Campbell International Development Group Title Registration Form

Transcription

1 Campbell International Development Group Title Registration Form Please complete this form to outline your proposal for a Campbell International Development Group systematic review. the completed form to Martina Vojtkova, Coordinator, Campbell International Development Group: mvojtkova@3ieimpact.org. Tel: Before completing this form: Make sure that your proposal falls within our scope, and that it has not already been covered in another Campbell or Cochrane review. Check existing registered titles at: and Authors are advised to use the Cochrane Handbook for Systematic Reviews of Interventions (see Be aware that preparing a Campbell review requires a significant, long-term commitment. At least two authors are required before a title can be registered. 1. Title of review Suggested format: [intervention/s] for [outcome/s] in [problem/population] in [location/situation] Example: Water and sanitation interventions for reducing child diarrhoea in low and middle income countries. The Impact of Public Information on the Electoral Behaviour of Citizens and the Decision-Making of Politicians in Developing Countries: A Systematic Review 2. Background and objective(s) of review Briefly describe the problem, the intervention(s), the relevance to policy and practice, and the objective(s) of the review, including important sub-questions. Will you develop a logic model (theory of change) to illustrate the hypothesized mechanism of action (that is, how the intervention is expected to work)? Is there potential for differences in relative effects between advantaged and disadvantaged populations? Problem and intervention Democracies around the world, whether in developing countries or not, struggle with how to motivate elected politicians to govern well. In the United States, for example, elected officials often have approval ratings well below 50% suggesting that the mechanism of elections alone does not ensure that politicians govern in the interests of the majority of their constituents. Hundreds of political science and political economy articles, both theoretical and empirical, have been written trying to explain why voters behave the way they do and why politicians behave the way they do in response. Ferejohn s (1986) Incumbent Performance and Electoral Control is a good example. International Development Group title registration form revised 11 th April

2 In developing countries, often with nascent democracies, the struggle to use elections as an effective mechanism for promoting good governance is often greater. Not only are the culture and practices of elections newer and less familiar, the challenge of good governance is harder. Corruption is generally more predominant in developing countries, as we see from indexes such as the Transparency International corruption index. Politicians and civil servants both often have less education and training specific to their roles as public servants at the same time that issues they must address can be extremely complex, such as deciding between competing program priorities when so many in their country are in need or balancing the demand of competing donors. In addition, just as the culture of elections may be new, the culture of public service may also be in a developing phase. It is generally believed that more and better information leads to more and better voting and can in turn motivate better governance by politicians. The literature on the developed country experience supports this. For example Matsusaka (1985) presents a theoretical model and empirical analysis for the link between information and voting. For the developing world, where information is generally more scarce, the emphasis on information, not only as an explanatory variable, but also as something around which a development intervention can be designed is widely accepted. Besley and Burgess (2002) develop a model and provide empirical evidence to show that having a more informed and politically active electorate strengthens incentives for governments to be responsive (p. 1415). Many donors fund democracy programs specifically designed to improve governance by providing information to voters. Just one example is the United States Agency for International Development s (USAID) Governance Enhancement Project in Guyana, which recently announced voter education sub-grants (Kaieteur News Online, 2011). Objective of the review The review will systematically collect and synthesize evidence from high quality impact evaluations of interventions designed to inform citizens on governance-related issues in order to assess their effect on politicians behaviour in developing countries. We also want to determine whether the same type of information impacts the behaviour of citizens, because theory predicts that more informed citizens are more likely to vote in order to punish bad incumbents and reward the good ones (see next section). We will examine this question along the causal chain (see next section) using both the quantitative and qualitative information available in rigorous impact evaluations. Specifically, the review aims to answer the following questions: 1) What is the impact of publicly distributed information about governance on the behaviour of citizens and politicians, and how relatively effective are different approaches of information dissemination (including types of information and methods of dissemination)? 2) Why are different approaches effective in different contexts? 3) What are the relative costs of different types of interventions and can we draw any conclusions related to cost effectiveness? Theory of change As noted above, we will synthesize both quantitative and qualitative evidence along the causal chain. The preliminary flowchart for the theory of change for the systematic review is presented in the figure below. In the theory of change, the information can lead to two intermediate outcomes: it can better inform citizens about the politicians as individuals and/or it can better inform citizens about the issues that the politicians influence. The arm of the causal chain indicated by the red (straight) lines corresponds to the theory that politicians will change their behaviour before an International Development Group title registration form revised 11 th April

3 election occurs if they know that their type has been or will be revealed. That is, the information and the threat of election lead to a change in type (or more clear good-type behaviour) that then leads to better governance. This theory is sometimes called the incentive effect (see Barro, 1973 and Ferejohn, 1986 for formal models). The black arm (dashed line) corresponds to the more standard theory that information leads to better governance after an election has occurred by helping citizens to elect better politicians. This theory is sometimes called the selection effect (see Fearon, 1999 and Besley, 2005). The green arrows (dotted line) in the causal chain, which directly link informed citizens to higher voter turnout, reflect the theoretical models and empirical evidence that citizens are more likely to vote if they feel better informed, or conversely, that some citizens refrain from voting when they do not feel informed enough to vote (Matsutaka, 1995; Fedderson and Pesendorfer 1996, 1999). One clear source of evidence for this effect is the fact that many citizens who have already incurred the transactions costs of voting still do not completely fill their ballots (Fedderson and Pesendorfer, 1996). We will examine this link between public information and voter turnout as part of the causal chain. There are various theoretical models in the literature that support the linkages in the theory of change depicted in the figure. We will discuss these in more detail in the protocol. There are also many interventions that seek to increase voter turnout directly, for example by increasing voter registration. We are not covering that causal effect in this systematic review just increased voter turnout due to increased political public information. Dis-advantaged populations Dis-advantaged populations are most likely to enter the analysis in three ways. First, the public information intervention may target certain groups; second, the intervention may not target any groups, but some groups have more or less access to public information; third, different groups may have different behavioural responses to the same information. Both theoretical and empirical studies have shown that voting is correlated with education (Matsutaka, 2005; Dee, 2004; Milligan, Moretti and Oreopoulos, 2004; Krishna, 2006; Fornos, Power and Garand, 2004). Some explanations advanced in the literature emphasize that education reduces the cost of acquiring and interpreting complex political information; it reduces the cost of manoeuvring among the bureaucratic and technological obstacles to civic participation (Dee, 2004); and increases voters confidence about making the right selection as discussed above. So, included studies may explore differences in effect by level of education, or between literate and illiterate citizens. The hypotheses would be that the more educated the voters are, the more informed they are likely to be even in the absence of intervention. Thus, the highest impact of information is expected to be observed among the less educated citizens. Also, potential differences in effect may be explored by gender, especially in areas where women and men do not have access to the same type of information. For example, women may have lower education than similar men. Women may also be restricted to participate in activities outside the household and hence their information may be more limited than that of comparable males. In addition, around the world women still carry the larger burden of household production and may be restricted by formal and informal rules to participate in the local political life and hence both the opportunity cost of acquiring new information and the returns from the elections may serve as disincentives to women s civic participation. International Development Group title registration form revised 11 th April

4 At the current stage of the review it is not clear that the included studies will consider differences in effect between sub-groups of the eligible population of interest. However, as we proceed into the search and collection of the relevant studies, we will pay particular attention to potential differences in relative effects between advantaged and disadvantaged groups, where advantaged groups may be defined in terms of socioeconomic status, education, gender, ethnicity, religious beliefs and others. Relevance for policy or practice As Sina Odugbemi states in the World Bank s Accountability Through Public Opinion elections are the main accountability mechanism in representative democracies. (Odugbemi and Lee, 2011, p. 26) Civil society organizations and international donors together spend tens if not hundreds of millions of dollars a year on interventions designed to improve the functioning of democracies in developing countries. While effective democracy is an objective in itself, as the quote above suggests, it is also seen as a mechanism to improve governance. Many of these interventions focus on or incorporate public information as an important element. We do not have an estimate of the amounts spent on the specific type of intervention we are studying, because these interventions are not classified as their own group by donors. Some are part of civic education programs; some are part of civil society support programs; some are part of media development program; and so on. Understanding whether and how much citizens can be empowered to influence governance is particularly important as donors are moving to a model of increased government ownership over development programs and donor funding. International Development Group title registration form revised 11 th April

5 Method of distribution Figure 1 Type of information Media About politicians Civil society/ NGOs Government/ politicians Public information About governancerelated issues About corruption Reveal politician types Improve citizens political knowledge Voter turnout increases Politicians change type or more clearly reveal type Individual votes better reflect voters preferences Electoral majority more closely reflects society s preferences Better politicians elected Better governance/ Reduced corruption International Development Group title registration form revised 11 th April

6 3. Existing reviews Briefly describe any existing systematic reviews on the topic, and justify the need for this review if existing reviews exist or are in progress. There are no existing systematic reviews on the impact of public information on the electoral behaviour of citizens and the decision-making of politicians in developing countries. To our best knowledge, one systematic review that comes close to our topic is a recently started review by Joshi et al. (2011). Based on the information we have received from the IDCG regarding this ongoing study, it aims to study interventions, such as social audit, community monitoring, citizen report card and community scorecard initiatives, that seek to improve government service provision. The authors are still in the process of completing the protocol for their review, but from communications with them, it is clear that their review differs from ours in three important ways: 1. While the review by Joshi et al. (2011) focuses on civil servants, we are focusing on politicians, that is, elected officials who are directly impacted by the electoral process. That is, we exclude studies that only consider civil servants. 2. In terms of interventions, our focus is on public information regarding politicians and governance. While this public information may include information coming from constructed data on social accountability, it is only one possible source of information. The Joshi et al. (2011) review excludes interventions that fall under broader definitions of social accountability than the constructed measures such as citizen report cards and community score cards and excludes interventions that provide more general information regarding governance than direct measures of social accountability. Most of the studies we have identified so far for our review focus on public information regarding campaign platforms, voting records, or resource allocation decisions, that is, reporting on decision making rather than constructed accountability measures. 3. The outcome measures we are looking at are quite different too, since Joshi et al. (2011) is focusing on service delivery and looking at measurements of service delivery outcomes. Our study will look at outcome measurements of decision making behaviour by politicians (e.g. budget allocations) and on intermediate outcomes of voter behaviour. That is, the Joshi et. al. review excludes studies that do not measure service delivery related outcomes, while we include studies that measure governance in other ways. There are also two systematic review protocols that focus on other aspects of governance: one protocol looks at the evidence of the effectiveness of anti-corruption policies (Hanna, Bishop, Durlacher, Nadel and Scheffler, 2010) and the other one analyses the impact of increasing salaries on improving the performance of public servants (Carr et al., 2010). Hanna et al. (2010) plan on analysing quantitative and qualitative impact evaluations of programs that aim to reduce corruption. Such programs may focus on improving monitoring of government employees and changing their incentives in order to pursue the interest of the public rather than their own, or the programs may focus on changing the underlying rules of the system (Hanna et al., 2010, p. 5), so that citizens have more power in decision making about local public goods. Some examples include the study by Bjorkman and Svensson (2009) in which communities are empowered to decide on the rules about running health centres, or the study by Alatas, Banerjee, Hanna, Olken and Tobias (2010) where local governments and communities are engaged in deciding on eligible program beneficiaries. These types of interventions are designed to reduce corruption among civil servants and that is exactly the outcome on which the authors focus. International Development Group title registration form revised 11 th April

7 Carr et al. (2010) propose to conduct a systematic review on the impact of higher salaries on the performance of civil servants including teachers, nurses and judges. Thus, this review examines better compensations as the mechanism which could align the interests of the civil servants with those of the public. Similarly to the previously discussed protocol, this review will also only look at the performance of public servants, not elected officials. Our systematic review differs significantly from these two systematic review protocols. As previously outlined, we are interested in the behaviour of elected officials, not civil servants. Also, we focus only on interventions that improve citizens information rather than on interventions that improve citizens rights in public sector management and participation. Finally, in our review the main mechanism through which citizens can act on the new information is elections, which allows them to selecting better candidates and to punish the bad ones. 4. Define the population Who is included and who is excluded? Are disadvantaged populations included, defined across PROGRESS-Plus categories? 1 The review includes citizens and politicians, belonging to developing (low-or middle-income) countries. By politicians we mean officials who are put in office or aspire for office via elections, and not civil servants or bureaucrats. With respect to citizens we mean citizens of developing countries who are eligible to vote (in most cases limited by age) spread across all occupations, religion, gender, ethnicity, socio-economic status, educational qualification, state/region/province, disability and sexual orientation. In reviewing the included articles we will pay attention to whether the interventions are reaching all the eligible populations with particular attention on disadvantaged groups within the voting population. In fact, some interventions only target disadvantaged citizens such as slum dwellers (Banerjee, Kumar et al., 2010) and low-educated rural voters (Banerjee, Green et al., 2010). 5. Define the intervention(s) What is given, by whom, to whom, and for how long? What are the comparison conditions (what is usually provided to control/comparison groups who don t receive the intervention)? Are interventions aimed at the disadvantaged? The review focuses on interventions that disseminate information regarding governance. The interventions can broadly be defined with respect to the type of information and the method of dissemination. In this review we are only concerned with public information about officials who are put in office or aspire for office via elections. We do not focus on the behaviour of government employees or bureaucrats. The kind of public information falling within the purview of our research includes the following: 1. Politicians background information such as qualifications, voting records, wealth, tax returns and criminal/corruption records; 1 Disadvantage can be measured across categories of social differentiation, using the mnemonic PROGRESS-Plus. PROGRESS is an acronym for Place of Residence, Race/Ethnicity, Occupation, Gender, Religion, Education, Socioeconomic Status, and Social Capital, and Plus represents additional categories such as Age, Disability, and Sexual Orientation. International Development Group title registration form revised 11 th April

8 2. Incumbents development spending in their constituencies, their record attending oversight meetings and participating in local initiatives; 3. Interventions which supply general information about the importance of public spending in certain areas such as education and healthcare; 4. Messages from the candidates campaign platforms. Such interventions may involve dissemination of information in the form of audit reports, individual tax returns, governance indexes, performance ratings, report cards, etc. The means of distribution of the information may also vary from different media sources such as radio and newspapers, through door to door canvassing and group discussions, to posters in public places. We pose no restrictions on the population targeted by the interventions as long as the information campaigns do reach citizens eligible to vote. Some interventions may be targeted at more poorly educated people who may be more likely to have inadequate information about the candidates and their responsibilities. Other interventions may be design to reach all eligible voters. Yet, throughout the review, we will pay particular attention to disadvantaged populations as discussed above. Moreover, in this review we will include interventions of varying duration. The duration of the program usually depends on the method of dissemination and the type of information. For example, audits reports may be published annually and hence a rich panel dataset may be available for such program evaluations. Yet, pamphlets about incumbents performance in office are likely to be most effective shortly before elections and therefore such interventions may be rather short-term. The type of information included in the campaigns is publicly available, that is both treated and control populations have access to it, but it is generally only in the treatment areas where the public information is actively disbursed through various channels. Thus, we expect that most of the identified interventions will evaluate the impact of active distribution of the public information versus no active distribution. In some studies comparison may made with respect to which type of information may have the highest impact on voter turnout and governance (for example, Chong et al., 2010). 6. Outcome(s) What are the intended effects of the intervention? What are the potential or unintended effects of the intervention? Primary and secondary (intermediate) outcomes for the review should all be mentioned, together with beneficial and, if applicable, adverse effects. Note relevant and important outcomes for the appropriate disadvantaged groups. Primary outcomes The public information interventions that we focus on aim to improve governance. We are primarily interested in assessing how information interventions affect the performance and behaviour of elected officials as well as the behaviour of citizens in electing officials. We will include indicators of politicians performance such as leaders responsiveness to constituents needs and complaints, spending allocation decisions, and corrupt behaviour. To capture these International Development Group title registration form revised 11 th April

9 indications, researchers use a variety of measures. Some examples are: meeting attendance, budget allocations, constituent responses regarding perception of corruption, citizen reports of vote buying, and so on. Intermediate outcomes In the interventions at the centre of the current review, the main mechanism that influences the performance of elected officials is elections (through either the incentive or the selection effect). Therefore, the intermediate outcome affected by the interventions is voter behaviour. Common measures of voter behaviour are voter registration, turnout, and incumbent share of votes. Potential unintended consequences A potential unintended effect of the interventions is voter disillusionment. This may occur if, for example, released information about corruption is so striking that citizens become even more discouraged from the political process and hence turnout decreases, rather than increases. Another potential adverse effect is clientelism and misdirected spending. For example, politicians may embark on a spree of spending in the treated areas only to influence the looming elections, or they may divert resources from necessary but less visible development projects towards those that are more visible to constituents (such as from infrastructure to food distribution). Observations of this behaviour would suggest that, rather than improving governance, the interventions actually lead to deteriorated governance. International Development Group title registration form revised 11 th April

10 7. Methodology What types of studies are to be included and excluded: please describe eligible study designs, measures, and duration of follow-ups. Briefly describe proposed data sources, search strategies and methods of synthesis. Where the review aims to include quantitative and qualitative evidence, specify which of the review questions noted in section 2 will be addressed using each type of evidence. Inclusion and eligibility The systematic review will consider both quantitative and qualitative evidence from rigorous impact evaluations. Studies that will be included in this systematic review to answer questions of effectiveness (question 1 above) will be impact evaluations of micro-level interventions in developing countries. Both randomized control trials (RCTs) and non-randomized studies will be considered as long as they evaluate a relevant intervention and utilize robust econometric methods to measure the impact. The studies should evaluate campaigns that disseminate information on governance-related issues as defined in the previous section. All high quality studies using quasi-experimental methods (regression discontinuity design, instrumental variable methods, difference-in-difference methods, statistical matching) will be included in the analysis. Studies will be excluded from the review if they examine the impact of macro-level policies or if they are not conducted in developing countries. Theoretical papers on the topic will be used only for background information but will not be considered in the synthesis. Qualitative evidence from impact evaluations will be used to determine why some interventions may have an effect and others not (research question 2). In addition, subgroup analysis of the quantitative data may contribute to answering this question. In order to address the question of cost effectiveness (research question 3), we will use the evidence from counterfactual impact evaluations as described above to estimate effectiveness, and look for data on cost both in the impact evaluations and in any project documents that we can access. Data sources and search strategies We will systematically search (both online and hand search) relevant databases and journals, using pre-determined search terms and combinations of terms. Some of the databases are: Business Source Complete, British Library for Development Studies (BLDS), Econlit, Eldis, Google Scholar, IDEAS (REPEC), JOLIS, Journal Storage (JSTOR), Science Direct, Social Sciences Citation Index, and SocIndex. Some of the specific journals for which we will search the tables of content are: Evaluation, World Development, Journal of Development Studies, Journal of Development Effectiveness, Economic Development and Cultural Change, Journal of International Development, Quarterly Journal of Economics, and American Economic Journal: Economic Policy. To find reports and evaluations on the topic, we will also look at websites and databases for organisations such as World Bank (PREM and DIME databases), JPAL and IPA (publications), IFPRI (publications), the Inter-American Development Bank and e-gap.org. We will conduct a bibliographic back-referencing and citation tracking of included studies. We will contact experts in the field via for additional guidance on the topic and potential working papers. The search, selection and assessment of articles will be conducted independently by two International Development Group title registration form revised 11 th April

11 researchers. A third reviewer may be called upon to resolve any disagreements that may occur during this stage. Methods of synthesis Using the included IE evidence (that evidence in relevant studies that meets the quality criteria for IE evidence) we plan to estimate effect sizes and conduct meta-analysis assuming random effects. Some of the studies identified so far do use cluster randomized designs, so we will need to correct for this in estimating effect sizes. We are sceptical that we will be able to run meta regressions given the small expected number of studies and the heterogeneity of interventions and outcomes. We will also use narrative synthesis methods to analyse the IE evidence drawing on Popay, et. al. (2006) and other guides. Some examples of these methods include groupings and clusters, tabulation, forest plots, moderator analysis, qualitative case descriptions, and checking the synthesis with the authors of the primary studies. We will synthesize the non-ie evidence using qualitative synthesis methods. The specific methods chosen will depend in part on the type of qualitative evidence that we find, but will certainly include narrative synthesis and will likely include qualitative metasummary. Bibliography Adserà, A., Boix, C. & Payne, M. (2003). Are you being served? Political accountability and quality of government. The Journal of Law, Economics, & Organization, 19(2): Alatas, V., Banerjee, A., Hanna, R., Olken, B. A. & Tobias, J. (May 2010). Targeting the poor: evidence from a field experiment in Indonesia. NBER Working Paper retrieved from Alesina, A. & Tabellini, G. (January 2004). Bureaucrats or politicians? NBER Working Paper Banerjee, A., Green, D., Green, J. and Pande, R. (2010). Can Voters be Primed to Choose Better Legislators? Experimental Evidence from Rural India. Working Paper. Banerjee, A. V., Kumar, S., Pande, R. & Su, F. (March 1, 2010). Do informed voters make better choices? Experimental evidence from urban India. Working Paper. Barnett-Page, E. & Thomas, J. (2009). Methods for the synthesis of qualitative research: a critical review. BMC Medical Research Methodology, 9(59): Barro, R. J. (1973). The Control of politicians: an economic model. Public Choice, 14: Besley, T. (2005). Political selection. The Journal of Economic Perspectives, 19 (3): Besley, T. (2006). Principled agents? The political economy of good government. Oxford: Oxford University Press. Besley, T. and Burgess, R. (2002). The political economy of government responsiveness: theory and evidence from India. The Quarterly Journal of Economics, 117(4), Besley, T. & Prat, A. (2006). Handcuffs for the grabbing hand? Media capture and government accountability. The American Economic Review, 96(3): Bjorkman, M. & Svensson, J. (2009). Power to the people: evidence from a randomized field experiment of community-based monitoring in Uganda. Quarterly Journal of Economics,124 (2): Bobonis, G. J., Fuertes, L. R. C. and Schwabe, R. (May 09, 2011). The Dynamic Effects of Information on Political Corruption: Theory and Evidence from Puerto Rico. Working Paper 428. International Development Group title registration form revised 11 th April

12 Capuno, J. and Garcia, M. M. (2009). What difference can performance ratings make? Difference-in-difference estimates of impact on local government responsiveness in the Philippines. Discussion Paper No Carr, S. C., MacLachlan, M., Clarke, M., Papola, T. S., Normand, C., Thomas, S., McAuliffe, E., & Leggat-Cook, C. (2010). What is the evidence of the impact of increasing aid salaries on improving the performance of public servants, including teachers, nurses and judges? Massey University, New Zealand. Chong, A., De La O, A., Karlan, D. & Wantchekron, L. (March 27, 2010). Information dissemination and local governments' electoral returns, evidence from a field experiment in Mexico. Preliminary draft. Dee, T. (2004). Are there civic returns to education? Journal of Public Economics, 88: Djankov, S., La Porta, R., Lopez-de-Silanes, F. & Shleifer, A. (February 2009). Disclosure by politicians. NBER Working Paper Djankov, S., McLiesh, C. Nenova, T. & Shleifer, A. (May 2001). Who owns the media? NBER Working Paper Downs, A. (1957). An economic theory of democracy. New York, NY: Harper & Row Publishers. Fearon, J. D. (1999). Electoral accountability and the control of politicians: Selecting good types versus sanctioning poor performance, pp in A. Przeworski, Stokes, S.C. & Manin, B. (eds.), Democracy, accountability, and representation. Cambridge: Cambridge University Press. Feddersen, T. J. & Pesendorfer, W. (1996). The swing voter's curse. The American Economic Review, 86 (3): Feddersen, T. J. & Pesendorfer, W. (1999). Abstention in elections with asymmetric information and diverse preferences. The American Political Science Review, 93 (2): Ferejohn, J. (1986). Incumbent performance and electoral control. Public Choice, 50(1/3), Carnegie Papers on Political Economy, Volume 6: Proceedings on the Carnegie Conference on Political Economy, Ferraz, C. & Finan, F. (2008). Exposing corrupt politicians: the effects of Brazil s publicly released audits on electoral outcomes. The Quarterly Journal of Economics, 123 (2), Ferraz, C. and Finan, F. (2009a). Electoral accountability and corruption: evidence from the audits of local governments. NBER Working Paper 14937: Ferraz, C. and Finan, F. (2009b). Motivating politicians: the impacts of monetary incentives on quality and performance. NBER Working Paper 14906: Fornos, C. A., Power, T. J. & Garand, J. C. (2004). Explaining voter turnout in Latin America, 1980 to Comparative Political Studies, 37 (8): Fiorina, M. (1981). Retrospective voting in American national elections. New Haven, CT: Yale University Press, in Besley, T. (2006). Principled agents? The political economy of good government. Oxford: Oxford University Press. Gentzkow, M. (2006). Television and voter turnout. The Quarterly Journal of Economics, 121 (3): Giné, X. and Mansuri, G. (February 2011). Together we will: evidence from a field experiment on female voter turnout in Pakistan. Working Paper. Hanna, R., Bishop, S., Durlacher, K., Nadel, S., Scheffler, G. (2010). The effectiveness of anti-corruption policy: what has worked, what hasn t, and what we don t know a systematic review protocol. DFID Systematic Review. Higgins, J. P. T. & Green, S. (Eds.). (2008). Cochrane handbook for systematic reviews of International Development Group title registration form revised 11 th April

13 interventions. West Sussex, England: John Wiley & Sons Ltd. Humphreys, M. & Weinstein, J. (March 2010). Policing politicians: citizen empowerment and political accountability in Uganda. Working Paper. Hyden, G. & Court, J. (2002). Governance and development. World Governance Survey Discussion Paper 1. Krishna, A. (2006). Poverty and democratic participation reconsidered: evidence from the local level in India. Comparative Politics, 38 (4): Larcinese, V. (2009). Information acquisition, ideology and turnout: theory and evidence from Britain. Journal of Theoretical Politics, 21(2), Lassen, D. D. (2005). The effect of information on voter turnout: evidence from a natural experiment. American Journal of Political Science, 49(1), Leeson, P. T. (2008). Media freedom, political knowledge, and participation. The Journal of Economic Perspectives, 22 (2): Matsusaka, J. (1995). Explaining voter turnout patterns: an information theory. Public Choice, 84(1/2), Milligan, K., Moretti, E. & Oreopoulos, P. (2004). Does education improve citizenship? Evidence from the United States and the United Kingdom. Journal of Public Economics, 88: Odugbemi, S. & Lee, T., (Eds.). (2011). Accountability Through Public Opinion. World Bank: Washington, DC. Palfrey, T. and Keith P. (1987). The relationship between information, ideology, and voting behavior. American Journal of Political Science 31(3): Pande, R. (2011). Can informed voters enforce better governance? Experiments in lowincome democracies. Annual Review of Economics, 3: Popay, J., Roberts, H., Sowden, A., Petticrew, M., Arai, L., Rodgers, M., Britten, N. (2006). Guidance on the Conduct of Narrative Synthesis in Systematic Reviews. ESRC Methods Programme, April Rogoff, K. (1990). Equilibrium political budget cycles. American Economic Review, 80(1): Spencer, L., Ritchie, J., Lewis, J. & Dillon, L. (2003). Quality in qualitative evaluation: a framework for assessing research evidence. London (UK): Government Chief Social Researcher s Office, Cabinet Office, Available from cm pdf Strömberg, D. (2004). Radio's impact on public spending. The Quarterly Journal of Economics 119 (1): Tendler J. (1997). Good Government in the Tropics. Johns Hopkins University Press, Baltimore, MD in Khemani, S. (2007). Can information Campaigns overcome political obstacles to serving the poor? Forthcoming in The Politics of Service Delivery in Democracies. Devarajan, S. and Widlund, I. (Eds.) United Nations Economic and Social Commission for Asia and the Pacific (UN ESCAP). What is good governance? Kaieteur News Online. May 28, Retrieved from: Waddington, H., Snilstveit, B., White, H. & Anderson, J. (January 2010). The impact of agricultural extension services. 3ie Synthetic Reviews SR00 9. Protocol. Wantchekon, L. (2003). Clientelism and voting behavior: Evidence from a field experiment in Benin. World Politics, 55 (3), World Bank. (1991). Managing development: the governance dimension Discussion Paper International Development Group title registration form revised 11 th April

14 8. Review team List names of those who will be cited as authors on the final publication. Lead reviewer Name: Annette N. Brown This is the person who Title: Chief, Advancement and Impact Evaluation Services develops and co-ordinates the Affiliation: International Initiative for Impact Evaluation review team, discusses and Address: 1625 Massachusetts Ave., NW Suite 450 assigns roles for individual State, Province or County: Washington, DC members of the review team, Postal Code: liaises with the editorial base Country: USA and takes responsibility for the Phone: X107 on-going updates of the review Co-author There should be at least one co-author Name: Vanya Slavchevska Affiliation: American University Country: USA Co-author If applicable Co-author If applicable Co-author If applicable Name: Anjini Mishra Affiliation: International Initiative for Impact Evaluation Country: India Name: Affiliation: Country: Name: Affiliation: Country: 9. Roles and responsibilities Please give brief description of content and methodological expertise within the review team. It is recommended to have at least one person on the review team who has content expertise, at least one person who has methodological expertise and at least one person who has statistical expertise. It is also recommended to have one person with information retrieval expertise. Please note that this is the recommended optimal review team composition. Content: Annette Brown has participated in project design, implementation, and management for numerous development projects in the field of governance. She is currently co-authoring a paper with Cyrus Samii and Monika Kulma titled Evaluating Stabilization Interventions which includes several categories of interventions aimed at improving governance in post-conflict situations. Brown is among the specialists presenting and participating in the DFID Governance workshop in September Anjini Mishra has researched the impact of governance interventions in development and authored the 3ie Enduring Questions Brief How to Turn the Tide on Corruption (June 2011). Methodology: Brown and Vanya Slavchevska received advanced methods training in systematic reviews at the August 2011 Campbell Colloquium. Slavchevska is the co-author for the systematic review How Effective are Cash Transfer Programs at Improving International Development Group title registration form revised 11 th April

15 Statistics: Search: Nutritional Status?A Systematic Review of Programs Effects on Anthropometric Outcomes that is currently being finalized for publication submission. Brown will receive additional training on estimating effect sizes at a one-day workshop at the November 2011 American Evaluation Association conference. Mishra is scheduled to attend a systematic review training in India in the next month. Brown has delivered a systematic review training workshop for U.S. policy makers and authored a short introductory article on systematic reviews for the Monday Developments magazine. All three authors have read several texts on systematic reviews in preparation for this study. Brown and Slavchevska have training in statistics and econometrics from their Ph.D. studies in economics. Brown has authored or co-authored numerous papers that employ econometrics. Slavchevska is conducting econometric analysis as part of her dissertation. Slavchevska has served as one of the two searchers in a previous systematic review. Mishra has been preparing for the search by reviewing several resource documents, including those produced by the 3ie systematic review program. 10. Potential conflicts of interest For example, have any of the authors been involved in the development of relevant interventions, primary research, or prior published reviews on the topic? There are no conflicts of interest. Although Brown has been involved in implementing governance interventions, she has had no direct involvement with activities like those being studied in this systematic review. Mishra s Enduring Questions brief was intended to provide motivation for the conduct of a systematic review of the topic and so is not in conflict. 11. Support Do you need support in any of these areas: methodology and causal inference, systematic searches, coding, statistical analysis (meta-analysis)? At the moment, the team does not need support. Once we have a sense of the possible effect size evidence that can be pulled from the included studies, we may need advice in estimating effect sizes or conducting meta regressions. 12. Funding Do you receive any financial support? If so, where from? If not, are you planning to apply for funding? Where? We received financial support from 3ie in the form of a paid internship for Slavchevska during the summer of We are not planning to apply for additional financial support specific to this systematic review. International Development Group title registration form revised 11 th April

16 13. Proposed deadlines Note, if the protocol or review are not submitted within 6 months and 18 months of title registration, respectively, the review area is opened up for other reviewers. Date you plan to submit a draft protocol: October 30, 2011 Date you plan to submit a draft review: December 31, Declaration Authors responsibilities By completing this form, you accept responsibility for preparing, maintaining and updating the review in accordance with Campbell Collaboration policy. The Campbell International Development Group will provide as much support as possible to assist with the preparation of the review. A draft protocol must be submitted to the Group within six months. If drafts are not submitted before the agreed deadlines, or if we are unable to contact you for an extended period, the Group has the right to de-register the title or transfer the title to alternative authors. The Group also has the right to de-register or transfer the title if it does not meet the standards of the Group and/or the Campbell Collaboration. You accept responsibility for maintaining the review in light of new evidence, comments and criticisms, and other developments, and updating the review at least once every three years, or, if requested, transferring responsibility for maintaining the review to others as agreed with the Group. Publication in the Campbell Library The support of the International Development Group in preparing your review is conditional upon your agreement to publish the protocol, finished review and subsequent updates in the Campbell Library. Concurrent publication in other journals is encouraged. However, a Campbell systematic review should be published either before, or at the same time as, its publication in other journals. Authors should not publish Campbell reviews in journals before they are ready for publication in CL. Authors should remember to include the statement: This is a version of a Campbell review, which is available in The Campbell Library. I understand the commitment required to undertake a Campbell review, and agree to publish in the Campbell Library. Signed on behalf of the authors: Form completed by: Vanya Slavchevska Revision completed by: Vanya Slavchevska Date: 08/23/2011 Date: 10/17/2011 International Development Group title registration form revised 11 th April

17 For Campbell use: Title registration submission date: Title registration approval date: International Development Group title registration form revised 11 th April