Peace polls are polls of publics who are parties to a violent conflict, they may be directly involved in the violent conflict themselves as potential victims of that conflict or indirectly involved as either the electorate of a government or constituents of a community engaged in a violent conflict.

Such polls can be undertaken at any point in the cycle of a conflict as the objective of a peace poll is to help parties involved in a conflict to bring an end to and/or prevent further harm associated with the destruction of property, injury and death. Ideally peace polls should be undertaken prior to the outbreak of violence when parties to a conflict may be threatening and/or preparing for violent action, in the hope that accurate measurement of the opinions and attitudes of all the effected publics can help resolve the conflict through peaceful negotiation. In practice such remedial action is frequently not undertaken until one party or another wishes to sue for peace after the prospects of their gaining some advantage through continuation of the violence has been lost. Critically peace polls are nonpartisan and therefore should be undertaken at any and all points in the cycle of a violent conflict with a view to preventing harm in accordance with international human rights and humanitarian law. Thus peace polls can and should be undertaken before violent conflicts begin, during hostilities and after the violence has ended to prevent, reduce and prevent the reoccurrence of violence and its harmful consequences.

Clearly there are many practical difficulties associated with such polling including, for example, freedom of association, free speech, the press and other media; the safety of those undertaking the polls and their informants; the neutrality of the researchers and their relationship to the parties to the conflict and how questions of independence and safety can be reconciled. Each conflict will present its own particular set of problems in these regards, and these guidelines are made with a view to helping the researchers navigate those problems.

Peace polls can serve a variety of different functions with a view to helping parties to peace. They can be used to track the attitude of publics involved in a conflict. Such timeline research should be able to identify those sections of the various publics that are commonly referred to as 'extremist elements' and 'silent majorities' along with their community and political affiliations. They can be used to rank order 'problems' associated with the causes of the conflict and 'solutions' or policies for dealing with those problems. Critically it is important to gauge the values that all publics have with regard to such problems and solutions so that all the parties to a conflict have an equal opportunity to know and weigh the views of their adversaries on key issues. In addition to such analysis and description peace polls can be used for public diplomacy by informing both publics and elites as to the nature of the conflict, the identification of common ground, points of most serious division and potential for compromise between various publics on these issues. Such research also clearly has an obvious academic dimension to it that might include a political analysis of leadership and party fortunes; however, if such research is to be used proactively as part of a peace process then care should be taken not to include questions that are politically partisan and/or can not be published as an otherwise transparent piece of research. These different functions will necessitate both the design of different research instruments and different methods of dissemination that will variously include summary and detailed reports to the interested parties, the media and academic press.

Peace polls undertaken as a part of a peace process should be published as soon as practically possible after the fieldwork is completed.  Any delay in disseminating the results will inevitably raise questions about the legitimacy of the effort as well as rendering the research and its conclusions less relevant over time.  For similar reasons the results should be disseminated to all the interested parties and publics at the same time as any preferential access to the results would be interpreted as partisanship. Secondary analysis for academic purposes, particularly as part of any comparative studies, is clearly not subject to these same imperatives.

Many national governments and Inter-Governmental Organizations (IGOs) place legal restrictions on providing assistance to terrorists. This does not and should not include research directed at identifying the opinions of such persons, their organizations and their respective constituencies. Identifying and measuring the extent of support for radical groups is a necessary part of conflict analysis and resolution. However, providing compensation to such groups for their cooperation and participation in public opinion research does raise a number of moral, legal and methodological problems. In this circumstance all forms of compensation to the various interested parties engaged with as part of the research should be avoided except for the research team themselves.