respects. …many people do not like conflict and prefer not to talk about politics with those who have conflicting views. But I diverge from their conclusion, which suggests that because people generally do not like it, we should not worry too much about whether it happens. (2006, 90)

Thompson is critical of some of the detractors of deliberative democracy on the basis of the methodology employed by them in reaching their conclusions. His point is that, based on isolated passages from various theoretical writings:

They …extract a simplified statement about one or more benefits of deliberative democracy, compress it into a testable hypothesis, find or (more often) artificially create a site in which people talk about politics, and conclude that deliberation does not produce the benefits the theory promised and may even be counterproductive. (2004, 498)

Although there is a dearth of proof for Hibbing and Theiss-Morse’s opinions, the important fact for virtual town meeting is that it is aimed at increasing participation, not seeking universal participation[7]. As underlined above, the fact that some citizens, or even many citizens, do not wish to participate in town meetings does not preclude the fact that there are some citizens who can contribute to and benefit from participation in deliberation and legislative voting via the mechanism of virtual town meeting. As pointed out by Rogowski in his discussion of Pateman’s work: “Those who decline to participate, or who participate only under duress or only perfunctorily, have made no commitment” (1981, 301). Surowiecki is also skeptical of trying to interest everyone in participating in deliberative democracy:

The ideal for “deliberative democracy” makes an easy target for criticism. It seems to rest on an unrealistic conception of people’s civic-mindedness. It endows deliberation with almost magical powers. And it has a schoolmarmish, eat your spinach air about it…. [people don’t] want to be told to take a holiday because it’s time to talk politics. (2004, 261)[8]

Practices

Bohman, who also provides a survey article of the deliberative democracy literature, draws conclusions about topics requiring careful study:

[D]emonstrating its feasibility and clearly understanding its limitations ultimately makes deliberative democracy a more, rather than less, appealing basis for genuine reform and innovation (1998, 423).

In addressing the social psychology research applicable to participatory and deliberative democracy, Delli Carpini, Cook and Jacobs, citing Mendelberg (2002) and others, arrive at the following [paraphrased] conclusions:

1. Face-to-face communications is the single greatest factor in increasing the likelihood of cooperation.
2. Talking allows group members to demonstrate their genuine willingness to cooperate and to determine others’ willingness to do so.
3. Deliberation helps them to see the connection between their individual interest and that of the group.
4. The group consensus that emerges from talk appears to lead to actual cooperative behavior, with more talk leading to more cooperation.
5. However, these studies cannot demonstrate that altruism (as opposed to self-interest) is the prime motivator for cooperative behavior (2004, 324).

The structure of virtual town meetings appears to be in line with that suggested by the social psychologists, particularly in that it makes decisions by majority rule rather than seeking unanimity.[9] Delli Carpini, Cook and Jacobs report on an artificial game theory experiment by Hibbing and Theiss-Morse (2002). They note that:

The Hibbing & Theiss-Morse experiments do not fully capture the action of public deliberation; for example, the interactions occurred only between subjects and decision makers rather than across subjects… Nonetheless these experiments add to the evidence that the positive impact of discursive participation is strongly context dependent and tied to both process and outcomes. (2004, 333)

Thompson lists standards of which we should be wary that sometimes are applied to participatory and deliberative democracy. These include consensus—which Mansbridge has pointed out is not necessary and may be counterproductive—and justice. As Thompson notes, “the empirical challenges of isolating the effects of the deliberation on justice … are formidable” (2008, 508). Barber goes further by arguing that: “where consensus stops… politics starts” (1984, 129).

One of the interesting points stressed by Delli Carpini, Cook and Jacobs is that the important way to improve deliberation is to make sure that the debate is complete and the public has an opportunity to be engaged. Increasing the level of participation through the mechanism of virtual town meeting fits Fishkin’s concept in the public engagement sense. It is the responsibility of the moderator to assure the completeness of debate. In referring to New England town meetings (putting Mansbridge’s 1983 work in context), Delli Carpini, Cook and Jacobs summarize Mansbridge’s distinction between unitary (consensus) and adversarial (majority rule) democracy. Mansbridge’s unitary democracy “is most effective when participants share common interests, and social bonds such as friendship” (2004, 329), and although virtual town meetings clearly possess these “common interests” and “social bonds,” they also adhere to Mansbridge’s dictum regarding adversarial democracy:

If there is no solution that serves everyone’s interest, more debate will not usually produce agreement, and it is often better to cut short a potentially bitter debate with a vote (1983, x-xi).

This is when a virtual town meeting moderator can accept a motion to “call the question.”

Thompson lists the empirical conditions that are necessary for good deliberation (2008, 509). He emphasizes that publicity (the requirement that the deliberative forum be open to scrutiny by citizens either directly or through the media) is vital (2008, 510). Virtual town meeting has, as a by-product, that it facilitates public scrutiny. Johnson, though, sounds cautionary notes. His careful argument for deliberation presents the important caveats that we must:

1. avoid being utopian;
2. not categorically exclude self-interest claims and the resultant conflicts;[10]
3. encourage communications to minimize devolution to bad outcomes; and
4. recognize that the structure of the deliberative forum affects success. (1998, 173-177).

It would appear that virtual town meeting is consistent with Johnson’s warnings. What is interesting about town meeting is that, even when proposals are primarily self-interested, the town meeting forum makes it necessary to couch the rationale in “common good” terms.

An aspect of public deliberation that has important consequences in the long run is the quality of the deliberation;[11] while quality of deliberation is an aspect of virtual town meetings that is worthy of future study, I will not address quality of deliberation in this study owing to the need for much more data[12]. Steiner et al. have made an important contribution to this subfield by establishing a Discourse Quality Index (DQI) along with examples of how to operationalize the DQI for analytical purposes. The DQI[13] constituents are:

- Participation: ……………….…Ability to participate freely in the debate.
- Level of justification: ………..Completeness of presented position rationale.
- Content of justification: …..…Location on the narrow interest to common good axis.
- Respect: ………………………..Do negative statements imply no respect, neutrality or disrespect?
- Constructive politics: ………..Is position strategic, an alternative or a proposed mediation? (2004 56-60)

Studies

Measuring the effects of participation in deliberative democracy has engaged the interest of several political science scholars.[14] Delli Carpini, Cook and Jacobs (2003) summarize their own look at (what they call) discursive participation surveys:

1. 4% of the adult public reported on having participated in on-line forums on public issues in the past year.
2. 24% had engaged in Internet or instant-message conversations about such issues.
3. 25% had attended a meeting to discuss such issues.
4. 31% had tried to persuade someone how to vote.
5. 47% had tried to persuade someone to alter their point of view on a public issue.
6. 68% had face-to-face or phone conversations about public issues at least a few times per month.

Putting Hibbing and Theiss-Morse’s opinions (2003) in perspective, Delli Carpini, Cook and Jacobs found that:

1. 19% of adults had not engaged in any of the (above listed) discursive activities in the past year.
2. 1% had engaged in all six (above listed) activities.
3. 58% had engaged in two or more of the (above listed) activities.
4. 36% had engaged in three or more of the (above listed) activities.

Considering these results in the context of attempting to increase town meeting participation by a few percent of the total number of eligible voters, leads me to conclude that the goals of virtual town meeting are achievable.

Mutz, in trying to narrow empirical deliberative democracy research, attempts to break out the constituent parts of deliberative democracy and the theorized benefits into separable “middle-range” approaches to deliberative hypotheses (2008, 530):

Figure 2-1: Mutz’s Requirements and Outcomes

The requirements of the left-hand column are mostly inherent in virtual town meeting. The accuracy of the information provided is not assured, but the participation of different interests is likely to question dubious information. Maintaining civility is one of the moderator’s duties. Virtual town meeting is public, group-based, face-to-face (plus cablecasting, webcasting and email), has equality of participation, often has disagreement with others, there is interactivity of discourse, a direct link to political action, it is binding, collective and has equal-status participation. The moderator is supposed to ensure reflection by summarizing the issues that have been raised prior to the start of voting. The right-hand column of Mutz’s table defines the desirable outcomes (most of which fall under the heading of political efficacy, discussed below); many of them are the targets of the case study survey questions. Consensual decisions are not a necessity in a virtual town meeting, and may not be desirable in all cases, as pointed out by Mansbridge. It is also not clear that it is possible to separate out the individual items in Mutz’s list in real-world virtual town meetings. Mutz is careful to point out:

The theories and evidence [that she reviewed] do not speak to whether deliberation itself is feasible, but rather to whether, even if they do manage to coax it into existence, its consequences are likely to be as advertised (2008, 533).

Her doubts about feasibility imply that she has not considered town meeting as a form of deliberative democracy. When virtual town meeting is coaxed into widespread existence, an object of future studies will be to measure at least some of consequences that have been theorized, although not advertised.[16]

Mutz is rightly concerned that, despite the “enthusiasm for deliberation” in current political science: “one would expect to know far more about when and why it works well to produce various outcomes” (2008, 535). Virtual town meeting aims to address this issue; however, it is not a test in which we can manipulate variables to ascertain the effect of that variable with all other variables held constant. Mutz has concerns with some proposed experiments about the need for too many necessary and sufficient conditions, which are each insufficiently well-specified concepts. She is also concerned about a lack of specification of the relationships among the parts comprising the deliberative whole (2008, 536). Meeting all of Mutz’s concerns is clearly beyond the scope of this study. Mutz also expresses a further concern that:

Attention has now turned to large-scale institutional implementations of deliberative practices… [T]hey are designed to spread an already accepted practice as widely as possible. I think this kind of action is premature. (2008, 536)[17]

I agree. I do not believe that the virtual town meeting test and its associated case study fall into this category but, instead, believe that they are in line with Mutz’s suggestions.

By contrast, Thompson[18] introduces his review of deliberative and participatory democracy by giving his perspective on the essence of deliberative democracy:

At the core of all theories of deliberative democracy is what may be called a reason-giving requirement. Citizens and their representatives are expected to justify the laws they would impose on one another by giving reasons for their political claims and responding to others’ reasons in return. (2008, 498)

Like Delli Carpini, Cook and Jacobs and Mutz, Thompson believes that:

[E]mpirical research … findings are mixed or inconclusive …the main reason for the mixed results is that the success or failure of deliberation depends so much on its context…. If only theorists can identify the right conditions, they can confidently continue to extol the virtues of deliberative democracy. (2004, 499-500)

It is the goal of the virtual town meeting research to provide the context in which limited empirical conclusions can be fairly drawn without concern for artificial settings, non-representative groups or manufactured incentives, and it should address Thompson’s fears that “the conditions under which deliberative democracy thrives may be quite rare and difficult to achieve” (2004, 500). Thompson further itemizes three methodological problems that need to be addressed, each of which he then goes on to discuss in detail (2004, 500):

1. Distinguishing the analytical elements of deliberation- its concept, standards and conditions.
2. Structural criteria: recognizing that the conditions that promote some values of deliberative democracy may undermine other values.
3. Examining the relationship between deliberative and non-deliberative practices in the political system as a whole and over time.

Connecticut town meeting, both because of its long history and the rules established by State statute, is an approach that meets Thompson’s elements concern that “better are those approaches that distinguish the definition from the evaluation of deliberation [the unit of analysis from the democratic quality]” (2008, 501). He goes on to break down the elements of deliberation analysis into (1) the conceptual criteria, (2) evaluative standards and (3) empirical conditions. Similar to Mutz, Thompson lists necessary conditions for effective deliberation as: a state of disagreement—virtual town meeting at least requires action to which all may or may not agree; a collective decision—a characteristic inherent in virtual town meeting; and the legitimacy of the decision—virtual town meeting has the legal legislative responsibility.

Thompson lists some “standards” that should be used to evaluate the success of deliberative discourse, including public-spiritedness, equal respect, common good, information quality, finding common ground, and equality of participation (2004, 506). He adds the caution that “one of the most consistent empirical findings is that unless special measures are taken, membership and participation are likely to be significantly unequal” (2008, 506). For the set of all eligible voters in a town meeting form of government, this warning is applicable. Participation in town meeting (standard or virtual) is voluntary. Only the willing will participate. The virtual town meeting test is designed to evaluate whether more eligible voters will participate and the case study is designed to measure what, if any, political efficacy effects are produced by making town meeting participation available to a larger group of electors.[19]

Dissemination

Because it is widely felt that participation by citizens in the workings of government beyond the periodic casting of votes is essential for our democracy, many proposals have been put forward to produce greater participation in deliberative democracy. Thompson suggests spreading participation by posing ways of re-structuring deliberation (2008, 514):

1. Distributed deliberation- spreading functions over different institutions.
2. Decentralized deliberation- having several groups deliberate.
3. Iterated deliberation—splitting between policy reviewers and policy implementers.[20]

All of these structures are aimed at breaking up the task amongst more manageable groups. Since virtual town meeting is manageable, unless many hundreds of people want to actively participate in the deliberation, none of these structures is applicable to the town meeting legislative function.

One of the best-known attempts at dissemination is Fishkin’s Deliberative Polling. Fishkin describes his suggestion for disseminating deliberative democracy as follows:

The idea is simple. Take a national random sample of the electorate and transport those people from all over the country to a single place. Immerse the sample in the issues, with carefully balanced briefing materials, with intensive discussions in small groups, and with the chance to question competing experts and politicians. At the end of several days of working through the issues face-to-face, poll the participants in detail. (1995, 162)

Referring to Fishkin’s deliberative polls, Delli Carpini, Cook and Jacobs conclude that not all responses to the deliberative poll have been positive. The durability of changes in attitudes, opinions and knowledge, and the practicality of the design as a means of increasing deliberation among the larger population have been challenged (2004, 334).

Virtual town meeting does not aim to include all of the public in the participatory deliberation process. Rather it aims to reduce the costs in the electors’ cost benefit ratio by eliminating obstacles to participation and by permitting tentative participants to contribute to the deliberation under less threatening conditions. As Wolfsfeld states:

Potential participants consider both the likelihood of success as well as the value of such success when deciding both if and how to act… The final decision, however, is most accurately explained by their own individual cost-benefit analysis. (1986, 125)

Another thread that is prominent in the literature is that technology can play a role in facilitating participatory democracy. Deliberative Polling, Deliberation Day and Citizen Assemblies are examples. Virtual town meeting applies very mundane technology (cable television, webcasting and email) to permit willing participants to participate remotely.[21] There are many current attempts to use technology as a way of enhancing deliberation and participation that are much more ambitious than virtual town meeting. Fishkin (1995), Gastil (2000), Fung (2007a) and Cohen (1998) are some of the leaders in the field.

Delli Carpini, Cook and Jacobs raise the question: “Can online deliberation capture the experience and benefits of the face-to-face ideal” (2004, 335)? They quote Iyengar (2003) as saying that it produces “roughly parallel results.” Barber’s fear that technology would focus only on defense and profitable commerce has not come to pass (1998, 80). Gastil, in discussing the technology of computer-mediated deliberation, draws the conclusions that:

1. Face-to-face may be more appropriate in the political arena.
2. Political deliberation usually involves diverse and unacquainted participants.
3. Computer-mediation can reduce the independent influence of social status (2000, 359).

Also, on the technology front, Shane and his associates have developed software they call “Delibra” (2007). It allows a networked group to communicate with audio and text; participants see themselves depicted around a conference table. Those who wish to speak press a button and are put in the queue. Meanwhile, they are able to ask questions and register responses with text messaging. Virtual town meeting differs from this kind of online participation in that the remote participants “see” the physical town meeting, can verify that their email inputs to the deliberation are displayed and/or read to all participants, and get feedback on the results of their votes. They feel like they are part of the town meeting. It may not be exactly like face-to-face deliberation, but it is certainly better than just watching or just reading about the results in the newspaper.

E-voting, as a substitute for in-person voting and mail-in voting, is practiced in Geneva, Switzerland. Chevallier (2009) has reported on the effects of this form of voting on the turnout at elections and found that it eliminates some of the “invisible barriers” to participation. [22] The techniques used by the Canton of Geneva to ensure the security of the electronic voting procedures are similar in outline to those used in the virtual town meeting.

Political Efficacy

The literature that addresses political efficacy includes one or more of the following topics: (1) definition of the concept; (2) measuring its level; (3) some generalized results; and (4) potential pitfalls in its evaluation. Political efficacy is a relatively new term but not a new concept, with a current consensus as to what the term means and fair agreement on how the concept can be measured through surveys. Wollman and Stouder define the most general forms of efficacy as follows: “Personal efficacy relates to the belief that one can personally affect events, whereas societal efficacy relates to the belief that people in general can affect events” (2001, 562). Thompson lists the items he believes should be included as constituents of political efficacy: (1) increase in political knowledge, (2) political views of other participants and their reasons for holding those views, and (3) the sense of the legitimacy of the outcome (2008, 507). The term ‘political efficacy’ was defined by Campbell, Gurin and Miller as:

“the feeling that individual political action does have, or can have, an impact upon the political process, … the feeling that political and social change is possible, and that the individual citizen can play a part in bringing about this change” (1954, 187).

In the current political science literature a distinction is made between internal and external political efficacy. Acock defines the distinction as: “[I]nternal efficacy indicates individuals’ self-perception that they are capable of understanding politics and competent to participate in political acts…external efficacy measures expressed beliefs about political institutions” (1985, 1064). Niemi, Craig and Mattei developed and refined the internal and external political efficacy questions adopted by the NES and used by many scholars. Their four internal efficacy survey questions are (1991, 1408):

1. I consider myself to be well qualified to participate in politics.
2. I feel that I have a pretty good understanding of the important political issues facing our country.
3. I feel that I could do as good a job in public office as most other people.
4. I think that I am better informed about politics and government than most people.

The three external efficacy survey questions are:

1. Sometimes politics and government seem so complicated that a person like me can’t really understand what’s going on.[23]
2. People like me don’t have any say about what the government does.
3. I don’t think that public officials care much what people like me think.[24]

The exact wording of these standard NES questions about both internal and external political efficacy are not directly suitable for the virtual town meeting case study because they do not specifically relate to the participation in town meetings. Instead, the survey questions in the case study are tailored to political efficacy in the context of the virtual town meeting and local government considerations.[25]

With regard to the correlation of political efficacy with other characteristics of citizens, Finkel reports that voting and campaign participation had external efficacy effects but little internal efficacy effect (1985, 907). Michelson believes that regional and cultural differences provide different efficacy responses (2000, 147). Morrell reports that, based on his analysis of several data sources, “Internal efficacy relates strongly to psychological involvement; moderately to participation, education, and external efficacy; and not at all to political trust” (2003, 597).

The difference between participating in a town meeting deliberation and merely voting in a referendum may have a different political efficacy effect on electors. In his own experimental study, Morrell concludes that citizens who engage in face-to-face deliberative decision-making, rather than just thinking about and voting on issues, will likely have higher levels of internal political efficacy in those specific face-to-face venues (2005, 63). Iyengar believes that, unlike political trust, the sense of external political efficacy does not appear to be closely intertwined with evaluation of the incumbent government: “Subjective efficacy is not a fleeting response to current political realities but is instead a more firmly embedded attitude concerning the responsiveness of the regime” (1980, 155). Attaining increased political efficacy can be considered as a form of education, or at least, the diminution of ignorance. It may increase or decrease the perception of the quality of governance being provided but it surely increases “individuals’ self-perception that they are capable of understanding politics and competent to participate in political acts” (Acock 1985, 1064).

Caution in the process of evaluating political efficacy is a common thread that runs through the writings of Delli Carpini, Shapiro, Thompson and Mutz. The questions used to measure it are indirect, open to varied interpretation, sometimes misread and limited by the choices of answers available to respondents. Reading too much into the statistical analysis of political efficacy surveys is discouraged. By contrast, Rosenstone and Hansen present (without reference) very precise numbers about the benefits of political efficacy (2003, 79):

1. People who feel a sense of personal competence to figure out what is going on in local politics are 5.4 percent more likely to attend local meetings.
2. People who believe that their actions will affect what local government does are 3.2 percent more likely to attend.
3. Likewise, people who believe that their actions will make the government respond are 5.7 percent more likely to write to public officials.
4. 2 percent are more likely to sign petitions.

Conclusions Drawn From The Literature

The primary conclusion I draw from this review of the political efficacy, participatory and deliberative democracy literature is that it appears that virtual town meeting could provide a laboratory for empirical testing of the underlying political theories about deliberation and participation and their combined effect on political efficacy. My hypothesis that virtual town meeting has some positive effects on political efficacy is neither denied nor strongly supported by the data that has been collected to date, but the results do provide some useful empirical data.

The virtual town meeting setting appears to meet all of the structural criteria for meaningful testing of deliberative and participatory democracy theories put forth by Delli Carpini, Cook and Jacobs, Thompson, and Mutz. It is a natural setting, with voluntary participation in a civic duty that has real consequences for the participants. Virtual town meeting can provide the deliberative and participatory quality that is essential for the generation of good empirical data.

The most useful part of the literature is the detailed reviews provided by Delli Carpini, Cook and Jacobs, Mutz, and Thompson. Delli Carpini, Cook and Jacobs summarize their findings in five areas:

1. Americans engage in public talk.
2. Social Psychology research indirectly provides support that deliberation can lead to some of the individual and collective benefits postulated by theorists.
3. Case studies, surveys and quasi-experiments lead to similar conclusions.
4. The Internet may prove a useful tool in increasing opportunities for utility.
5. Most importantly, the impact of deliberation is highly context dependent. It depends on the purpose of the deliberation, the subject under discussion, who participates and the connection to authoritative decision makers (2004, 336).[26]

They believe that “we should take great advantage of the myriad real-world deliberative experiments that occur every day” (2004, 335). Virtual town meeting is one such candidate.

Virtual town meeting is a potential way to increase participation in town meeting. A potential by-product of increased participation is that political efficacy may be expanded to cover more citizens. The aim is not to try to extend it to all eligible citizens. If ultimately virtual town meeting is shown to lead to increased political efficacy for the remote participants and/or those who attend, then the question will arise whether the benefits could be made applicable in non-town meeting settings. Fung (b) considers how you would set up Minipublics as a way of implementing town meeting type of forums for participatory deliberative democracy (2007,156). Fung (a) also believes that there should be a comprehensive national system of local assemblies in every rural, suburban and urban place (2007, 450). The advantage of virtual town meeting is that one does not need to invent artificial settings and means in order to analyze participatory deliberative democracy. One may consider the alternative of devolving legislative prerogatives from the States to local municipalities and neighborhoods. Local control of schools, open space planning, police and fire protection, local taxation, local zoning and wetlands preservation are all potentially devolveable to town meeting types of legislative control.

One of the recurrent themes in the literature is caution. One should take small steps, have low expectations and be skeptical of the validity of apparent empirical results. One should also not set expectations too high in that, as Wollman says: “[T]he fact that people in general feel that they can be effective by taking a particular action does not necessarily imply that they will act” (2001,565). A related thread is that some authors appear to consider deliberation and participation as the answer to all of our democratic shortcomings. Sanders somewhat addresses this from her elite perspective:

Average people can be improved in a number of ways through their involvement in politics. Not only might they develop basic competency at citizenship, they also are likely to become better human beings, acquiring both individual autonomy and a sense of common involvement. Many contemporary democrats extend this hope to all citizens: they want everyone involved in politics, but they want everyone to deliberate about it. (1997, 359)

On the positive side of the ledger, virtual town meeting addresses Ryfe’s requirement that “we must learn what deliberation actually looks like… by investigating deliberation in the natural political contexts in which it takes place” (2005, 64).

With regard to the questions raised about how the discipline ought to address deliberative and participatory democracy, Thompson makes the plea (also voiced by Delli Carpini, Cook and Jacobs and Mutz) that:

The division of deliberative labor [between theorists and empiricists] may or may not serve the practice well… But even with the division of labor that is likely to persist, collaboration can still go forward if theorists and empiricists engage with each other’s work…theory and empirical research might then more often progress hand in hand. (2008, 516)

That seems like the hard way to accomplish an important task. Better government is a pragmatic task. [27] Allowing arbitrary field specialization to interfere with progress towards this vital goal seems counterproductive.

I hope that I have made a reasonable case for town meeting as a logical setting to investigate participatory and deliberative democracy. And yet, town meeting is not mentioned at all in most of the literature; when it is mentioned it is either misidentified or mentioned to put Mansbridge’s work in context (without providing any details). Hibbing and Theiss-Morse revile town meetings for their “vitriol” (2003, 241) without referencing their source. And, none of the authors cite either Bryan or Zimmerman, who have written excellent works about town meeting. I find this puzzling, and hope that the report on this field test and case study will underscore their important contributions to participatory and deliberative democracy scholarship.

Chapter 3: The Tests

The objective of increasing town meeting participation coupled with the availability of live cable broadcasting (cablecasting) and citizen’s wide use of email communications led to the initial test of remote live town meeting participation (virtual town meeting) in 2008. Remote participants could watch the proceedings on their television sets and could provide written inputs (comments, questions or points-of-order) and vote via email.

As a precursor to the initiation of virtual town meetings in Salem, my Masters Thesis (Sielman, 2006) investigated whether or not eligible voters who do not currently participate in town meetings would do so if they could participate from home, a business trip, college or while serving in the armed forces. In 2007, based on that survey’s results[28], I proposed virtual town meeting tests to the Salem Board of Selectmen[29]. Included in the presentation to the Board of Selectmen was the Town Attorney’s ruling that a virtual town meeting would be legal in the State of Connecticut based on the Freedom of Information Act which explicitly recognizes that a meeting can take place “by means of electronic equipment”, and Section 1-200 of the General Statutes which grants the Town’s explicit authority to adopt rules of order for the conduct of its meetings, coupled with the State Constitution’s recognition that the people have an undeniable and indefeasible right to alter their form of government in such manner as they may think expedient (Article 1, Section 2). Further, Connecticut General Statute §7-5 says, “the place of holding town meetings may be determined by a majority of the voters present and voting at any town meeting specially warned and held for that purpose”. The Board of Selectmen approved the virtual town meeting tests, provided that there would be no costs[30] to the Town, and brought the matter up at the May 2008 annual town meeting which initially[31] approved three virtual town meetings.

The Application of Electronics

In the period 2001-3 local cable franchises were awarded to cable companies in Connecticut in return for which the cable companies provided studio facilities, equipment and training in the use of video equipment to facilitate public access to locally generated material, including educational and governmental materials. The cable company (Comcast) provided a free video uplink facility (including some electronics) at the Salem Town Office Building (TOB) as part of a party line cable franchise that also includes the towns of East Haddam, Haddam Neck, Lyme, and Old Lyme. The facility in the Salem Town Office Building initially included a choir microphone hung centrally from the ceiling of the large conference room (capacity = 75), a tripod mounted camcorder and a VCR recorder/monitor. This equipment suite is shown schematically in Figure 3-1. With this facility the Town of Salem was able to broadcast, live, town meetings and Board and Commission meetings from the Town Office Building to homes that had signed up for cable service (approximately 60% of Salem households are cable subscribers). These broadcasts have continued to this date.

Figure 3-1: Cablecasting from Town Office Building

In order to test the feasibility and utility of remote participation in town meetings, the equipment at the Salem Town Office Building was augmented with a second tripod-mounted camcorder (to permit separate video of the moderator and audience participants), a laptop computer equipped to receive email; a large projection screen (to permit audience viewing of spreadsheets/PowerPoint presentations and incoming emails; a second monitor to view all potential video from which the operator can select what to transmit; and a switcher to select moderator the video source for transmission via cable. A second slaved laptop was provided for the to select incoming email messages. This equipment suite is shown schematically in Figure 3-2.

Feedback received from the initial virtual town meetings, based on cable transmission for outgoing video (and audio) and email for incoming questions/comments and votes, highlighted the following shortcomings of the system:

1. Only 60% of Salem households (that possess cable) could participate.
2. Both a computer (~90% of households have computers) and a television set with a cable source were required to participate. Even where both existed in a household, they were often located in different rooms.
3. Potential participants away on business trips, away at school or serving in the armed forces could not participate.

Figure 3-2: TOB Equipment

Based on this feedback, it was decided to augment the system by providing webcasting of the video. This entailed the addition of a server for uploading video; a device for converting from analog to digital signal format; some software (Wirecast); and the services of a video provider (Livestream, who provides basic Live and On Demand video free of charge in return for periodic advertisements).

Figure 3-3: TOB Upgraded for Webcasting

This expanded system equipment suite is shown schematically in Figure 3-3. A domain name of www.salemct.org was reserved for the webcasting activities. With both the input (video) and output (email) on a single computer, the popular way to participate remotely is to place both the website and the email software simultaneously on a horizontally split monitor.

It was further desired by the Board of Selectmen that meetings that are too large for the Town Office Building, and are consequently held at the School or Firehouse, should be capable of being run as virtual town meetings. Since the cable provider (Comcast) set a price of $90,000 for a video feed at the School or Firehouse, it was decided to forego the cable casting and to configure a webcasting-only (outgoing)-email (incoming) virtual town meeting capability at the School. The School is centrally located and is familiar to the many families who have (or had) children that matriculated through it. This all-digital capability is depicted schematically in Figure 3-4.

Figure 3-4: Mobile All-Digital Remote Participation Equipment

A picture of this mobile facility (capable of being used throughout the school) is shown in Figure 3-5.

A characteristic of the assembly of hardware and software for the virtual town meeting tests was the necessity to provide the equipment without the expenditure of significant funds. Likewise, the operation of the equipment, both during tests and actual virtual town meetings, was dependent on volunteer help[32]. The funding limitations also affected the choice of transmission quality resulting in periodic video dropouts on the webcasting uplink[33]. A further equipment modification is undergoing development because not all of the attendees at town meetings heed the moderator’s exhortation to “stand up and speak up” when asking questions or making comments. As a result they are not always readily understood by the remote participants. An approach that is being developed to address this problem is the use of a parabolic microphone antenna that is slaved to the camcorder which pans to and zooms in on attendees when they are speaking.

Virtual Town Meeting Processes

In addition to the provision of hardware and software to enable virtual town meetings, it has been necessary to develop operational processes. The processes that we have evolved for virtual town meetings are in response to a number of requirements that must be fulfilled if the virtual town meeting is to achieve its goals. Requirements that we have set include:

1. The system must be easily utilized by those who choose to participate remotely.
2. The system must have a minimum adverse impact on those who attend the virtual town meeting in person.
3. The system must be protected from fraud by excluding ineligible participants.
4. The system must cast out multiple votes from a single validated email address.
5. The system must provide transparency so that attendees will be aware of the number of remote participants.
6. Since Internet voting inserts a delay (of up to approximately 90 seconds) the votes of attendees and the votes of remote participants must not influence each other through non-simultaneity.
7. A record of remote participant inputs must be created and stored.

Easily utilized by remote participants

More than 85% of Salem residents have and use email[34]. That percentage is far greater than the number of voters who can attend an evening meeting on any given date set by the Board of Selectmen. Hence, the use of email in virtual town meeting participation represents a net increase in the potential number of town meeting participants. It does not cover 100% of potential participants but does meet the virtual town meeting goal of increasing town meeting participation. However, if the process is complicated, cumbersome or protracted, new remote participants may fear getting involved. If the system is complex, inexperienced remote participants are likely to make errors and potentially invalidate their vote or prevent their comment or question from being a part of the deliberation process. Therefore, the process that we have evolved is simple, easily understood and requires a minimum of actions on the part of the remote participant.

To simplify the remote participation process and eliminate the need to remember the appropriate e-mail address, the system sends out five emails to each registered participant on the day of the meeting. These blank emails have subject lines that say, respectively:

- Present (this email also contains the town meeting agenda)
- Speak
- Point-of-Order
- Yes
- No

At the beginning of the meeting, the moderator asks all of the remote participants to signify their participation by selecting the Present email and then clicking on “Reply” followed by “Send”. The computer automatically counts the number of remote respondents[35]. The number is displayed on the moderator’s laptop and reported to the attendees by the moderator.

Figure 3-6: Present Email Response

During deliberations, remote participants who wish to submit a comment, question or point-of-order select either the Speak or Point-of-Order email (as appropriate[36]) and then click on “Reply”, type their name and message in the body of the email, and then click on “Send”. The moderator has access to a display (on his laptop) that shows the number and source of remote participants, and uses this information to select inputs to the town meeting.

Figure 3-7: Use of the Speak Email

During virtual town meeting votes, the moderator will instruct remote participants when they should send in their votes. To vote in favor of the current motion, a remote participant selects the Yes email and then clicks on “Reply” followed by “Send”. To vote against the current motion, a remote participant selects the No email and then clicks on “Reply” followed by “Send”. The computer automatically counts the votes[37]. Once both local and remote votes are in, the moderator announces the result. A sample “Yes” vote is shown in Figure 3-8.

Figure 3-8: Email Voting Procedure

Minimum adverse impact on attenders

Inevitably, the moderator’s instructions to remote participants is an added function that non-virtual town meetings do not require and which could be viewed by some as an intrusion on a “traditional” town meeting. This must be weighed against the increased legitimacy of a town meeting that encompasses more participants. As the number of virtual town meetings increases and the number of eligible voters who participate in them increases, the need for detailed instructions for remote participants will diminish. As a general practice, most town meeting moderators instruct attendees on how to cast their vote and what both the ‘yeah’ and ‘nay’ votes mean. With virtual town meetings the moderator provides instructions to remote participants on how to let their virtual presence be known (email Present), how to raise their virtual hands during debates (email Speak or Point-of-Order) and how to vote (email Yes or No)[38].

Protected from fraud

The most frequently asked question about virtual town meetings is: How will you prevent unauthorized voters from voting or even disturbing the deliberation process? In conventional town meetings, unauthorized voting is most frequently predicated on the knowledge of the moderator for voice and show-of-hands voting. At large meetings, moderators will frequently ask those ineligible to vote to identify themselves. In the case of challenges or close vote, citizens can ask for a “checklist and ballot” where a Town official, using the voter registration list, verifies the eligibility of voters who then cast paper ballots[39].

In the case of virtual town meeting, the capabilities offered by the email service provider (Comcast) permit us to restrict incoming emails to a Trusted Address List (TAL). Only those who register their email address and who have been certified to be on the official Town voter registration list have their email address put on the TAL. Therefore, only emails from addresses on the TAL will be received in the Inbox of the virtual town meeting computer. Emails from other email addresses (not part of the Trusted Address List) are funneled into the “Junk Email” folder. This is not a foolproof system- it has been shown that if an individual knows the email address of a member of the TAL, it is possible, with some effort, for that individual to appear to be sending an email from the TAL member. To thwart this possibility, the messages that are sent out on the day of the virtual town meeting are coded. Thus, in effect, the Yes and No emails are individual (reusable) ballots sent only to valid remote participants. Since only legitimate TAL members receive these messages, and only replies with the coding are accepted, fraud has been made less likely.[40]

Cast out multiple votes

The one man- one vote principle is applied by requiring each member of a household that has multiple eligible voters to register a unique email address for each eligible voter. In cases where someone attempts to “stuff the ballot box” with multiple votes, the computer system funnels all of the Yes and No votes into separate folders. The folders are sorted by source, so that multiple votes from a common email source will be contiguous. Duplicated votes are deleted from the count[41].

Transparency

One of the hallmarks of town meeting is that it is a group process where people interact and change their positions based on what they hear there neighbors say. It is therefore important that the virtual town meeting attendees are fully aware of the remote participants and that the remote participants feel “a part of” the meeting. To that end the tests that we have run are based on three potential video outputs (selected, when appropriate, by the operator). One camera is focused on the moderator. A second camera (located near the moderator) is focused on speakers in the audience when they are contributing to the deliberation process. A third video source is the computer that serves the dual purpose of (1) projecting illustrative material (such as spreadsheets or diagrams), and (2) processing email inputs and projecting counts or individual remote participant inputs under the direction of the moderator. Thus, attendees are aware of who is participating remotely, what comments/questions or points-of-order are being made and what the vote tallies are. The remote participants receive the video and audio of the moderator and attendee participants and see the same displays that the local audience sees.

No influence on votes by other group

The Internet is not instantaneous. The time between transmission of an email and its reception varies, but is usually less than 90 seconds. The solution of our moderator has been to instruct remote participants to vote on each motion and wait 90 seconds before having the attendees vote. In this way, the votes of attendees will not influence remote participants. The votes of remote participants are not displayed until so directed by the moderator. Our moderator has used the 90-second interval to provide useful town information to both the attendees and the remote participants[42].

Record of remote participant inputs created

Although official minutes are created by the Secretary of the town meeting (appointed by the moderator at the beginning of the town meeting), these minutes are augmented by the recorded video that was transmitted during the virtual town meeting and by the stored record of all emails received during the meeting[43]. From the point of view of an accurate historical record of town meetings, the virtual town meeting, as we have evolved it, will provide a much richer source.

A further capability is that the video of virtual town meetings is stored on the www.salemct.org website and is available for ON DEMAND replay at any time. As a consequence voters who were not able to participate in the virtual town meeting, live, still have the ability to review it after the fact. This is particularly valuable for questions that must be adjourned to a charter mandated town-wide referendum because it provides an opportunity for referendum voters to educate themselves on the issues prior to casting their referendum vote.

Measurement Of Participation Impact

The goal of the virtual town meeting tests was to determine whether the removal of some of the perceived barriers to participation would indeed lead to an increase in participation. Anecdotally, people have emailed me, or said to me, that they participated remotely because they were unable to leave the house on the evening of a virtual town meeting. Figure 3-9 provides a summary of the participation numbers for the first 4 virtual town meetings that have been conducted up to this writing. In the table the third and fourth column titles have asterisks because these virtual town meetings covered only routine matters such as the approval of the sale of an old fire truck upon the delivery of its new replacement.

Figure 3-9: Participation Summary

Of significance in the table is the large number of attendees (130) at the January 2010 virtual town meeting whose call included the approval of further virtual town meetings, the acceptance of a grant and consideration of a major school renovation project. It was the last of these three items that brought out the largest attendance that I have encountered in 22 years of town meeting attendance[44]. Of significance in the third row is the steady, albeit slow, increase in the number of eligible remote participants in virtual town meetings. It is my rough estimate that approximately half of the eligible remote participants attended the January 2010 virtual town meeting. The following row provides the number of remote participants who responded to the moderator’s request to respond to the Present email. From the records of the webpage for the January 2010 virtual town meeting, we know that there were 40 “hits” on the website[45] during the time of the meeting. Some of these we know were non-eligible participants, but some were probably eligible to participate but chose not make their presence known.

The next to last row of the table shows the percentage increase of participation provided by the availability of remote participation. While the percentages for the first three virtual town meetings are larger, the most significant number is the 16% in the last column because the issues at that town meeting were far from mundane. This is reinforced by the last row that shows the percentage those who had signed up to potentially participate remotely that actually did so. Considering that approximately half of the eligible remote participants attended the meeting, one can reasonably double the 12% figure as representing the number of eligible participants who availed themselves of the opportunity to participate remotely.

Conclusions About The Tests

The hardware assembled for the tests of the virtual town meeting appears to work satisfactorily although there are some concerns still about the reliability of the School’s Internet uplink[46]. Clearly, with more available funds the hardware could be improved and the Operator tasks thereby simplified. But it does the job.

The processes that have been developed appear to work sufficiently well to give both the attendees and the remote participants the sense that they are in a common meeting. While the security measures are quite simple they appear to be adequate for the purposes of a town meeting and those participants who have asked about security seem satisfied with the explanations of the measures used. All of the feedback from the January 2010 virtual town meeting is detailed in Appendix C.

Although there has been some local quarterly publication news about virtual town meeting, there has been relatively little effort devoted to increasing the eligible voter enrollment numbers. It seemed prudent to first see if the tests would provide measurable positive results and whether the process and hardware bugs could be overcome, before a major effort to increase enrollment was undertaken. In my estimation, that point has now been reached.

Chapter 4: The Case Study

Overview

A second objective of the virtual town meeting study was to ascertain whether or not there was a measurable relationship between town meeting participation and individual feelings of political efficacy. The mechanism employed to address this question was a survey questionnaire. The opportunity to respond to the questionnaire was offered on several occasions. Not all respondents to the questionnaire signed up to participate remotely. Many, but not all, eligible remote participants filled out a questionnaire. Some respondents provided no identification thereby eliminating the possibility of following up with a post participation survey (which would complete the inputs required for a true PrePost survey). The result of the survey questionnaire is a set of data that provides insights into the relationship between participation and feelings of political efficacy, that correlates well with the data of Niemi et al, but that provides only sparse PrePost data that might establish the direction of the causal arrow between participation and political efficacy.

Survey Questionnaire

In this study the survey questions were answered by respondents checking one of 5 columns for each of the 19 questions[47]. In consonance with the NES political efficacy questionnaire, the columns were labeled (in order from left to right): “agree strongly”, “agree somewhat”, “neither agree nor disagree”, “disagree somewhat”, and “disagree strongly”. The 5-choice answer approach represents a compromise between a 3-choice regimen (”agree”, “no opinion”, “disagree”) whose interpretation is unambiguous and a graded choice (select a number between 1 an 10, with 10 representing “strong agreement” and 1 representing “strong disagreement” with an optional “no opinion”) in which the respondent can shade the level of agreement. The former lumps all shades of agreement (or disagreement) into a single category. The latter provides shading but poses the problem of consistency- if 2 respondents select an answer of “8”, do they really have the same level of agreement or are their shading scales different? The 5-choice answer approach compromises in a way that retains both the benefits and difficulties of the 3-choice and the10-choice approaches. All respondents may not have the same threshold between “disagree somewhat” and “disagree strongly”. The hope is that with a sufficient number of respondents the “noise” associated with this variability will tend to be cancelled out. A further potential source of “noise” is that the choice of “I neither agree nor disagree” can be selected in the case where the respondent has no opinion or in the case where the respondent does not understand the question. Some respondents handled this latter case by leaving that questionnaire row blank.

In accordance with the NES model, the three external political efficacy questions were negatively worded; that is, a response of “strongly disagree” represented the greatest level of external political efficacy. Likewise, one of the community-related (CO) questions and three of the town meeting specific (TM) questions were negatively worded, resulting in a total of 37% negatively worded questions[48]. This issue is highlighted because the questionnaires were generally filled out by respondents who elected to do so on the spur of the moment, sometimes under the time pressure of having to leave for another endeavor such as a job, or cooking dinner or picking up a child. There is some evidence (based on a lack of consistency) that the questions were not always read carefully and that the negative word (“not”) may have been missed. One respondent, probably out of haste, responded “agree somewhat” to all 19 questions.

Beyond this limitation of the questionnaire as a whole, the individual questions each present potential ambiguities that may explain the lack of full correlation between responses to the different questions. Only 5 of 159 respondents (3%) answered all 19 questions with “agree strongly” for positively worded and “disagree strongly” for negatively worded questions. The corresponding percentages for the 4 groups of questions are presented in the Table below:

Figure 4-1: Distribution of “Strong” Political Efficacy Responses

It can be seen from this Table that individual questions are triggering somewhat different aspects of political efficacy feelings in most respondents. Therefore, analysis of the individual questions in each group may be illuminating.

Internal Efficacy Questions:

For purposes of consistency with well-established practice, the wording of the internal efficacy questions used in the survey questionnaire is similar to those employed by NES with town (generic) or Town (specifically the Town of Salem, Connecticut) inserted where appropriate. As proof of their equivalence, consider the cross correlations between questions as presented by Niemi, et al and the study survey:

Figure 4-2: Correlation Comparisons Among Internal Political Efficacy Responses

It can be seen from the Figure (comparing the first two rows- Niemi, et al and all respondents) that the relative correlation values are similar although there is substantial difference among the different groups of respondents as seen in rows 3 through 6. This is shown graphically in the curves (below). It can be seen that the cross correlations between Internal Political Efficacy question responses of the Attenders (at town meeting) and the voters at the November Election closely parallel those of Niemi, et al, whereas the responses of the voters at Referendum and the responders to the advertisement have considerably lower cross correlations. It should, however, be remembered that the data were collected sequentially (Attenders è Election è Referendum è Advertisement) with respondents being asked to complete the questionnaire only if they had not done so previously. Therefore, it should not be inferred that all voters at Referendum are less consistent in their responses to internal political efficacy questions, but rather that those voters at Referendum who did not attend the virtual town meeting or avail themselves of the opportunity to sign up for virtual town meeting at the November Election[49], are less consistent than Attenders and the average of voters at Election who signed up for virtual town meeting. This group, and those responding to the Advertisement show substantial variation from the Niemi, et al data.

Figure 4-3: Internal Political Efficacy Cross Correlations Responses

Baseline Survey Analysis

Because the survey used differs from the conventional NES survey by focusing on local governmental function and differs in details of how the questions are phrased, this section addresses each of the questions and raises issues that must be taken into consideration when interpreting the results[50].

INT1: I feel that I could serve on a Town Board or Commission [PUBOFF][51].

On reflection, the wording of the questionnaire question and the NES question could be open to different interpretations. The NES question reads: ” I feel that I could do as good a job in public office as most other people”. This could be interpreted not so much as a personal political efficacy evaluation but rather as a poor evaluation of those who do serve in public office. The questionnaire question is much more directly personal. As can be seen from the table (below), a greater percentage of Attenders at the virtual town meeting “strongly agreed” with the questionnaire statement than the other groups.

Figure 4-4: “Strong Agreement” Distribution For PUBOFF

In part, this is true because many of them are current or past locally elected or appointed Board or Commission members. The NES question is hypothetical and relative to other voters, whereas, the questionnaire question is specific to the individual and represents a real option for the respondent[52].

INT2: I feel that I have a pretty good understanding of the important issues facing our town [UNDERSTAND].

The NES question reads: ”I feel that I have a pretty good understanding of the important political issues facing our country”. The questions sound very similar, but a response is dependent on the availability of information sources that describe the political issues facing the country (in the case of NES) and facing the Town (in the case of the questionnaire). The former are presented and debated in newspapers, on television and on the Internet, whereas the latter rarely make the papers or television news and are more dependent on inter-personal communications via meetings, telephone conversations or email. Thus, an understanding of issues facing the town requires the respondent to either be connected with others who understand the issues or to make an effort to seek out the information. On balance, a positive response to the NES question can be quite passive and a reflection of self evaluation, whereas the questionnaire question is a measure of how active the respondent is in educating them self on local political issues. Again, as can be seen from the table (below), the Attenders “strongly agree” at between 2 and 4 times the rate of the other groups of respondents. They make it their business to understand.

Figure 4-5: “Strong Agreement” Distribution for UNDERSTAND

INT3: I consider myself to be qualified to participate in Town affairs [SELFQUAL].

The NES question reads: ”I consider myself to be well qualified to participate in politics”. The only differences between these two forms of the SELFQUAL question is the NES reference to “well qualified” and “politics” vice the questionnaire reference to “qualified” and “Town affairs”. The NES question’s dependence on “politics” is open to interpretations as narrow as feeling qualified to vote periodically or as broad as feeling qualified to be a politician. Therefore, the same response by different respondents may represent vastly different feelings of self-qualification by the respondents. The questionnaire’s dependence on “Town affairs” participation is less likely to be interpreted as merely voting in national elections (Salem turnout > 80%) and more likely to be interpreted as serving on a Board, Commission, volunteer organization or Committee[53]. As can be seen from the table (below), the spread of “strongly agree” responses for SELFQUAL is smaller than that of the two previous questions and the levels are considerably higher. The SELFQUAL question is the most direct query into respondents’ self esteem divorced from considerations of willingness to expend effort on matters of politics.Figure 4-6: “Strong Agreement” Distribution for SELFQUAL

Figure 4-6: “Strong Agreement” Distribution for SELFQUAL

INT4: I think that I am as well informed about Town affairs as most people. [INFORMED].

The NES question reads: “I think that I am better informed about politics and government than most people”. The major difference between the wording of these two forms of the question is the distinction between “better than” (NES) and “as well as” (questionnaire). The NES version, in part, rests on a class distinction, whereas the questionnaire question is dependent on an information distinction. The questionnaire form of the question is very much related to the UNDERSTAND question with the difference that the UNDERSTAND question implies both information and comprehension (“I can make sense of this”) whereas the INFORMED question requires only access to information without the need for comprehension of the rationale behind various alternatives. Also, the response to this question is not absolute, but rather relative to the respondent’s evaluation of fellow citizens. The table (below) bears out this relationship of the questions with regard to the number who provided the same answer to both questions and it shows the much stronger sense of both information and understanding of town meeting attenders compared to the other groups of responders.

Figure 4-7: Comparison of INFORMED and UNDERSTAND Responses

External Efficacy Questions:

EXT1: I don’t think that Town officials care much about what people like me think [NOCARE].

When comparing the various groups of respondents, the graph (below) shows an approximate 25% difference (30% – 55%) in the number of respondents who answered this question with “strongly disagree”. The gap is driven by the advertisement respondents (Ad), more of who probably feel divorced from Town affairs. The NES question reads: “I don’t think public officials care much what people like me think”, which is open to interpretation as a class distinction rather than the feeling expressed in an earlier survey that “they” run the town, and “they“ would not listen to the interviewee if the interviewee did participate (Sielman 2006, 26) which expresses the feeling that a clique runs the Town.

Figure 4-8: NOCARE Comparison

EXT2: People like me don’t have a say about what the Town does [NOSAY]

This question addresses the sense of empowerment felt by respondents. Clearly, one would expect that attenders at town meeting would “disagree” with this statement both absolutely and relative to other groups of respondents who are not among “the usual suspects” who do participate in town meetings. This is borne out by the relative scores of the various groups of respondents depicted in the graph. The NES question reads: “People like me don’t have any say about what government does”. In responding to this question the respondent is free to interpret ‘government’ as being local, state or federal, representing 3 very different degrees of remoteness.

Figure 4-9: NOSAY Comparisons

My supposition is that most respondents would assume that ‘government’ refers to Washington. Further, some may assume that Congress is the referent and others may assume it is the federal bureaucracy. In most cases there is a feeling of remoteness and insularity which is in contrast to local government that has offices in the local town hall and whose office holders may be met at the super market or a little league game.

EXT3: Sometimes I do not understand the complexities of Town government [COMPLEX].

As can be seen from the graph, this question elicited a very wide spectrum of responses amongst all of the groups with relatively little difference between the groups.

Figure 4-10: COMPLEX Comparisons

Beyond the uncertainty posed by Niemi, et al, there exists an additional ambiguity in the survey question wording. It is the interpretation that the respondent gives to “sometimes”. It could refer to approximately half of all local government functions or a single function with which the respondent has had a less than satisfactory interaction [such as: “the Transfer Station attendant would not let me dispose of my dehumidifier without my first having its refrigerant drained by a licensed organization”]. To some respondents, the ‘complexities’ may be a substitute for ‘procedures’ or even ‘personalities’ of those administering town government functions. As a result the mean score (“disagree strongly” = 1, “disagree somewhat” = 0.75, “neither agree nor disagree” = 0.50, “agree somewhat” = 0.25 and “agree strongly” = 0), for the COMPLEX question by all respondents was 0.46, which was the only score that had a mean below 0.50 for all of the 19 survey questions. The next lowest score was 0.59 and the average weighted score was 0.77.

Cross Correlations

Figure 4-11: Internal and External Political Efficacy Cross Correlations

The graph (above) compares the Niemi, et al cross correlations among and between the internal and external efficacy questions against the same cross correlations for the various survey respondent groups (in descending order for Niemi, et al). It is seen that the NOCARE question responses generally have the lowest cross correlations with other question responses and that, except for the Referendum respondents, the NOCARE cross correlations for the survey are somewhat higher than the Niemi, et al data. This difference may be attributable also to the difference in question wording.

Community Related Efficacy Questions (CO)

I have labeled this set of questions “Community Related Efficacy Questions” because they address the respondent’s sense of how the community interacts with individual citizens. Although a previous survey (Sielman, 2006) found no correlation between length of residence in the community and willingness to participate in town meetings if it could be accomplished remotely, these questions are related to the sense of belonging that the respondent feels.

CO1: I don’t trust my fellow citizens.

This is a negatively worded question. The bar graph of scores in response to this question shows that there was little variation among the various groups of respondents with approximately 80% of each group disagreeing at least somewhat with the question statement.

Figure 4-12: Trust Comparisons

The Advertisement group was more tentative in their disagreement and even had a few respondents who agreed with the statement. In evaluating these responses one needs to take into account that the Town of Salem is fairly homogenous and affluent with the largest percentage of voters registering as ‘Unaffiliated’. The Town experienced rapid growth in the 1970s and 1980s leading to some division between old-timers and in-comers but that pattern has not been sustained in the last 20 years. Since trust is an essential ingredient in town meeting participation, its absence would present a potential obstacle to significantly greater participation in town meetings.

CO2: I feel that I am part of the Salem community.

This question directly measures the respondent’s sense of belonging, and probably the respondent’s sense of caring about community issues. Almost everyone cares about local property taxes because they are high and must be paid every July and January.

Figure 4-13: “Part of Community” Comparisons

There are other community issues (some of which translate into tax issues) that are addressed by the Town’s legislature (town meeting) that voters who do not feel that they are a part of the community are unlikely to feel the need to participate in. The scores associated with this question differ markedly from those relating to TRUST. Almost 80% of Attenders strongly feel part of the Town community, whereas less than 60% of the voters at the November Election and May Referendum share that feeling, with less than 30% of responders to the Advertisement feeling that way. There is tentativeness to the Advertisement group with nearly 70% agreeing only “somewhat” with the question statement. An aim of this study is to see if some of those who remotely participate in a virtual town meeting will migrate their response from “agree somewhat” to “agree strongly”. Will participation foster a sense of belonging?

CO3: I believe that it is important to discuss differences with people who disagree with me.

This statement is only peripherally related to the community, and primarily related to a philosophical belief in how to deal with differences of opinion. It was not included in the town meeting (TM) group of questions because it is not a matter exclusive to the workings of town meeting. As can be seen from the bar chart, there is little divergence of scores between the various groups of respondents. 50-60% “strongly agree” that discussion (deliberation) is an appropriate way to deal with differences of opinion. Even 40% of the town meeting attenders only “agree somewhat” that discussion with those we disagree with is important.

Figure 4-14: Comparison of DISCUSS

Their emphasis is probably more on majority rule inherent in town meeting decisions than on the process of deliberation itself.

Town Meeting Efficacy Questions:

- TM1: I feel that town meetings are not a good way to make local decisions.
- TM2: I think that town meeting is a good form of democracy.

TM1 and TM2 were never contiguously placed in the sequence of questions as they were presented to the respondents. Approximately 60% of all respondent groups provided the same response to both statements. Owing to the similarity of the questions, one would expect that almost all respondents would provide the same response to these two statements. Of the 30-40% who provided different responses 5-8% can be attributed to missing the negative wording (“not”) in CO1. This is shown in the table by the percentage of respondents who failed to answer TM1 with a “disagreement” response whereas they gave an “agreement” response to TM2. The remaining approximately 30% of respondents answered one of the two questions “strongly” and the other “somewhat”, thereby underscoring the concern that responses are not necessarily always consistent[54].

Figure 4-15: Comparison of Related Question Responses

- TM3: I would be willing to participate in town meetings in the future.
- TM4: I think that operating under Robert’s Rules of Order keeps town meeting orderly and efficient.
- TM5: I think that the town meeting moderator is important for making the deliberations fair.
- TM6: I don’t feel that I can contribute usefully to town meeting discussions of issues.
- TM7: I think that town meeting participants consider what is good for the Town in deciding how to vote.

These 5 statements in the questionnaire address the mechanics of how town meetings actually work. Attenders of town meetings[55] would fully understand the meanings of the posed statements and provide responses accordingly. Most of the group that responded to the Advertisement had never attended a Salem town meeting. This difference of background is borne out in the percentage of responses that “neither agree nor disagree” with the statement shown in the table below:

Figure 4-16: Comparison of Questions About the Mechanics of Town Meeting

It is seen that the responders from the Advertisement group had by far the largest percentage of ‘neither agree nor disagree” responses reflecting a greater lack of opinion about a process with which some of them are unfamiliar. This group of questions is a measure of efficacy in the sense that the responder understands how town meetings work and has opinions about the efficacy of the process. Of this group, TM3 (FUTURE PARTICIPATION) is a direct measure of the potential that virtual town meeting provides for increased participation in town meetings.

Figure 4-17: “Future Participation” Comparison

Probably none of those in the “agree somewhat” group are likely candidates for actual participation in future virtual town meetings, but the significant percentage of those who responded “agree strongly” are likely candidates[56].

TM8: I don’t think that everyone who wants to be heard at town meeting gets to present their views.

This seemingly straightforward statement had responses substantially different from all of the other questions except EXT 3 (COMPLEX). The scores relative to this question are summarized below:

Figure 4-18: “Everyone Being Heard” Comparison

Clearly, there are several threads that are discernable in these responses. The responses to “neither agree nor disagree” are large and are probably driven by an absence of experience on the part of some respondents with the workings of town meetings. Because TM8 is one of the negatively worded questions it is likely that approximately 5-8% of the respondents ‘agreed” with the statement because they missed the “don’t” in the question statement as was the apparent case in comparing TM1 and TM2 (above). The remainder of the “agree” responses may be attributable to unhappiness with the conduct of the town meetings by the moderator, reflecting a first hand experience or second hand report of an instance where a legislator felt unable to present an input to the deliberations[57]. The responses may also reflect different interpretations of the word “everyone” in the sense that it could be interpreted as 100% of the time at all town meetings, or as almost everyone at almost all town meetings.

TM9: Participating in town meeting was an enjoyable experience.

Figure 4-19: Town Meeting Experience Comparison

By comparison with TM8, the responses to TM9 are understandable and predictable. The statement is positively worded so a potential source of uncertainty is eliminated. The “neither agree nor disagree” responses likely represent an increasing unfamiliarity with, or total ignorance of, town meeting functionality. If one has never attended a town meeting, the logical response to town meeting was an enjoyable experience is the absence of agreement and disagreement. The mean weighted scores are as expected and the variance is reasonably low for all of the groups.

Age and Gender Analysis

Although the survey respondents were self-selected, almost an equal number of males and females filled out a questionnaire. There were 74 female respondents, 67 male respondents and 18 who did not provide identification from which age and gender could be ascertained. For the 141 respondents whose gender could be determined, the base EXCEL spreadsheet data was sorted to segregate the genders and then the mean of each gender’s score was calculated for each question and for each of the grouped sums for the INTERNAL, EXTERNAL, COMMUNITY and TOWN MEETING groups as well as for the combined (INT + EXT, CO + TM and Overall PEI).

Figure 4-20: Gender Comparison

It is clearly seen from the graph that there are no discernible differences in the response of men and women respondents. It is particularly significant because the survey did not have a very large number of respondents (141) thereby exposing the comparison to random phenomena.

Another potential aspect of gender differences was investigated to see if the strength of convictions varied as a function of gender. To evaluate this possibility the percentage of respondents who indicated a “strong” political efficacy response to each question (and the various sums thereof) was plotted. As shown in the figure below there is no meaningful difference between the response of men and the responses of women.

Figure 4-21: Gender Strong Response Comparison

By contrast, plotting the data for 10-year age cohorts provides stark differences. Although the age cohort sub-sample numbers are necessarily smaller than the gender samples, some clear distinctions can be made between the age cohorts. The 20’s age cohort has substantially lower mean scores for almost every question, and thereby, for the various sums. The over 69 cohort (70’s and 80’s) has substantially higher mean scores than all of the other groups. There is little to distinguish the 60’s, 50’s and 40’s cohort, but the 30’’s cohort appears to rest between the 20’s cohort and the middle group.

Figure 4-22: AGE Cohort Comparison

Figure 4-23: “Strong” Responses by Age Cohort

When one compares the “strong” responses by age cohort, the picture is even starker with a discernible “layering” of percentages as a function of age. Perhaps, as we age there is a hardening of perspectives and a (perhaps unwarranted) feeling of the need to make definitive statements.

Political Efficacy Index

The rationale for creating virtual town meetings is to eliminate some barriers to participation, both physical and mental, for some of the eligible participants who do not participate in some or all conventional town meetings. The value of enhanced participation is predicated on the widely held assumptions that (1) it provides greater legitimacy to the actions of the town meeting, (2) the actions taken by the town meeting will be “better” owing to the added perspectives provided by the additional participants, and (3) some of the added participants will have their political efficacy enhanced through the participation experience.

Niemi, et al claim that their “measurement of internal political efficacy measures a single concept distinct from external efficacy and political trust” (1991, 1407). They define internal efficacy as “referring to beliefs about one’s own competence to understand, and to participate effectively in, politics” (Ibid). The four internal political efficacy survey statements that they have evolved and that are a part of the National Election Studies (NES) are SELFQUAL, UNDERSTAND, PUBOFF and INFORMED.[58]

The first question that arises is: why are the NES items used in place of a direct response to the statement: “I am competent to understand and participate effectively in politics”? The NES questions break this statement into two constituent statements about UNDERSTAND and SELFQUAL which the Niemi, et al data show to be highly correlated (1991, 1408). It is not clear what additional value is gleaned by breaking the concept into two distinct survey items.

The second question that arises is what is the added value of the INFORMED item? How can one claim to understand without being informed or vice versa? Further the INFORMED statement that respondents are asked to respond to is not worded to inquire about a sense of competence (absolute) but rather uses the relative statement “better than most people”. Without delving into what constitutes “competence”, it is theoretically possible to be better informed than most people (whom the respondent can compare to) and yet be incompetent to participate effectively; and also, it is possible to not be better informed than most people (whom the respondent can compare to) and still be competent to participate effectively. So the INFORMED item is tinged with a comparative self worth evaluation rather than measuring the ability of the individual to be an effective participant in a democracy.

The third question that arises has to do with the applicability of the PUBOFF question. It seems to me that this item (at least partially) suffers from the weakness of the INFORMED item in that it is comparative to “most people” (although it uses “as good as” vice “better than” in INFORMED) and it further introduces the issue of serving in public office, which is a very narrow aspect of “participating effectively in politics”. As a veteran of having served in public office, both elected and appointed, I am keenly aware of aspects of serving in public office that tap into skills that are not required to participate in politics in the sense of participating in town meetings or election campaigns or furthering preferred policy options. These public office skills include dealing with constituents, managing a budget, managing staff, moving the bureaucracy, running meetings, communicating policies, etc. One can be a very effective participant in politics without having to posses these skills.

I raise these three questions concerning the measurement of internal political efficacy because Niemi, et al in effect create an internal political efficacy index by (1) scoring the responses to the four statements from 0 to 4 (“disagree strongly” to “agree strongly”), and (2) summing up the scores to create an index that covers the range of 0 to 16 (1991, 1411). Inherent in this process is the assumption of a linear process with equal weight provided to each statement. I believe that this choice is sub-optimum on two grounds. First, the range should be normalized to between 0 and 1, which allows interpretation as a percentage of maximum possible responses. A respondent who “agrees strongly” with all of the statements should be considered to have an internal political efficacy (IPE) of 100% and a respondent who “disagrees strongly” with all of the statements should have an IPE of 0%. The linear addition of responses is mathematically the most tractable and I am unaware of attempts to develop non-linear models. Since internal political efficacy is a concept rather than a directly measurable variable, such as how an individual voted, it would be difficult to develop and prove the validity of a non-linear model. However, as a second point, I raise the question of the utility of INFORMED, as being largely redundant with UNDERSTAND; and PUBOFF, as tapping into skills that are not necessary to “participate effectively in politics”. As a result, I believe that a preferable IPE would be the linear sum of SELFQUAL and UNDERSTAND normalized to the range of 0-100%. Ultimately, since I have raised the question of the utility of dividing the NES questions into separate SELFQUAL and UNDERSTAND, my real preference would be for the single statement: “I am competent to understand and participate effectively in politics” with the responses scored as:

Figure 4-24: Internal Political Efficacy Scoring

Notwithstanding my preferences (above), the survey data that I collected was modeled on Niemi, et al political efficacy statements so that comparisons could be made. The Niemi, et al numbers are presumably representative of the entire population in the late 1980s. The numbers derived from the 2008/2009 local survey are not a random selection of all eligible town meeting attenders. Rather, they are a self selected group who volunteered in response to: (1) a verbal request at the first virtual town meeting (Attenders- October 2008), (2) a sign in the lobby of the polling station for the November 2008 Election and (3) the same sign at the town budget Referendum in May 2009, and (4) in response to an advertisement in the local quarterly newsletter (Ad June 2009). The solicitation of survey responses was spread in time and only new respondents were solicited[59]. The time order of the collected responses was Attenders è Election è Referendum è Advertisement. The table below summarizes the relationship between the groups of respondents, their voting behavior and their calculated Internal Political Efficacy (IPE) score.Over 80% turnout of registered voters- less the 39 who had already signed up.

Figure 4-25: Voting Behavior and IPE of Respondent Groups

17.4 % turnout of registered voters less 110 who had already signed up. Up to 1000 camper owners were also eligible.
80% of 2585 registered voters less the 140 who had already signed up.
Calculated by multiplying scores of 1.00, 0.75, 0.50, 0.25 and 0.00 times the percentage of responses from “agree strongly” to “disagree strongly”.

When one compares the Niemi, et al IPE distribution (re-scaled to the range of 0 – 1) to the data collected in the survey (distributed to match the Niemi, et al scoring intervals), one sees that there are both similarities and differences in the distribution.

Figure 4-26: Internal Political Efficacy Sum Comparison

Two conclusions can be fairly drawn from the survey data and its comparison to the Niemi, et al data. The Niemi data is normally distributed whereas the survey data is heavily weighted towards higher levels of IPE. Secondly, Attender and Election IPE’s peak at “strong agreement” IPE’s (with Attenders having a much larger peak) whereas the Referendum and Advertisement peaks center on the “agree somewhat” IPE’s.

Figure 4-27: Cumulative IPE Scores

Another way to view the data is shown in the score comparisons in which the cumulative percentage of respondents with scores greater than or equal to each succeeding decrease in score is plotted. It is seen that all groups in the virtual town meeting survey had a greater percentage of maximum scores than the Niemi, et al data, and the virtual town meeting attenders were as much as 60% higher than the Niemi, et al data.

The survey data encompassed items concerning internal efficacy and external efficacy following the Niemi, et al model, but also included questions aimed at a sense of community (CO) and feelings about town meeting (TM). The graph below shows the scores for these separate categories of survey question.

Figure 4-28: Political Efficacy Scores by Category

In these data the curve marked “Av. PEI” represents the weighted sum of all of the 19 survey question scores with the ordinate equaling the percentage of all of the respondents who checked the selected score.[60] Several conclusions can be drawn from this chart. First, respondents felt most positive about the community statements (CO) with almost 60% “agreeing strongly”. Second, the shape of the responses for each of the groups is similar with the exception of the above-mentioned community “agree strongly” response and the larger number of “disagree somewhat” responses for external efficacy questions. The latter can be traced to the varied interpretations of the COMPLEX statement. Third, it is seen that the overall average PEI curve is a good approximation for the separate group of statements and can be used as a surrogate for the more detailed data.

I have suggested (above) that internal efficacy could be measured directly by requesting responses to “I am competent to understand and participate effectively in politics”. Since the survey did not use this combined statement, the closest alternative is to take the mean of the SELFQUAL and UNDERSTAND responses. A comparison of the Niemi, et al index and this surrogate index is shown in the graph below: These distributions are similar. The difference that does exist is driven by responses to the PUBOFF statement (also shown) which drags down the summed Niemi, et al-like index at the high end and, as mentioned above, involves considerations beyond the ability to participate effectively in politics. Since internal political efficacy is a “concept” rather than a measurable action, perhaps the most salient conclusion to be drawn from these Pretest data is the reverse of the Niemi, et al objective. People who currently attend town meetings have higher political efficacy indices than those who have not yet participated in town meetings.

Figure 4-29: SELFQUAL + UNDERSTAND

It is a form of validation of the internal political efficacy concept. What remains to be measured in the Posttest is whether or not remote participation increases the sense of political efficacy for current non-attenders.

Post Remote Participation Survey Analysis:

At this writing, the January 13, 2010 virtual town meeting represents the only opportunity, so far, to conduct a PrePost test to measure the impact of remote virtual town meeting participation on feelings of political efficacy. The data is encumbered with several weaknesses. First, although there were 21 remote participants[61] at this meeting (representing a 16% increase in participation over the already very large attendance) only eight of them had filled out Pretest questionnaires. The remaining 13 remote participants had signed up their email addresses in the two days preceding the virtual town meeting and, hence, no Pretest questionnaire data exists for them.[62] A second weakness is that the structure of a 5-choice questionnaire requires substantial changes in opinion before the mean delta becomes a significant number. The combination of a small number of inputs and the coarseness of response choices makes it unclear as to whether the data represents significant change or merely “noise”. The data is presented in Figure 4-27, below.

Several conclusions can be drawn from the PrePost test data:

1. The means of the deltas (Posttest minus Pretest) are generally positive. The summed political efficacy (INT, EXT, CO and TM) delta means are positive except for CO in which CO3 [“I believe that it is important to discuss differences with people who disagree with me”] is slightly negative owing to a single input where all of the other inputs were identical.
2. The largest negative delta mean occurs for question TM9 [“Participating in town meeting was enjoyable”] where notes were attached to some returns indicating frustration with the video dropouts.[63]
3. The largest mean delta increases occurred for TM3 [“I would be willing to participate in town meetings in the future”] and TM4 [“I think that operating under Robert’s Rules of Order keeps town meeting orderly and efficient”].
4. External political efficacy questions had the highest average increase
5. Internal political efficacy showed significant increases in UNDERSTAND and INFORMED question mean deltas with decreases in SELFQUAL and PUBOFF. The SELQUAL decreases started from a “strong” initial response to a “somewhat” response in the Posttest responses. PUBOFF (as discussed earlier) is an outlier because it encompasses skills that are beyond those required for a feeling of internal political efficacy

Figure 4-30: PrePost Test Political Efficacy Score Comparisons

Figure 4-31: Summary of All Posttest Responses

The figure above presents all of the Posttest responses with separate mean calculations for earlier signers-up (for whom Pretest responses exist) and the last minute signers-up who had not provided Pretest survey responses[64]. In the figure below, the comparison of the means for these groups of respondents is shown graphically.

Figure 4-32: Comparison of All Posttest Means

It can be seen from the figure that the means track fairly well except that the earlier respondents (those who had filled out Pretest surveys) had exhibited higher internal political efficacy and lower town meeting political efficacy for the questions TM7 [common good], TM8 [not everyone gets heard] and TM9 [participation as enjoyable experience]. The lower mean for TM7 probably reflects a higher level of cynicism about people’s motives; for TM8 the difference is probably attributable to a sense that debate was cut off at the January virtual town meeting earlier than those voting remotely desired; and, the response to TM9 reflected the concern about video dropouts.

Conclusions About The Case Study:

It would have been nice to have more PrePost test data on which to base conclusions about the political efficacy impact of remote participation. As the virtual town meeting concept spreads (which it likely will in our increasingly internet dependent culture) additional data will be collected and analyzed. A larger data set will allow the law of large numbers to swamp out the noise effects and thereby will provide more definitive data about the existence of a causal relationship between increased participation and a sense of increased internal political efficacy. In Chapter 2, I summarized the relevant scholarly literature. Perhaps most relevant to this issue is the reference to Acock:

Attaining increased political efficacy can be considered as a form of education, or at least, the diminution of ignorance. It may increase or decrease the perception of the quality of governance being provided but it surely increases “individuals’ self-perception that they are capable of understanding politics and competent to participate in political acts” (Acock 1985, 1064).

The conclusion from the PrePost test data (Figure 4-27) is that remote participation resulted in an increase in the UNDERSTAND and INFORMED mean responses, which corroborates Acock’ s conjecture about participation as an educational process.

Analysis of the survey data has left a residue of a concern about the use of negatively worded questions. Their use has the advantage of not appearing to slant the answers to more or less agreement, but has the disadvantage that the negative word can be overlooked (particularly when the questionnaire is being filled out in a hurry), resulting in a false picture of the respondent’s feelings of political efficacy.

The consistency of the data correlations with those collected by Niemi, et al provides a sense of assurance that the survey data are rooted in reality. However, whereas the Niemi, et al data is representative of the voting population as a whole, the study data is confined to a self-selected group of respondents who show much higher levels of political efficacy than the average citizen. This difference is not of concern in that the aim of the virtual town meeting concept is to increase participation for those who have perceived a barrier to their participation; it does not have as its aim a desire to facilitate participation by all of the eligible participants. The targets are the willing but not able, the interested but tentative, the knowledgeable but insecure, the busy but time-limited, and the cautious but curious.

A clear and convincing conclusion from the survey responses is that there is no difference in feelings of political efficacy between men and women. The responses to each and every question in the survey track closely between the males and females. On the other hand, there is a marked difference in response by the different age cohorts. We older voters express a significantly higher sense of political efficacy than all other age cohorts. The youngest cohorts (20’s and 30’s) express a substantially lower feeling of political efficacy than all the other cohorts[65].

A final conclusion from the case study is about the survey makeup. I firmly believe that if our objective is to measure internal political efficacy, defined as:

“I am competent to understand and participate effectively in politics”

then that is the question that should be asked in surveys- with the option of “strongly agreeing”, ”agreeing somewhat”, “neither agreeing nor disagreeing”, “disagreeing somewhat” or “strongly disagreeing. Doing so would avoid the overlap of UNDERSTAND and INFORMED, would eliminate the extraneous considerations implicit in PUBOFF and would incorporate the SELFQUAL political efficacy consideration without making it comparative to the respondent’s estimation of “other people”. A single question covering internal political efficacy would also eliminate the need to construct an internal political efficacy index based on an arbitrary linear addition of responses to 4 questions. Even if my suggestion above is not accepted, any political efficacy index ought to be scaled to the 0% – 100% range since “strong agreement” with positive worded statements (and ”strong disagreement” with negatively worded statement) represent the maximum responses available to respondents. Likewise, responses at the opposite extreme represent the least politically efficacious responses available to survey respondents and, therefore, should be scored at 0%. The concept of expressing political efficacy feelings as a percentage conveys the idea that a respondent ranks him or her self within a limited range and within a fixed scale.

Chapter 5: Conclusions

As recounted in chapter 2, it is a generally, but not universally, held belief among political scientists that the quality of governance is enhanced if there is substantial public participation in politics and if decisions are reached as the result of deliberative processes. The belief extends to encouraging political participation in the deliberative process by members of the public. This is predicated on the belief that greater public participation increases the legitimacy of decisions, leads to better decision making based on an increased knowledge and experience base of a larger pool of participants, and the conjecture that increased participation will enhance individual feelings of political efficacy. Governmental processes that incorporate increased public participation represent the goals of advocates of participatory and deliberative democracy. This process is alive, but ailing, in the New England town meeting form of government. The virtual town meeting tests and case study, reported on herein, have provided a preliminary evaluation of the application of technology to revive this theoretically attractive form of participatory and deliberative democracy.

Virtual town meeting is a natural application of widely available and heavily used technology. Almost everyone has access to a computer[66] and/or a television set and a very large percentage of the public use email as their primary means of communication. As described in Chapter 3, the processes developed for virtual town meeting participation are extremely “low tech” requiring minimal effort on the part of remote participants. The hardware and software required to set up a virtual town meeting is not expensive[67] and is easy to operate. Most importantly, the availability of virtual town meeting has increased the number of participants in each of the 4 tests of the concept conducted to date. This is but the first baby step in a process that, in my opinion, is bound to grow.

The absolute number and percentages of increased participation has varied substantially, depending on such factors as weather, the topics on the meeting call and the spread of information about the availability of remote participation. As of this writing, only 6% of the eligible voters in the Town of Salem, Connecticut have signed up to take advantage of the possibility to participate remotely in town meetings. Of these, approximately 50% attended the last (January, 2010) town meeting in person and 12% participated remotely. It is fair to conclude, however, that the availability of virtual town meeting has increased the number of participants. It is my belief that the level of remote participation will climb in the future as more of the remaining 94% learn about virtual town meeting and decide to sign up their email addresses. It is also fair to conclude that the security, vote counting and hardware constituents of the virtual town meeting process work well enough, while simultaneously recognizing that there is room for improvement, particularly if increased funding can be made available.

As reported in Chapter 4, measuring and analyzing political efficacy, although widely accepted in the discipline, incorporates some anomalies that became evident in its application to virtual town meeting participants. The basic question addressed in the case study is whether or not remote participation in virtual town meetings has an impact on feelings of internal political efficacy. The mechanism for addressing this question was a PrePost survey. At this writing, there are not enough PrePost test inputs (8 at the January 2010 virtual town meeting) to claim a statistically meaningful result. The trend in this limited data set seems to indicate a consistent increase of internal political efficacy feelings with regard to the INFORMED and UNDERSTAND questions. This trend is not evident in the PUBOFF and SELFQUAL responses but, as pointed out in Chapter 4, these questions present anomalies with regard to interpretation; with PUBOFF potentially alluding to skills that are unique to serving in public office but which are not requisite to having feelings of internal political efficacy. As a consequence, I am recommending that consideration be given to changing the NES internal political efficacy question to a direct mimicking of the accepted definition of internal political efficacy, namely:

“I am competent to understand and participate effectively in politics”

159 political efficacy survey questionnaires were filled out in the Pretest portion of the case study. This statistically significant set of responses revealed a number of conclusions that should be of interest to the discipline. First, there is absolutely no difference in responses to any of the 19 political efficacy questions between males and females. Second, there is a very substantial layering of feelings of political efficacy based on age, with feelings of political efficacy increasing with age cohort. The oldest cohort (>69) has higher political efficacy feelings than all the other age cohorts and the youngest cohort (20’s) has the lowest feelings of political efficacy. The cohorts on either side of these cohorts (60’s and 30’s) exhibit political efficacy feelings between the extremes and a middle group (40’s and 50’s) that exhibit little difference between them. Third, the cross correlations among all of the internal and external political efficacy questions are consistent with the data provided by Niemi, et al although there is an important difference. The Niemi, et al data is applied to a random sampling of voters, whereas the virtual town meeting case study is of a self-selected set of volunteers who chose to sign up for virtual town meeting remote participation. The very act of signing up is indicative of a level of political participation beyond that of randomly selected voters.

The data analysis of the questionnaire responses indicates a very substantial difference in feelings of political efficacy between those who signed up after attending a town meeting and those who signed up on Election Day, on Referendum day and in response to an ad in the local quarterly newsletter. The feelings associated with the community-related questions differed only minimally amongst these groups. But, the feelings of internal political efficacy amongst the attenders of town meeting were much higher than those of other groups and vastly greater than the Niemi, et al random voter data. Clearly, the causal arrow points from feelings of internal political efficacy toward participation in the political process. The data from the PrePost test analysis seems to indicate that the causal arrow also points from participation to enhanced feelings of internal political efficacy, but further data needs to be collected to statistically firm up this tentative conclusion.

Many of the authors cited in Chapter 2 caution the discipline to proceed carefully in the investigation of participatory and deliberative democracy and political efficacy. Included in this caution is an exhortation to avoid the introduction of extraneous influences by avoiding tests that incorporate artificial environments and incentives for the test subjects. The beauty of virtual town meeting is that is not artificial- these are real citizens discharging their legal legislative responsibilities in an environment prescribed by state statutes and local charters. The video recordings and email records of virtual town meetings provide a non-intrusive means for analysis of the proceedings.

My motivation in pursuing the virtual town meeting test and associated case study is to remove some of the barriers to participation in town meetings, thereby enhancing the legitimacy of the legislative process, improving the quality of the legislative decisions and, perhaps, affecting the willingness of some citizens to further involve themselves in other political processes because their sense of political efficacy has increased. A subsidiary motivation is that virtual town meetings may prove to be both a viable means of extending democratic institutions to a larger number of people (even where the institution does not currently exist) and may prove to be an ideal testing environment for studying deliberative and participatory democracy.

References

Acock, Alan, Harold D. Clarke and Marianne C. Stewart. 1985. “A New Model for Old Measures: A Covariance Structure Analysis of Political Efficacy.” The Journal of Politics 47 (4): 1062-1084.

Barabas, Jason. 2004. “How Deliberation Affects Policy Opinions.” American Political Science Review 98 (4): 687-701.

Barber, Benjamin. 1984. Strong Democracy. Berkley: University of California Press.

Barber, Benjamin. 1998. A Place for Us. New York: Hill and Wang.

Bennett, Stephen Earl. 1997. “Knowledge of Politics and Sense of Subjective Political Competence.” American Politics Quarterly 25 (2) 230-240.

Berry, Jeffrey M., Kent E. Portney and Ken Thomson. 1993. The Rebirth of Urban Democracy. Washington, D.C.: The Brookings Institution.

Bohman. James. 1998. “The Coming of Age of Deliberative Democracy.” The Journal of Political Philosophy 6 (4): 400-425.

Bowler, Shaun and Todd Donovan. 2002. “Democracy, Institutions and Attitudes about Citizen Influence on Government.” British Journal of Political Science 32 (2): 371-390.

Brown, Steven R. 1971. “A Review of Pateman’s Participation and Democratic Theory.” Public Opinion Quarterly 35: 512-513.

Bryan, Frank M. 2003. Real Democracy. Chicago: The University of Chicago Press.

Campbell, Angus, Gerald Gurin and Warren E. Miller. 1954. The Voter Decides. Evanston, IL: Row, Peterson.

Campbell, Angus, Phillip E. Converse,, Warren E. Miller and Donald E. Stokes. 1960. The American Voter. Chicago: The University of Chicago Press.

Chambers, Simone. 2003. “Deliberative Democratic Theory.” Annual Review of Political Science 6: 307-326.

Chevallier, Michel. 2009. “Internet Voting, Turnout and Deliberation: A Study.” Journal of e-Government Volume 7,Issue 1, 29-44 (available online at http://www.ejeg.com).

Cohen, Joshua. 1998. “Democracy and Liberty” in Elster, John. Deliberative Democracy. New York: Cambridge University Press.

Cronin, Thomas E. 1989. Direct Democracy. Cambridge: Harvard University Press.

Dahl, Robert A. 1998. On Democracy. New Haven: Yale University Press.

Delli Carpini, Michael X. and Scott Keefer. 1996. What Americans Know about Politics and Why it Matters. New Haven: Yale University Press.

Delli Carpini, Michael X., Fay Lomax Cook, and Lawrence R. Jacobs. 2004. “Public Deliberation, Discursive Participation, and Citizen Engagement.” Annual Review of Political Science 7: 315-344.

De-Shalit, Avner. 2006. Power to the People. New York: Lanham.

Finkel, Steven E. 1985. “Reciprocal Effects of Participation and Political Efficacy.” American Journal of Political Science 29 (4): 891-913.

Fishkin, James S. 1991. Democracy and Deliberation. New Haven: Yale University Press.

Fishkin, James S. 1995. The Voice of the People. New Haven: Yale University Press.

Fung, Archon, 2007a. ”Minipublics: Deliberative Designs and their Consequences”in Rosenberg, Shawn W. 2007. Deliberation, Participation and Democracy. New York: Palgrave Macmillan.

Fung, Archon. 2007b. “Democratic Theory and Political Science: A Pragmatic Method of Constructive Engagement.” American Political Science Review 101 (3): 443-458.

Gambetta, Diego. 1998. “Claro! An Essay on Discursive Machismo” in Elster, John. Deliberative Democracy. New York: Cambridge University Press.

Gargarella, Roberto. 1998 “ Full Representation, Deliberation and Impartiality” in Elster, John. Deliberative Democracy. New York: Cambridge University Press.

Gastil, John and, James P. Dillard. 1999. “Increasing Political Sophistication Through Public Deliberation.” Political Communications 16: 3-23.

Gastil, John. 2000. “Is Face-to-Face Citizen Deliberation a Luxury or a Necessity”? Political Communication 17: 357-361.

Goodin, Robert E. 2003. Reflective Democracy. New York: Oxford University Press.

Gutman, Amy and, Dennis Thompson. 2004. Why Deliberative Democracy. Princeton: Princeton University Press.

Hibbing, John R. and Elizabeth Theiss-Morse. 2003. Stealth Democracy. New York: Cambridge University Press.

Iyengar, Shanto. 1980. “Subjective Political Efficacy as a Measure of Diffuse Support.” Public Opinion Quarterly 44 (2): 249-256.

Johnson, James. 1998, “Arguing for Deliberation: Some Skeptical Considerations” in Elster, John. Deliberative Democracy. New York: Cambridge University Press.

Mansbridge, Jane J. 1980. Beyond Adversary Democracy. Chicago: The University of Chicago Press.

Mendelberg, Tali. 2002. “The Deliberative Citizen: Theory and Evidence.” In Research in Micropolitics: Political Decisionmaking, Deliberation and Participation. Edited by M.X. Delli Carpini.

Michelson, Melissa R. 2000. “Political Efficacy and Electoral Participation of Chicago Latinos.” Social Science Quarterly 81 (1): 136-150.

Morrell, Michael E. 2003. “Survey and Experimental Evidence for a Reliable and Valid Measure of Internal Efficacy.” Public Opinion Quarterly 67: 589-602.

Morrell, Michael E. 2005. “Deliberation, Democratic Decision-Making and Internal Political Efficacy.” Political Behavior 27 (1): 49-68.

Morrell, Michael E. 2008. “Empathy, Winning and Losing & Citizens’ Perceptions of Deliberation.” [unpublished manuscript].

Mutz, Diana C. 2006. Hearing the Other Side. New York: Cambridge University Press.

Mutz, Diana C. 2008. “Is Deliberative Democracy a Falsifiable Theory?” Annual Review of Political Science 11: 521-538.

Niemi, Richard G., Stephen C. Craig, and Franco Mattei. 1991. “Measuring Internal Political Efficacy in the 1988 National Election Study.” American Political Science Review 85 (4): 1407-1413.

Pateman, Carol. 1970. Participation and Democratic Theory. New York: Cambridge University Press.

Renshon, Stanley Allen. 1974. Psychological Needs and Political Behavior. New York: The Free Press.

Rogowski, Ronald. 1981. “The Obligations of Liberalism.” Ethics, 91: 296-301.

Rosenstone, Steven J. and John Mark Hansen. 2003. Mobilization, Participation, and Democracy in America. New York: Longman.

Ryfe, David M. 2005. “Does Deliberative Democracy Work”? Annual Review of Political Science 8: 49-71.

Sanders, Lynn M. 1997. “Against Deliberation.” Political Theory, 25 (3): 347-376.

Shapiro, Ian. 2003. The State of Democratic Theory. Princeton: Princeton University Press.

Shane, Peter. 2007 “ The Virtual Agora Project.” Ohio State University.

Sielman, Peter F. 2006. “Investigation of Non-Participation in the Town Meeting of a Small Connecticut Town.” University of Connecticut, Master’s Thesis.

Steiner, Jürg, André Bächtiger, Markus Spörndli, and Marco R. Steenbergen. 2004. Deliberative Politics in Action. New York: Cambridge University Press.

Sunstein, Cass. 1995. Democracy and the Problem of Free Speech. New York: The Free Press.

Thompson, Dennis F. 2008. “Deliberative Democratic Theory and Empirical Political Science.” Annual Review of Political Science 11: 497-520.

Van Mill, Davis. 2006. Deliberation, Social Choice and Absolutist Democracy. New York: Routledge.

Verba, Sidney, Kay Lehman Schlozman, and Henry Brady. 1995. Voice and Equality. Cambridge: Harvard University Press.

Wolfsfeld, Gadi. 1986. “Political Action Repertoires: The Role of Efficacy” Comparative Political Studies 19 (1): 104-129.

Wollman, Neil and Robin Stauder. 2001. “Believed Efficacy and Political Activity; A Test of the Specificity Hypothesis.” The Journal of Social Psychology 131 (4): 557-566.

Zimmerman, Joseph F. 1999. The New England Town Meeting. Westport, CT: Praeger.

Zimmerman, Joseph F. 2002. “The New Hampshire Referendum Town Meeting.” Current Municipal Problems, 28: 425-437.

Zittel, Thomas and Dieter Fuchs,. 2007. Participatory Democracy and Political Participation. New York: Routledge.

APPENDIX A: TOWN OF SALEM, CT CHARTER – CHAPTER IV – THE TOWN MEETING

SECTION 4.01 GENERAL

The Town Meeting is the primary legislative body of the Town. The Town Meeting shall have authority for final approval of those actions of the Town hereinafter enumerated.

Electors shall be eligible to vote on all matters in Town Meetings and referenda called pursuant to this Charter. Voters who are not electors shall be eligible to vote only on financial matters in Town Meetings and referenda called pursuant to this Charter.

SECTION 4.02 ANNUAL TOWN MEETING

The Annual Town Meeting shall be convened in accordance with the provisions of Section 10.07 of this Charter. The Board of Selectmen may require at such meeting the presence of department heads and/or representatives of each board of the Town. In addition to the consideration of the Town Budget, the Annual Town Meeting may consider any of the items contained in Section 4.03 of this Charter.

SECTION 4.03 TOWN MEETINGS

A. Town Meetings shall be called by the Board of Selectmen for consideration of the following:

The issuance of bonds and all other forms of financing, the terms of which are in excess of one year.
Supplemental appropriations as provided for in Section 10.09 of this Charter.
Any appropriation greater than $3,000 from the Capital Reserve Fund.
Any purchase by the Town of an interest in real estate purchase at a cost in excess of $5,000.
The sale or disposition of any interest in any Town-owned real property, or the sale or disposition of any Town-owned personal property whose value exceeds $15,000.
Real estate leases to which the Town is party and which involve a term in excess of one year.
The creation, termination or modification of any permanent agency or department of the Town not provided for in this Charter.
Proposals for municipal improvements, including committees to plan and oversee such improvements.
The acceptance of any interest in real property.
Such other matters or proposals as may be required by the provisions of this Charter, the General Statutes, ordinance, or as the Board of Selectmen, in its discretion, shall decide are of sufficient importance to be submitted to a Town Meeting, including the adoption, modification or repeal of any ordinance.
Approval of the Town’s Plan of Conservation and Development.

B. Action taken by a Town Meeting, including Town Meeting items adopted at referendum, shall not become effective until the seventh calendar day following such action, unless otherwise provided by the Town Meeting action or by this Charter.

SECTION 4.04 ELECTORS’ POWER TO REQUIRE TOWN MEETINGS

A. The electors of the Town shall have the power to call a Town Meeting to propose ordinances, repeal or modify existing ordinances, propose Town Meeting actions on matters enumerated in Section 4.03 of this Charter, or consider such other matters or proposals that the petitioners deem to be of sufficient importance to be submitted to a Town Meeting.

B. Electors may call a Town Meeting in the following manner:

A petition may be filed by any elector of the Town with the Town Clerk, and except as otherwise provided herein, such petition shall conform to the requirements of Section 7-9 and 7-9a of the General Statutes.
Said petition, or counterparts thereof, shall contain the full text of the proposal and/or ordinance. Said petition shall be on a form prescribed or approved by the Town Clerk and shall be signed in ink by no fewer than fifty (50) electors of the Town.
Said petition, or counterparts thereof, shall be accompanied with affidavits signed and sworn to by each circulator as prescribed in Section 7-9 of the General Statutes.
The Town Clerk shall, within five calendar days after receipt of said petition,determine whether the petition and counterparts thereof received are sufficient as required by law and, if so, shall certify said petition to the Board of Selectmen.
Within thirty calendar days of the certification of said petition, the Board of Selectmen shall call a Town Meeting to take action on said petition.
At said Town Meeting, the matter included in the petition shall be approved if a majority of those eligible to vote thereon shall have voted in the affirmative.

SECTION 4.05 PROCEDURE

A. All Town Meetings shall be called in accordance with Section 7-3 of the General Statutes, by resolution of the Board of Selectmen. The Board of Selectmen shall fix the time and place of said meeting. The First Selectman shall give public notice of any such meeting at least ten calendar days but not more than 17 calendar days in advance by publication in a newspaper having general circulation in the Town and by posting a notice in a public place.

B. All Town Meetings shall be called to order by the First Selectman, if present, or by a member of the Board of Selectmen. A moderator shall be elected and all business conducted pursuant to Robert’s Rules of Order, as revised.

C. The Town Clerk shall serve as Clerk of all Town Meetings but, in the Town Clerk’s absence, a Clerk shall be designated by the moderator. Any Town Meeting may be recessed from time to time as the interests of the Town may require, and the moderator may entertain a motion to recess such meeting to a specific date and time.

SECTION 4.06 VOTING

Prior to taking the vote on individual motions put before any Town Meeting, the moderator may entertain a motion that the qualifications of those present and voting be demonstrated before the vote is cast.

SECTION 4.07 TOWN MEETING ACTION BY REFERENDUM

A. Except for the adoption of the Town Budget and such other matters required by this Charter to be decided at referendum, all Town Meeting actions which are not of a procedural nature (election of moderator, setting time and place for referendum, recess, adjournment, etc.) shall be submitted to referendum if either of the events set forth in (1) or (2) below occur.

The Board of Selectmen may, no fewer than five days prior to the day of a Town Meeting, order that any item or items on the call of such meeting be submitted to referendum in the manner set forth in this section.
A written petition may be filed by any elector with the Town Clerk not later than five days prior to the date of any Town Meeting requesting that any item or items on the call of such Town Meeting be referred to the Town voters at a referendum.

- A. Any such petition shall conform to the requirements of Section 7-9 and 7-9a of the General Statutes. Said petition shall be on a form prescribed or approved by the Town Clerk and signed in ink by no fewer than fifty (50) electors of the Town.
- B. Said petition, or counterparts thereof, shall be accompanied with affidavits signed and sworn to by each circulator as provided in Section 7-9 of the General Statutes.
- C. Not later than the date and time scheduled for the call of the Town Meeting to order, the Town Clerk shall determine whether said petition and affidavits are sufficient as prescribed by law and this Charter and, if so, shall certify said petition to the Board of Selectmen.

B. The First Selectman, or his designee, shall notify the moderator of the Town Meeting of the item or items on the call of the Town Meeting that shall be adjourned to referendum.

C. Any item on the call of the Town Meeting which is to be submitted to referendum, except any item that pertains to any specific amount of proposed expenditure, appropriation, borrowing or bonding, may be amended at the Town Meeting and shall be submitted to referendum as so amended.

D. The Town Meeting or, if the Town Meeting fails to do so, the Board of Selectmen shall fix the time and place of referendum, which shall be held no fewer than seven nor more than 14 calendar days after the date of the Town Meeting. Hours of voting shall be from six o’clock a.m., at the latest, until eight o’clock p.m., at the earliest. The Town Meeting will have the power to extend hours for voting.

E. The Board of Selectmen shall determine whether voting shall be by paper ballot or voting machine. Appropriate ballots or voting machine labels shall be prepared by the Town Clerk. The Registrars of Voters or, if they fail to do so by five calendar days prior to the referendum, the Town Clerk shall designate a moderator for the referendum, who may be the Town Clerk. The moderator of the Town Meeting or, if he fails to do so within twenty four hours of the adjournment of the Town Meeting, the Town Clerk shall phrase each item submitted to the referendum in the form of a question to be answered “YES” or “NO.”

F. Notice of the referendum shall be given in the manner provided by law and in such other manner, if any, as the Board of Selectmen may approve.

APPENDIX B: BOARD OF SELECTMEN PRESENTATION

Background

The purpose of the research that I am proposing for my PHD dissertation at UCONN is to ascertain whether providing eligible voters with the opportunity to participate in town meetings remotely will result in a net increase in the number of eligible voters who do participate in town meetings. The form of the proposed research is a case study and test in Salem. The effectiveness, as measured by comparing participation before and during the experiment, will be measured and participant attitudes will be solicited both before and during the test.

The rationales for seeking to increase the level of participation in town meetings include:

1. Greater participation enhances the legitimacy of decisions.
2. Greater participation improves the quality of decisions.
3. Participation enhances the political efficacy of the participants.

The proposed research stems from a previous survey I conducted of 40 eligible voters in Salem (of 134 contacted) who do not currently participate in town meetings. That survey concluded that a meaningful number (13% ± 8% with a confidence level of 95%) of them indicated that they would participate if they could do so from home. The proposed research is intended to verify the survey conclusions.

People who do not participate give one of the following reasons for their non-participation:

- “I can’t attend”
- “Nobody asked me to attend”
- “I don’t want to attend”

Survey Results: Participation from Home versus Reason for Non-Participation

Note: Numbers in parentheses are row percentages.

Legality (John Butts written opinion excerpts):

Many concerned citizens now express the opinion that town meeting participation in Connecticut is on the decline. The question arises whether, in the interest of expanded participation, municipalities may, without statutory modification, expand participation by eligible citizens via electronic or by other means. In the opinion of the undersigned, the answer is a qualified “yes.”

The Freedom of Information Act explicitly recognizes that a meeting can take place “by means of electronic equipment.” Section 1-200 of the General Statutes, therefore, read together with the town’s explicit authority to adopt rules of order for the conduct of its meetings (Section 7-7), and the Constitution’s recognition that the people have an undeniable and indefeasible right to alter their form of government in such manner as they may think expedient (Article 1, Section 2) lead the undersigned to conclude a Connecticut Town could conduct “virtual” town meetings.

Permission:

The Connecticut General Statute §7-5 says, “the place of holding town meetings may be determined by a majority of the voters present and voting at any town meeting specially warned and held for that purpose”. Since the Board of Selectmen call and warn town meetings, it is necessary to seek the Board’s permission to conduct the experiment. I also seek your permission to conduct mock “virtual” town meetings, using town equipment, to work out the mechanics of integrating remote participants.

APPENDIX C: BOARD OF SELECTMEN PRESENTATION – Point-of-Order Inputs to the January 13th Virtual Town Meeting

7:46 Point-of-Order: Lost connection….

7:49 Point-of-Order: You just came back!!!

8:00 Point-of-Order: We can’t see the presentation at home.

8:04 Point-of-Order: The online connection has been lost. I am no longer receiving the video.

8:53 Point-of-Order: Gene went blank and we’re left with a Salem commercial on Virtual Meetings.

9:05 Point-of-Order: I agree with tabling this item and come back with a differrent plan. Thanks!

9:14 Point-of-order: the feed is very inconsistent…we keep losing the feed, swant to
vote…o therefore.

9:15 Point-of-Order: feed just came back. Did we miss the whole vote?

9:17 Point-of-Order: Frustrating that the feed went in and out and we missed the final vote!

9:23 Point-of-Order: Has the full presentation been made tonight or interrupted by the questions?

Speak Inputs to the January 13^th Virtual Town Meeting

7:52 Speak: Please ask people who are answering questions to do so with the microphone; they are very difficult to understand! Thanks.

8:04 Speak: We have lost the live broadcast — argh!

8:06: Speak: Back on . . .

9:20 Speak: Listening from home, it sounds like a large part of the audience is not in favor of this project at this time. I understand and appreciate the financial concerns expressed. I would like to see this go to referendum to allow all to vote.

9:37 Speak: Excellent meeting! The virtual worked well and John Butts moderated superbly. Thank you.

January 14^th email: I have gone back over my e-mail from yesterday and found the YES email. All the others came in together, but the YES had come in earlier, separated from the others by a number of other e-mails. Sorry for my mistake.

January 20^th email: As you can see from the attached, the promised emails to participate in the virtual town meeting last night did not reach you because I failed to include the “1” in your email address. My apologies. I hope that you were able to watch the meeting and would be interested in any feedback if you did. Peter Sielman

January 20^th email: No problem – we had another commitment, but were able to join at about 8:15pm and watch the rest of the meeting. Generally worked very well, although we did lose the live feed temporarily a couple times. Thanks for organizing – I think it is a great idea.

APPENDIX D: SURVEY QUESTIONNAIRE ORDERS

The following 4 pages provide samples of the 4 different survey question orders used with the different 4 different groups of survey respondents. For analysis purposes, the responses were transposed into a common question order starting with internal political efficacy, followed in turn by external political efficacy, community political efficacy and town meeting political efficacy.

The data was entered into an EXCEL spreadsheet and the weighted sums were calculated by imbedded spreadsheet functions.

Footnotes

[1] There is a delay of approximately 4 seconds for both webcast and cablecast video and audio.

[2] Unfortunately not all scholars use the term “town meeting” in the sense of the New England town meetings. For instance, Fishkin: defines a town meeting as an event “ where candidates or office holders talk with eclectic groups of voters.” Fishkin, in footnote 1, elaborates by saying: “I distinguish town meetings from ’electronic town halls’ in Ross Perot’s sense” (1991, 107).

[3] In my over 20 years of town meeting attendance, the largest single meeting was related to the possibility of substantial State-provided revenues in return for the storage of low level nuclear waste material.

[4] Virtual town meeting employs some modern communications technology but, for reasons of both cost and adaptability of citizens, the technology component is kept to a minimum.

[5] De-Shalit worries that: “When the dynamics is toward reaching a wide agreement, majorities might lose patience, and minorities and their opinions may be overlooked or oppressed” (2006, 143).

[6] In this paragraph I have underlined the words “some” and “many” to emphasize the key premise of this work, which is that the goal is to provide opportunities for increased participation and deliberation. It should not be confused with the idea that everyone should participate in virtual town meetings.

[7] Shapiro echoes this position when he says: “Deliberation that is defended for its inherent benefits is best seen by the government, then, as a consumption good; people should be free – but not forced – to engage in it” (2003,43).

[8] In reference to Fishkin’s proposed Deliberation Day.

[9] Van Mill, without ever considering town meeting, concludes that: “It is very difficult to find an institutional setting that resembles the procedural requirements of the competing theories” (2006, 75). Having not considered town meeting, he goes on to analyze the workings of the Continental Congress!

[10] This is echoed by Barber in his definition of Strong Democracy as: “[P]olitics in the participatory mode where conflict is resolved in the absence of an independent ground through a participatory process of ongoing, proximate self-legislation and the creation of a political community capable of transforming dependent, private individuals into free citizens and partial and private interests into public goods” (1984, 132, emphasis in original).

[11] In virtual town meetings this is dependent upon teamwork between the moderator and the participants.

[12] See the discussion of Mutz’s concerns about the need for sufficient data (2008, 535) discussed below. Also, readers can judge the quality of deliberation of a virtual town meeting by viewing the January 10, 2010 virtual town meeting available ON DEMAND at www.salemct.org.

[13] Because the DQI evaluation is subjective, Steiner et al. go to great lengths to ensure the reliability of their coding (2004, 68).

[14] Campbell et al. recognized early on that political discussion was widespread and: “…one of the most significant forms of political behavior by a mass public” (1960, 92).

[15] Goodin elaborates on this point: “ … once people get used to making their choices with a view of having to justify them publicly to all and sundry, they might just stop trying to calculate who is likely to ask what, when” (2003, 223).

[16] The consequences of legitimacy and quality are advertised and have been observed in the past. The measurement of at least some of the consequences that I have lumped under the heading of “political efficacy”, are the goals of the case study reported on herein.

[17] See the discussion of the Dissemination literature (below).

[18] Thompson’s article appeared in the same 2008 issue of the Annual Review of Political Science as the Mutz article, with each commenting on the other’s text.

[19] Possible outcomes include that fewer participants will attend, preferring to participate remotely, and that first-time remote participants may feel that their fears of participation were unfounded and may decide to attend future meetings.

[20] Gutman and Thompson describe a limitation in the example they give of iterative deliberation with respect to healthcare in Oregon: “The fact that deliberative democracy does not in itself define a unique method for bringing deliberation to a justified conclusion (short of moral consensus) means that it acknowledges that no single method can justify whatever results from its implementation” (2004, 19).

[21] Remote participation refers to such alternatives as: home, at college, on a business trip or serving in the armed forces.

[22] Via email contact, Chevallier has kindly provided a reference to the details of Geneva e-voting design and safeguards which we have used to evaluate the “virtual’ town meeting security procedures.

[23] Niemi, et al include this question (labeled COMPLEX) in their list of retained political efficacy questions (1991, 1408) with the qualifier that COMPLEX, NOSAY and NOCARE survey questions “tap some combination of internal efficacy, external efficacy and political trust (though COMPLEX does appear to be at least a weak measure of the dimension).” My analysis in the dissertation corroborates that the survey equivalent of the NES COMPLEX question is an outlier with regard to its correlation with other survey questions.

[24] The response operationalization assigned numerical values to: agree strongly, agree somewhat, neither agree nor disagree, disagree somewhat, and disagree strongly (agree is considered more efficacious for the internal efficacy questions and disagree is considered more efficacious for the external efficacy questions).

[25] Tailoring to fit the context of the survey is consistent with Morrell’s approach (2003); he has shown that situation-specific wording of these items produces valid and reliable measures. The exact wordings of the questions used in the survey are in Appendix B. Comparison of the NES questions with the survey questions is discussed in detail in Chapter 4.

[26] Sunstein generalizes that: “What counts is the force of the argument. Politics is emphatically not a process in which desires and interests remain frozen, before or during politics” (1995, 245).

[27] I subscribe to pragmatism and limited goals when it comes to implementation of governing procedures. Shapiro sounds this same note: “That deliberation is not the cure-all proponents often believe it to be does not mean it is worthless in politics. … Machiavelli intimates that the common good is that which those with an interest in avoiding domination share…. If we embrace it, the question whether deliberation promotes the common good is reframed as the question whether it diminishes domination” (2003, 35).

[28] The survey was conducted of 40 eligible voters in Salem (of 134 contacted) who do not currently participate in town meetings. That survey concluded that a meaningful number (13% ± 8% with a confidence level of 95%) of them indicated that they would participate if they could do so remotely.

[29] See Appendix B for the presentation made to the Salem Board of Selectmen.

[30] Costs have been borne by the author and a grant from the Old Lyme Cable Council.

[31] A subsequent town meeting held on January 10, 2010 approved an unlimited number of virtual town meetings to be held at the discretion of the Board of Selectmen.

[32] I am grateful for the dedicated help of Will Spang, Tony Griggs, Chris Regan, Jay McKenney and Robyn McKenney.

[33] The exact cause and cure for this problem is under investigation. During the January 13, 2010 virtual town meeting several remote participants submitted emailed points-of-order indicating that the video had dropped out. Operational procedures have been modified to include live monitoring of the webpage so that the operators can rapidly reset the uplink transmission.

[34] In the survey of random households conducted by the author (Sielman, 2006) 88% of those queried had and use email.

[35] Microsoft Outlook permits the inclusion of “Rules” that automatically channel incoming emails to selected folders based on the contents of the email’s Subject Line. Microsoft Outlook further counts and displays the number of the unread email in each folder.

[36] Based on discussions with the moderator, it was decided to offer Point-of-Order as a distinct virtual hand-raising because it allows the moderator to give priority to points-of-Order in accordance with Robert’s Rules of Order.

[37] Using Microsoft Outlook, the computer applies message rules to forward incoming messages to the YES VOTE and NO VOTE folders

[38] Remote participants in effect are always voting by “checklist and ballot”, whereas attenders may vote by voice, show of hands or “checklist and ballot” at the moderator’s discretion.

[39] In 22 years of town meeting attendance this has occurred only once in my experience. It has the by-product that how individual people vote is not public knowledge.

[40] These virtual town meeting tests are conducted on a shoestring budget. More elaborate encryption devices could be applied.

[41] In these shoestring tests, this function is performed manually. Clearly, software could be written to automate this process.

[42] An alternative being considered is to turn off the video and audio transmission during the voting period that would make the process more simultaneous.

[43] Although the computer sorts the incoming emails by category (Present, Speak. Point-of-Order, Yes and No), all emails end up in the INBOX.

[44] The only larger attendance was at an informational meeting about the potential of storing low level nuclear waste material in the town.

[45] The website records also indicate that the website has been scrutinized by individuals in Russia, Latvia and Romania!

[46] This is a state-provided capability at the School, so fixing it may prove to be a daunting task. Since momentary outages are tolerable, a satisfactory work-around is careful monitoring of the downlink video and rapid reset if it is seen to drop out.

[47] Appendix D provides the four different question orders used in the survey.

[48] Professor Delli Carpini suggested the use of some negatively worded questions in an email to the author.

[49] Almost every voter at Referendum voted in the November Election.

[50] The Base Data (spreadsheet) upon which all of the following analysis is based can be found in Appendix E. In the spreadsheet each respondent was assigned a number to protect his or her identity.

[51] The names in [brackets] are those used by Niemi, Craig and Mattei (1991, 1410).

[52] It must be remembered that the questionnaire respondents represent a self-selected 6 % of the eligible respondents.

[53] Signs on major roads entering the Town of Salem proclaim it as “A Community of Volunteers”.

[54] Further evidence of this phenomenon is provided by the 3 individuals who responded twice. One of these respondents consistently provided “strong” responses to all questions representative of someone who feels maximally politically efficacious. The other two respondents, however, each provided different responses in their 2 submissions to at least 3 of the questions.

[55] It should be remembered that some of the responders who provided responses at the November election and May Referendum had attended some town meetings, but were not among those who filled out questionnaires at the October virtual town meeting.

[56] It should be remembered that the respondents to the questionnaire are a self-selected 6% of the eligible Salem voters. Each of these respondents signed up their email address for potential participation in future virtual town meetings.

[57] My own experience with over 20 years of town meeting participation is that moderators have leaned over backwards to ensure adequate chances for deliberation inputs with deliberation being quashed only when the question has ‘been called” and received a majority support reflecting a desire to vote on the question and move on to the next topic on the call of the meeting.

[58] Defined earlier in this dissertation.

[59] There were 3 repeat responses, one of which was identical to its precursor, but two of which provided somewhat different responses between the two submissions. It seems to me that this is a clear reminder that responses must be viewed as having a random noise component based on other influences on the respondents at the time of response.

[60] For positively worded statements, “agree strongly” = 1.00, “agree somewhat” = 0.75, “neither agree nor disagree” = 0.50, “disagree somewhat” = 0.25 and “disagree strongly” = 0.00. Negatively worded statement responses are multiplied by –1.

[61] There were 40 “hits” on the website during this virtual town meeting with an unknown distribution for the 19 additional viewers who did not reply to the Present email. Some were probably not eligible to participate, but some may have merely watched without wishing to indicate that they were doing so. There is no way of retrieving this mix.

[62] All 8 of the remote participants who had previously filled out Pretest questionnaires filled out Posttest questionnaires. 9 of the 13 Posttest-only participants filled out survey questionnaires despite email and snailmail (with self-addressed and stamped envelope) requests having been made.

[63] These responses are highlighted in yellow in the Figure.

[64] Those who made comments about the video dropouts are highlighted in yellow.

[65] The number of self-selected respondents in these age cohorts is also much smaller than their percentage in the eligible voter category.

[66] The term “computer” nowadays encompasses such items as Ipads and Iphones.

[67] The webcasting cart used at the Salem School cost less than $3,000 coupled with several man-hours of assembly and test.

or

Something Interesting