© 2011 Foundation Review.All rights reserved.

Reproductive Health Matters 2011;19(38):146–1620968-8080/11 $ – see front matter

DOI: 10.1016/S0968-8080(11)38582-5www.rhm-elsevier.com www.rhmjournal.org.uk

Effective social justice advocacy: a theory-of-changeframework for assessing progress

Barbara Klugman

Associate Professor, School of Public Health, Faculty of Health Sciences, University of the Witwatersrand,Johannesburg, South Africa. E-mail: bklugman@mweb.co.za

Abstract: This article offers a theory-of-change framework for social justice advocacy. It describesbroad outcome categories against which activists, donors and evaluators can assess progress (or lackthereof) in an ongoing manner: changes in organisational capacity, base of support, alliances,data and analysis from a social justice perspective, problem definition and potential policy options,visibility, public norms, and population level impacts. Using these for evaluation enables activistsand donors to learn from and rethink their strategies as the political context and/or actorschange over time. The paper presents a case study comparing factors that facilitated reproductiverights policy wins during the transition from apartheid to democracy in South Africa and factorsthat undermined their implementation in the post-apartheid period. It argues that after legal andpolicy victories had been won, failure to maintain strong organizations and continually rethinkstrategies contributed to the loss of government focus on and resources for implementation ofnew policies. By implication, evaluating effectiveness only by an actual policy change does notallow for ongoing learning to ensure appropriate strategies. It also fails to recognise that apolicy win can be overturned and needs vigilant monitoring and advocacy for implementation.This means that funding and organising advocacy should seldom be undertaken as a short-termproposition. It also suggests that the building and maintenance of organisational andleadership capacity is as important as any other of the outcome categories in enabling success.©2011 Foundation Review. All rights reserved.

Keywords: reproductive and sexual rights, advocacy, social justice, monitoring and evaluation,South Africa

AT one of my interviews with the FordFoundation for a job as a program officerin reproductive health and rights, I was

asked whether I had any experience in success-fully influencing policy change.“Yes,” I said. “On abortion in South Africa.”I proceeded to tell the story of the campaign

to increase access to abortion and other repro-ductive health services in South Africa. Thecampaign culminated in the Choice on Termina-tion of Pregnancy Act 1996 and in significant,related policy changes that gave the public freeaccess to primary health care, an increasedrange of contraceptives for free, the right to

screening and treatment to prevent cervicalcancer, and more.The interviewers then asked whether I'd had

experience of such advocacy going wrong.“Yes,” I said. “On abortion in South Africa!”And I described how, despite winning passage

of a law that should have led (and did, to a limitedextent) to a significant decrease in the numberof maternal deaths and ill-health in the country,the campaign had not managed to address themajor barriers to countrywide implementation.This story highlights a number of the chal-

lenges to advocates, donors, and evaluators inunderstanding and advocating for social change.

146

This article, therefore, aims to unravel some ofthe key components of policy advocacy as away of reflecting on what kinds of outcomescan be used as markers of progress towardsachieving the goals of social justice. Beforeconsidering the nature of policy advocacy, it isworth making explicit the values underlyingmy use of the term “social justice”, which drawson the analysis of Nancy Fraser1,2 and on prin-ciples of human rights.3 Social justice advocacydescribes efforts to: a) increase fairness in thedistribution of resources;* b) end discriminationagainst all groups, fostering values that recog-nize all people as equal; and c) promote theparticipation of people in policy and implemen-tation processes that affect their lives, and trans-parency and accountability for how decisionsare made and how they impact on society.†

Theory of change for policy advocacyThere is a general consensus that advocacy andadvocacy evaluation cannot be done without atheory of change (whether explicit or implicit)and that this needs to be grounded in socialscience research4 and in experience, in order todraw on field learning, without which it can beweak or problematic.5 One has to understandwhatthe organization, coalition or network thought itwas doing and what it hoped to achieve by itsactions in the short and medium term in order tobe able to evaluate it. A number of evaluators useKingdon's6 approach to policy analysis as thebasis of a theory of change to explain the natureand complexity of policy processes.7–10

Kingdon points out that there are a world ofproblems which never get onto the politicalagenda, and similarly a world of potential solu-tions. In tandem, “political events flow along ontheir own schedule and according to their ownrules, whether or not they are related to prob-lems or proposals”(p.20).8 Hence, the processof problem identification, the process of deve-loping solutions, and the political process arenot sequential but should be understood as

“multiple streams” that flow independently andsimultaneously – and in each, different actorsmay take part.For this reason, it is necessary to analyse how

a problem gains recognition as a problem to beaddressed in the political terrain, how specificsolutions get onto the political agenda, and whypoliticians are concernedwith certain issues ratherthan others at a particular moment in time. I haveadded a fourth stream, that of bureaucracies andadministration, since implementation is as mucha site of policy making as is law; and bureaucratsand administrators, as with policy makers, act onthe basis of personal and institutional concernsthat may bear no relationship to the problemsand desired solutions of those who are most inneed or marginalized.11 Hence, the focus of advo-cacy is on the processes required to influenceproblem definition and identify matching solu-tions and then to get these onto the agendas ofthe politicians, bureaucrats and other decision-makers who determine policies and their imple-mentation, and keep them there in the face ofopposition or bureaucratic apathy (see Figure 1).Kingdon suggests that “policy entrepreneurs”

have the role of creating connections betweenthese streams, working with the media and lob-byists as critical components of this process. Inrelation to social justice advocacy, I frame theseas “policy activists”12 to denote the link to socialmovements, and the recognition that mobiliza-tion of those most affected can in itself changethe policy environment, in particular the publicdiscourse, to get specific problems and preferredsolutions onto public and policy agendas.It is here that the question of values comes

into play. Pastor and Ortiz specifically critiquethe process of “policy entrepreneurs” writingpapers and engaging policy makers withoutengaging and generating grassroots leadershipso that the social movement can “make sureto directly involve those with ‘skin in the game’and make sure that the frames and values arederived from them and not from focus groupsconducted by distant intermediaries”(p.2).13 Thechoice of the term “policy activist” aims to sig-nal the desirability, from a social justice pointof view, of building the capacity of individualsand groups who are part of or closely tied tograssroots movements, to play this role – to getsolutions onto the political agenda that matchproblems identified by those who are most

*For example, in relation to services such as health oreducation, ensure equity in their availability, accessi-bility, acceptability and quality.†Hence the term “social justice” is used broadly to incorpo-rate social, economic, cultural, civil and political rights.

B Klugman / Reproductive Health Matters 2011;19(38):146–162

147

B Klugman / Reproductive Health Matters 2011;19(38):146–162

marginalized or negatively affected by a policy. Measurable outcomes of policy advocacy

Thus to what extent potential beneficiaries ofpolicy are active in advocacy and to what extentthe content of resultant policy proposals reflectstheir perspectives, are both significant markersfor evaluating social justice advocacy.Donors, evaluators and, indeed, many advo-

cates tend to focus on a policy win. But policywins are usually the result of multiple strategiescoupled with windows of opportunity that arevery seldom predictable. In addition, the verysame factors that influence policy wins need tobe sustained in order to support policy implemen-tation, to address challenges to policy, and toachieve the ultimate goals of advocacy campaigns.A theory-of-change framework of measurable

outcomes, presented below, can be used by thosewho are supporting or undertaking social justiceadvocacy to reflect onwhether the diverse factorsthat influence change are being addressed, andwhether social justice values are being retainedin the process. It encourages, in particular, reflec-tion on whether the needs and aspirations ofthose who are most excluded in society remainat the forefront of advocacy campaigns. Withouta theory of change – without clarity about whatadvocacy hopes to achieve and what strategieswill be pursued to get there – it is not possibleto assess progress and adjust strategies.

A number of reviews of the literature on suc-cessful advocacy initiatives group together theoutcomes that advocacy campaigns tend to aimfor.7,14,15 Reisman et al call these groups “out-come categories”. The first four lay the ground-work for effective advocacy:

• strengthened organizational capacity,• strengthened base of support, and• strengthened alliances, which in turn draw on• increased data and analysis from a social jus-

tice perspective.

These four outcomes form the basis for con-ducting advocacy, sometimes quietly within thecorridors of power and sometimes from the out-side, through the mobilization of constituencies,public actions, and the engagement of the media.They enable the following outcome, which is amarker of significant progress in advocacy:

• the development of consensus around a commondefinition of the problem and possible policyoptions by an ever-widening constituency ofpeople (both of which will also evolve over timewith new insights, data, and constituenciesinforming them).

These, in turn, form the basis for the advocacymovement as a whole, comprising individuals,

148

organizations, and alliances that are continuallyadapting to changes in context, in order to ensurethe “readiness” of their organizational capacity,messages, and strategies. They enable effectiveengagement in the policy process, which fallswithin the sixth outcome category:

• increased visibility of the issue in policy pro-cesses, resulting in positive policy outcomes,including maintaining gains, and maintain-ing pressure through ongoing monitoring ofthe implementation of policy.

Ultimate impacts, usually beyond the timeframe of any grant or set of grants, would be:

• shifts in social norms, such as decreased dis-crimination against a specific group or increasedbelief that the state should provide high qualityeducation. That said, along the way, one maystart to see shifts in public understanding andvisibility of the issues, as the problem defini-tion or potential solutions gain social accep-tance over time; and

• shifts in population-level impact indicators,such as decreased violence against women,fewer suicides among gay youth, or increasededucational achievement among groups withhistorically poor achievement (see Figure 2,which draws on a range of sources).

B Klugman / Reproductive Health Matters 2011;19(38):146–162

149

What are the dynamics that influence out-comes in each category, and the kinds of out-comes donors, grantees, and evaluators mayseek from a social justice values perspective?The case of reproductive rights* advocacy inSouth Africa during the transition from apart-heid to democracy and in the 15-year periodthereafter – the “transition” and “post-apartheid”periods – provides a case study through which toexplore these dynamics. The information is basedon a range of policy analyses undertaken regard-ing the role of civil society in these periods,12,16–19

including a recent reviewof the state of civil societyadvocacy on these issues, conducted by the authorin collaboration with Khathatso Mokoetle.20

The case of reproductive rights advocacy inSouth Africa: 1990–2010From the mid- to late-1990s, donor support formobilizing grassroots constituencies, undertak-ing policy-oriented research, and establishingpolicy advocacy NGOs and coalitions played acritical role in the achievement of wide-rangingsexual and reproductive rights policy changesduring the era of transition from apartheid todemocracy in South Africa. Many donors werenot “reproductive rights” donors, but were moreconcerned with ending apartheid and buildingmovements to enable development under apart-heid and in the transition to democracy.The mobilization of civil society, coupled with

civil society leadership's entry into politicalpower and government, enabled the achieve-ment of a group of policies that are among thestrongest in the world from both public healthand human rights perspectives. The Choice onTermination of Pregnancy Act 1996 allows awoman to choose to terminate a pregnancywithin the first 12 weeks, or to do so in consulta-tion with a medical practitioner between 13 and20 weeks. It also allows midwives to conduct

abortions and does not require minors to getparental consent before having an abortion. Itenables the realization of a number of humanrights for women, in particular the right to life,liberty, autonomy, and security of the person; toequality and nondiscrimination; to privacy; tothe highest attainable standard of health (includ-ing sexual health); and the right to decide thenumber and spacing of their children. One resultof the legislation was better training of mid-wives. In addition, the Department of Healthestablished a National Committee for Confiden-tial Enquiries into Maternal Deaths, which pub-lishes reports on maternal deaths to enable theproviders of health care to review their guidelinesfor themanagement of common causes of death.21

Together these initiatives led to a 90% reduction inabortion-related maternal mortality by 2001,21

and a decline in the loss of dignity and autonomyassociated with women's lack of access to safeabortions. It had a particularly strong impact onmorbidity in young women.22

Policies were made and judicial findings pub-lished on a range of other related issues duringthis time as well, including a replacement of theexisting population control policy with a rights-based population policy,23 guidelines for con-traceptive provision24 and the introduction ofa cervical screening programme.25 By the late1990s, organizing around the HIV/AIDS pan-demic in the country also came into its own.Donor funding played a critical role in support-ing the legal, research, and grassroots capacitiesneeded to challenge a government in denial aswell as international pricing regimes that madeantiretroviral treatment for people in developingcountries unaffordable. Successful advocacy,including litigation, ultimately resulted in nationalroll-out of testing and treatment programmes.18

Yet, despite the fact that sexual and reproduc-tive rights issues are central to effective preven-tion of HIV transmission, and that people livingwith HIV/AIDS face daily challenges regardingtheir sexual and reproductive decision-making,these issues have slowly shifted from centrestage during the transitional period to outsidethe public health agenda in the post-apartheidperiod. As a result, since 2003, there has beenan increase in maternal deaths.26 This is attrib-uted to a number of factors, including hyper-tension, that could be prevented with strongerhealth service interventions and, most notably,

*In the South African context, the language of “repro-ductive rights” is used to describe the rights to accessi-ble, affordable, appropriate, and quality health services;to information; to autonomy in sexual and reproductivedecision-making; and to freedom from discrimination,coercion, and violence as they relate to reproduction. Theserights are undermined by poverty as well as by discrimina-tion on the basis of gender, race or ethnicity, sexual orien-tation, disability, and other bases of discrimination.24

B Klugman / Reproductive Health Matters 2011;19(38):146–162

150

deaths due to non-pregnancy-related infections,specifically HIV.27–29 The situation has beenexacerbated by a decline in access to abortionsin the public sector – from 45% of communityhealth centres providing services in 2008–09 to25% in 2009–10, in part due to a shortage ofnurses trained to perform first trimester termi-nations.30 Even efforts in the post-apartheidperiod have ground to a halt. For example,guidelines for the introduction of medical abor-tion that were drafted in 2004 were only final-ised in 2008, and at the time of writing onlyone province has officially implemented medi-cal abortion.31

Between 2002 and 2008, three of the leadingsexual and reproductive rights organizations inthe country – the Progressive Primary HealthCare Network, the Women's Health Project andthe Planned Parenthood Association closeddown, and a fourth, the Reproductive RightsAlliance, closed its office and let its staff go.*While the loss of these advocacy organiza-

tions does not explain the loss of service out-reach within the public health system, it doespartly explain why these failures were allowedto happen with impunity, and why issues ofsexual and reproductive rights were not incor-porated into the HIV/AIDS civil society andpolicy agendas. At moments of crisis, such asanti-abortion challenges to the abortion law atthe Constitutional Court and in Parliament,ad hoc mobilisation by legal groups, individualswho were active in the original legal process, andinternational NGOs such as Ipas has managed togarner enough momentum to provide inputs togovernment, politicians, and legal advocacyorganizations to keep the 1996 abortion lawin place. But they have not managed to garnersufficient institutional, public or political momen-tum to take forward the achievements of thetransition period.In reviewing the outcome categories asso-

ciated with advocacy processes, the factors that

facilitated and constrained the achievement andimplementation of reproductive rights policiesduring this period in South Africa are discussedby way of illustration.

Strengthened organizational capacityStrong organizational capacity in non-profitorganizations and coalitions is a pre-conditionfor successful social justice advocacy efforts.32

Hence assessing improvements in organisationalcapacity over time is a key evaluation outcomecategory. Organizational capacity involves anumber of essential components that donors,activists, and evaluators would assess: strategicand evaluation capacity with their associatedleadership capacity and ability to generate newleaders; fundraising capacity, and financialmanagement capacity; and networking andcommunications capacity.Less easy to measure, but arguably most impor-

tant, is the extent to which a leading organizationin an advocacy campaign or coalition has “adap-tive capacity”(p.135)33 – the capacity to learn asthe situation changes.34–36 This requires an inclu-sive style of leadership that is able to take anorganization and its coalition members througha process of reflection together.In relation to the case study, three of the four

reproductive rights organizations that closed inSouth Africa between 2002 and 2008 had lostthe strategic leadership that had enabled themto make an impact in the transition period. Noneof them had managed to groom second genera-tion leadership that would be able to adapt tothe post-apartheid environment – in particularleadership of women living with HIV/AIDS,given the enormity of the challenges HIV posesfor reproductive rights. In interviews with theleadership of the organizations that closed andwith members of other donor-funded sexual andreproductive rights groups, while some argued thatthe problem had been a lack of funds, most recog-nized that the overriding factor was a lack ofleadership, vision, and organizational capacity.20

While major policy victories had been achievedand the new challenge was to enable and monitorimplementation, organizations were not able toadjust to the changing environment, particularlyin terms of rethinking their strategies.Similarly, the reality of the HIV pandemic

required reproductive rights organizations toreassess the terrain and redefine their demands

*These were not the only groups advocating for sexualand reproductive rights and health, but were key nationalvoices with a strong rights orientation and sophisticatedstrategies that mobilised diverse constituencies in supportof these rights. Significantly, in the post-apartheid periodmore international NGOs established themselves in SouthAfrica, predominantly focusing on strengthening serviceprovision rather than on advocacy.

B Klugman / Reproductive Health Matters 2011;19(38):146–162

151

to take account of HIV and AIDS. It also requiredthem to shift their mode of engagement withdonors and even to change donors. Many donorswho were working to an anti-apartheid briefstopped funding in the country or, in the caseof bilateral and multilateral donors, shifted theirfunds to government. New donors were comingin, but in the health field, a large proportionwere concerned with addressing HIV as one ofthe country's biggest challenges and did notrecognize the centrality of sexual and reproduc-tive rights to preventing HIV.The emergence and growth of many HIV orga-

nizations was indeed appropriate for this time, andcritical in addressing the right to treatment forpeople living with HIV and AIDS. But allowingsexual and reproductive health and rights to falloff the agenda at the same time was a serious loss.

Strengthened base of support and alliancesThe “base of support” for an issue refers to grass-roots, leadership, and institutional support, whichincludes the breadth, depth, and influence ofsupport among the general public, interest groups,and opinion leaders. The social justice valuesdimension of building a base of support pertainsto the participation of those who are most affectedin defining the problem and potential policy pro-posals or options for implementing policy. Here,effectiveness overlaps with values. While the par-ticipation of those most affected is a values-basedprinciple of a social justice approach, it is alsopragmatic because of the need to ensure appropri-ate policies and maintain the degree of mobiliza-tion necessary for policy victories and for holdinggovernments accountable for implementation.As an advocacy process gets under way, dif-

ferent organizations representing different inter-ests and bringing in a wider range of insights,contacts, and relationships will almost certainlyneed to be mobilized.8 However, this breadth ofsupport also creates challenges. Alliances need tobe nurtured. The larger a coalition, the wider therange of interests held within it. If new perspec-tives, new research findings and new experiencesfrom the piloting of interventions are brought inby new allies, achieving consensus on the problemdefinition and possible policy options as part ofthe alliance-building process becomes more com-plicated. Howmuch the agreed solutions continueto represent the interests of those most affectedis an important question for evaluation, as the

dilution of policy demands is a risk in the processof seeking consensus among ever wider sectors.In the case study, the process of winning sup-

port for reproductive rights during the period oftransition involved mobilizing a wide range ofconstituencies, including workers, doctors andnurses, women (including rural women), youngpeople, lesbians, disabled people, faith-basedorganizations, and the structures of the newlylegalized African National Congress, includingits Women's League. These groups were drawntogether by the Women's Health Project17 in aprocess of deliberation about the reproductiveand sexual health and rights problems facingwomen. Through interactions between experts(both academics and practitioners) and theseorganized constituencies, policy proposals weredeveloped within months of the establishmentof democracy in April 1994. These expertsincluded many people who, during this periodof deliberation, were elected into parliamentor joined the new government administration.As a result, there was wide-ranging support forthe issues, and many contentious issues had beendebated and resolved before they ever got ontothe official political agenda. Given the breadthof the base of support and the diversity of allies,a substantial negotiation of values was necessary.In the debates about the impact of unsafe andillegal abortions on women's health and lives,for example, the need for safe abortion becameclear from a public health point of view andfrom the point of view of saving women's lives,because of high rates of death and illness causedby illegal abortions.37,38

At the time, the overarching frame was theneed to end discrimination on the basis of race.This provided a powerful motivation as, underthe previous regime, only those with accessto substantial resources – in effect, only whitewomen – had access to safe abortions, whetherlegally or illegally. Thus, ending discriminationwas the overriding value that enabled consensus.But beyond that, there were differences. Thedoctors involved in the deliberations arguedstrongly that only doctors should be able to doabortions, even though there is ample evidencein the clinical literature, including from SouthAfrica, that trained midwives and nurses cancarry out and manage safe abortions. Thisposition was more a reflection of doctors' profes-sional interests. In contrast, nurses and midwives

B Klugman / Reproductive Health Matters 2011;19(38):146–162

152

participating in the process saw the opportunityof increasing their areas of responsibility andpotential income. Women's rights groups recog-nized that the absence of doctors in rural areaswould mean that rural and poor women wouldcontinue to have unsafe abortions. Thus, the finalproposal, and the law that was ultimately passed,included allowing trained midwives to performabortions, thereby increasing access for thepoorest and most marginalized women.39

This example illustrates how the process ofbuilding a base of support and alliances requiresthe negotiation of values. From a social justiceperspective, it shows how the needs and rightsof those on the margins need to take precedence,and organizations representing them need to holdmore “mainstream” advocacy groups accountable.By 2006, however, there was no longer any

national organization or coalition systemati-cally bringing together groups concerned withreproductive rights in this way. Reproductiverights organizations describe a situation inwhich competition between groups had becomethe norm, particularly in a context of scarcity ofdonor funds for this work.20 Equally signifi-cantly, few of these organizations had retainedor built formal and ongoing linkages with grass-roots constituencies, such as HIV and AIDSorganizations, let alone managed to mobilisethem to express moral outrage at the dire stateof women's reproductive health and how theHIV pandemic was exacerbating it. Reproduc-tive rights networks in fact declined, as didtheir means of communication, such as sharednewsletters. Also, many of those who had beenemployed in non-governmental organizations(NGOs) and research institutions focusing onreproductive rights were now working on HIVand AIDS, without taking sexual and reproduc-tive rights concerns into those spaces.The lack of attention to sexual and repro-

ductive rights in the context of HIV requireda major refocusing of attention by reproductiverights activists as to who needed to be drawninto a base of support and alliances for repro-ductive rights, in order to keep reproductiverights demands on the public and politicalagenda. The HIV terrain was itself fraught withconflict over strategies, particularly the balanceof attention between prevention and treat-ment, and there was competition betweengroups.40 This no doubt exacerbated the diffi-

culty of reproductive rights groups in broaden-ing their alliances.One of the critical social justice values that

supports the establishment and maintenance ofalliances is collaboration, and this is an areawhere donor expectations can exacerbate asituation. Given the complexity of theories ofchange in relation to policy advocacy, no oneorganization should ever be expected or expectitself to deliver all of the outcomes, but ratherthat a mix of organizations would collectivelywork to achieve these. Collaboration betweenorganizations in recognition that they eachhave something to contribute, rather thancompetition for attribution of victories, is notonly a “good” value, but is what is needed tobe effective. Effective advocacy puts learningand working together above competition. Thisis a challenge for advocacy groups, particu-larly in the context of competition for scarcedonor resources.*It also poses one of the biggest challenges to

donors – how to create enough financial securityfor a group of organizations that they are willingand able to acknowledge and support each others'strengths, without fearing that their own con-tributions will be devalued. And above all, thesocial justice value of participation – the inclu-sion of all the relevant issues and the peoplewho are most affected – is particularly importantas the range of allies increases and strategic com-promises are made. This holds true even if policyvictories are won; organizations representingthose at the margins need to establish sustainedmechanisms for holding decision-makers –whether government, donors, or even nationalNGOs − accountable for their decisions.

Increased data and analysis from a socialjustice perspectiveEvidence, whether in the form of research find-ings or personal testimonies, is frequently help-ful and sometimes essential to strengthen thecapacity of the base and allies to pin down theproblem definition, to shape and test potentialpolicy options, and to analyze a changing policyenvironment. Hence, to what extent information

*That said, there are times when groups have fundamentaldifferences in their goals and strategies, and a donorrequirement for collaboration could be equally problematic.

B Klugman / Reproductive Health Matters 2011;19(38):146–162

153

and analysis are available and used to informadvocacy is a question for assessing the develop-ment of an advocacy strategy. Both the mediaand policymakers may feel that research find-ings legitimize policy demands. Producing dataand analysis may also carry a significant valuesdimension to the extent that it provides infor-mation on previously ignored groups or is pro-duced by and carries the perspectives of suchgroups. Moreover, the orientation and perspec-tive of researchers influences what questionsthey ask and how they interpret their findings.So research should be generated by or in closerelationship with those who are the target ofpolicies, to ensure accurate information onwhich to shape policy options.A strategy that donors can use here is enabling

advocacy groups to commission the necessaryresearch so that the findings are more likelyto be used. On the other hand, gathering databeyond those needed immediately for advocacycan be essential for gaining deeper insights intoproblems, policy options, and indeed the natureof the policy terrain. Research may need to bedone over some months or even years, and there-fore cannot be commissioned at the momentthe findings are needed. Advocacy NGOs andmembership-based organizations also need tobe identifying research questions and allyingwith research entities in order to get researchdone. Evaluating the information dimensionof advocacy requires not only looking at itssource, but also its effectiveness in advocacy.41

The lack of linkages between researchers andresearch findings with groups able to use researchfor advocacy is usually the result of a poor theoryof change, such as the assumption that researchershave an interest in their findings being used topromote change, or the assumption by researchersthat new evidence will inevitably lead to improvedpolicy outcomes.42 As a result, researchers oftenfail to put an effort into building the necessarylinkages with a base of support and allies onthe issues.In South Africa in the transition period, advo-

cates put substantial energy into identifyingand forming alliances with researchers withkey knowledge. For example, a historian wasable to give evidence in Parliament about thelong history of abortion among all ethnicitiesin the country, so that it could not be con-strued as “un-African”. Also, reproductive health

researchers anticipated the need for healthsystems data that would inform politiciansabout the costs of unsafe abortion to the publichealth system, and produced it before the issuewent to Parliament. Advocates also gatheredkey anecdotal evidence, for example, from reli-gious women whose clergy had supported themin having abortions, thus undermining the viewthat abortion was necessarily “anti-Christian”.Legal experts studied international law to for-mulate arguments about constitutionality andlegal interpretation. All of this helped to framemessages for the media and in policy debates inways that kept the issues of discrimination againstwomen and public health at the forefront. Thisillustrates the tight relationship that was deve-loped between researchers and activists as wellas the strategic nature of research. In addition, anumber of research institutions, sometimes incollaboration with reproductive rights NGOs,developed and evaluated methods for buildinghealth service and community support for repro-ductive rights and particularly for abortion.43,44In contrast, some of the gaps identified in the

post-apartheid period are the lack of connectionbetween activists and researchers, and the lackof a forum for identifying research priorities andsharing research findings as the basis of reproduc-tive rights advocacy.20 As a result, there is nomechanism for bringing new research findingsinto NGO and community activism. There hasalso been extremely limited mobilization to getgovernment and civil society groups to use thetested methods developed of building communityand health worker support for implementing thereproductive rights policies described above.*There has been one initiative, conducted underthe auspices of a project to develop treatmentguidelines for HIV-positive women, which ranan e-list that fostered debate on reproductiverights issues as they pertain to HIV/AIDS andestablished a collaborative projectwith researchers,clinicians, and some activists to develop guide-lines on specific topics. But it, too, has faltered,as the host organization expressed discomfortwith the “movement-building” and advocacydimensions of the project, wanting to run it asa more dispassionate research project.

*Although the NGO Ipas has made some headway in train-ing public health services in parts of some provinces.

B Klugman / Reproductive Health Matters 2011;19(38):146–162

154

This disconnect – between those who seeresearch as closely linked with advocacy andmobilization and those keen to keep researchpolitically neutral – is a common problem facedby advocates and a reason why advocacy theo-ries of change need to include it as an outcomearea worthy of continual monitoring.45,46

Increased support for a specific problemdefinition and policy optionsThis takes me back to the values dimension ofresearch for advocacy. Allan, McAdam andPellow talk about the need for a “motivationalframe” that will persuade people to take action,making them want to get involved.47 This iswhere research findings are reframed to supportcampaign language. Activists frame issues fortheir base of support and allies and try to gettheir frame, or perspective, taken up by themedia and those in a position to influencepublic attitudes and government policy.48

Increased standardization in the articulationof the problem and potential policy options isa marker of the coalescing of a base of supportand alliances, particularly as numbers grow.In the search for early indicators of effectiveadvocacy, the ability to cohere a growing groupof people who recognize a problem, and thencome to agreement around a specific problemdefinition that is drawn from and talks to theexperience of those most affected, is key. Itbrings together the efforts in the previous out-come categories and is therefore a solid indica-tor of movement towards the goal(s). Monitoringhow a problem definition and policy options arerenegotiated as new allies are found or as thepolitical, social, or economic context changes,and the extent to which they remain true to theconcerns of those suffering the greatest dis-crimination or lack of access to resources, is coreto the maintenance of a social justice perspec-tive in evaluation. Changes in the degree ofsupport for the activists' problem definitionor policy option by policy thinktanks or othergroups with resources and power that activistshave targeted, are an indicator of the effective-ness of their advocacy.As noted in South Africa during the transition

period, reproductive rights activists were slow inrecognising and responding to the scope of theHIV crisis.49 HIV activists were having to copewith stigma, deaths of leaders and a government

in denial. Leadership of the HIV movement waspredominantly by men whose perspectives onHIV struggles were influenced by the early framingof HIV in the US,48 which gave little attention tothe dynamics of heterosexual sexual relationshipsin fostering HIV transmission. In South Africa in2001, reproductive rights activists and lawyersdecided towork in alliancewithHIV/AIDS activistsand lawyers to ensure that potential litigationfor the implementation of programmes to pre-vent mother-to-child transmission of HIV wouldbe framed around women's reproductive rights,including the right to have healthy babies. Butover time the lawyers let go of this frame in favourof arguments based on the right to health care, andparticularly to HIV treatment.16 In general, AIDSactivists' dominant claimwas for the right to treat-ment, which “constructed women as bearers ofchildren, and as patients, rather than as activeagents in their own right”.16 At times women wereblamed for the escalation of HIV and in general thefocus on preventing mother-to-child transmissionfailed to acknowledge the breaches in sexual andreproductive rights that made women particularlyvulnerable,50 a phenomenon not particular toSouth Africa.51 The leading HIV organizations –the Treatment Action Campaign – was successfulin mobilising activities that brought media atten-tion and creating a “moral consensus” regard-ing the right to treatment.40 In the context ofAIDS denialism, much of the media focus wason the conflict between the government and HIVactivists,48,52 with very little attention to the livedrealities of people living with HIV, nor to the highlevels of sexual violence,53 and cultural impera-tives to have children, all of which were key deter-minants in the escalation of HIV. Claims regardingthe need to promote sexual and reproductiverights as a key dimension to preventing HIVbrought a level of complexity that this call couldnot contain, or, argued differently, neither AIDSactivists nor reproductive rights activists wereable to frame these issues in ways that caughtthe public and media imagination. Hence, a criti-cal opportunity for broadening and deepeningpublic understanding and legal precedent regard-ing the scope of reproductive rights and women'srights in particular, was lost. In the process,public and policy recognition of the right to treat-ment eclipsed the issues underlying the HIV/AIDSepidemic, in particular the lack of mutuality insexual and reproductive relationships.

B Klugman / Reproductive Health Matters 2011;19(38):146–162

155

There was information, there were researchers,but there was no frame around which to build acollective understanding and set of demands. Inthe post-apartheid era and in the context of thepandemic, a new frame is needed. In the recentassessment of the state of reproductive rightsorganizing in South Africa,20 what became clearwas that there is no longer a shared problemdefinition among remaining reproductive rightsadvocates. Each interviewee had a somewhatdifferent focus: the absence of attention to men,the absence of attention to sexual orientation, theabsence of lesbians' voices among those con-cerned with sexual orientation, the absence ofattention to young people's interests, the absenceof attention to women living with HIV, and soon.20 At this writing a new initiative is underwayto bring these together under a broader frame, inrecognition that mutually respectful sexual andreproductive relationships and the ability of allpeople to make decisions about their sexual andreproductive lives are fundamental to achievingequality for all people and to effectively prevent-ing HIV and AIDS.*Struggles aroundmeaning, around what aspects

of an issue are most pressing and should havepriority in policy demands, are common. Whatthe media chooses to cover brings further com-plexity. For this reason, tracing shifts in problemdefinition and policy options, and the extent towhich they garner increased support by diverseconstituencies, is a helpful indicator for activistsand donors to assess progress.

Increased visibility of the issue in policyprocesses resulting in positive policy outcomesThe issue of “readiness” is particularly impor-tant in policy advocacy because the actualmoments for policy change often come and goas political and economic contexts change. It ismuch easier for policy activists to move theiragenda when there are changes in context thatcreate windows of opportunity,9 for example in

the run-up to elections or after the election of anew party or president. Windows can also becreated through activism54 – mobilizing publicconcern about an issue, or using litigation toforce state action, as occurred in South Africain the struggle for AIDS treatment.18 This iswhy solid and coordinated strategies amonggroups and coalitions aiming for change are soimportant. Windows of opportunity are criticalmoments for policy activists seeking to push atransformative agenda, compared to “politicsas usual”, where they may only be able to pushfor small improvements to existing policy. Shawdescribes the need for “tactical activism”,55

where policy advocates use a window of oppor-tunity by finding a way to link their solution toa problem that is on the political agenda. Suchtactical activism assumes advocacy groups arealready prepared so that they can take advan-tage of any such windows.Winning the right of civil society to partici-

pate in certain policy forums is a critical out-come in itself. Achieving representation orparticipation specifically of marginalized groupsin those forums, is an additional critical out-come. Once the right to participation has beenwon, a quality outcome would be the ability ofthese participants to be heard in these forums.Groups will frequently claim participation asan achievement, which it is. But the moreimportant question is whether the groups' par-ticipation has been of a quality that it is beingtaken seriously by other stakeholders, whatevertheir position on the topic, and, whether it isinfluencing the debate, and ultimately, whetherit has “strengthened the accountability of stateinstitutions to civil society groups”(p.51).56

In South Africa, the high degree of access ofcivil society reproductive rights groups to policy-makers during the transition period was a keyfactor in the movement's success. And in thepost-apartheid period, one of the victories ofthe HIV/AIDS movement has been the right toparticipation. The South African National AIDSCouncil (SANAC) comprises sectors where groupsrepresenting diverse interests – women, youth,men, people living with HIV/AIDS – can advo-cate for their issues and participate in nego-tiations and debates with government aboutpolicy and its implementation. There is no simi-lar forum for reproductive health and rights.But, given the prevalence of HIV/AIDS, SANAC

*The review of civil society perspectives referred to earlier20

has resulted in a decision to establish a new organizationscalled the Sexual Health and Rights Initiative – SouthAfrica (SHARISA) to help reframe the issues and mobilisediverse constituencies around an analysis of the connec-tions between diverse struggles pertaining to sexual andreproductive rights, including HIV.

B Klugman / Reproductive Health Matters 2011;19(38):146–162

156

is an essential forum for bringing reproductiverights issues to the fore. Yet the existing sectorgroups have not brought it into the debate. Thewomen's sector – which is one where one mightexpect to see reproductive rights issues raised –has been noticeably disorganized and silent onthese issues, reflecting the lack of organizationalcapacity, base, allies, and shared message.Beyond evaluating readiness, donors, advo-

cates, and evaluators would also need to assessif and how the policy message gains traction ina policy agenda and policy debate, and whetherincreased numbers of policymakers show aninterest in and ultimately take up a social justiceperspective on the issue. Ultimately, they wouldbe looking to see this perspective being adopted,funded and implemented, with effective mecha-nisms for monitoring implementation. Notethat a positive policy outcome may not be a newpolicy but “maintaining the status quo”, wherean existing policy that supports the advocacycoalition's values has been under threat.14,57 It isalso worth monitoring unexpected victories andhow these came about. Once a policy is won,ongoing reflection on the theory of change wouldbe needed to assess the extent to which advocacyis effectively targeted to ensure that the newpolicy is resourced to ensure equity in implemen-tation. Policy advocates have to continue towatch the political process, because policies cannot only fail through lack of implementationbut can also be overturned at any time. This iswhy grantmaking to influence policy has toassume long-term planning and commitment.

Shift in social normsIn the long term, to sustain policy victories, oneneeds to build public support for an issue. Hence,the identification of shifts in social norms is akey indicator of long-term impact. In the earlystages of an advocacy process, evaluation mightassess this outcome in terms of greater visibilityof a social justice perspective on the policy issue.As the campaign strengthens, building publicoutcry over an issue can be a means for policyactivists to create a window of opportunitywhere politicians are pushed to take up a policyissue. But despite the conventional wisdom thatpublic communication strategies are essentialto successful advocacy, visibility alone is not adeterminant of policy success. Public opinionmay or may not be amenable to change through

media influence.58,59 Indeed, sometimes thevisibility of a highly provocative issue canmake it harder to maneuver it into a policy pro-cess, because visibility serves to mobilize theopposition – as is frequently the case in relationto abortion. In addition, one sometimes achievessocial justice policy victories without majoritysupport of the public; hence, there is no causalpredictability between opening public debateand winning policy change, let alone implemen-tation. Where dominant social norms are not insupport of an issue, policy activists may decidenot to engage the public at large. Advocates needto ensure that in developing their theory ofchange for a specific advocacy goal, they inter-rogate whether engaging the public throughmedia and other public spaces would be a helpor hindrance in creating a conducive policy envi-ronment for the desired change.In South Africa, the change in the abortion

law in 1996 was won in the context of an over-arching frame of ending discrimination and itsresultant impacts on women's health and lives. Anumber of doctors who were also anti-apartheidactivists took positions of power in the firstdemocratic government – the Minister of Health,the head of the Health Portfolio Committee inParliament, and the head of Maternal Health inthe Department of Health. In addition, reproduc-tive rights organizations had mobilised grass-roots organizations around this issue and wonthe support of their leadership.17 However,despite the prevalence of abortion across all reli-gious and cultural traditions in the country, it isa silent phenomenon and causes disquiet whenbrought into the public sphere. Women whoknow about the availability of free abortions,and can afford transport to the few facilities pro-viding them, make use of them. Organizing suchwomen and the public in general to speak outabout the lack of services is quite another chal-lenge. Reproductive rights activists recognisedthe need to build public and bureaucratic sup-port, and most particularly support of health careproviders. They developed and tested interven-tions to build community understanding ofreproductive rights, ways of preventing preg-nancy, and the limited role of abortion withinthat, such as Communities for Choice,43 andvalues clarification.60 These proved effective.But, neither these nor subsequent activists man-aged to implement these interventions on a wide

B Klugman / Reproductive Health Matters 2011;19(38):146–162

157

scale, nor to persuade government to do so. Theyhave similarly made very limited inroads inbuilding support among nurses and midwives,and here, too, well-tested interventions locallyand internationally, such as Stepping Stones,61Health Workers for Change62 and values clarifi-cation63 have not been institutionalized. In par-ticular, the idea of the right of people to choosewhether or not to have children is a critical framefor the issues, and it has not been won. As a result,one irony of the new abortion law is that nursesoften put pressure on women living with HIV tohave abortions or to be sterilized after givingbirth, not out of support for women's right tomake reproductive choices, but as a result of thestigma of HIV and AIDS. The failure of govern-ment to implement is itself a product of dominantsocial norms. Only when there have been strongadvocates within government has implementa-tion been strengthened.31

This illustrates the complexity of the chal-lenge of developing an effective frame withinwhich to promote reproductive rights in this con-text, and the broader point that shifting publicnorms is a particularly context-specific challenge.

Changes in impactThis final outcome category refers to the hoped-for impacts of policy change on the lives of thepopulation and the conditions under which peopleare living.15 From a social justice point of view,one would be looking for declines in discrimina-tion against and stigmatization of specific groupsof people, increases in equity of distribution ofresources across the population as a whole, e.g.in access to health care, and the institutionaliza-tion of mechanisms for participation in policy-making and monitoring. Most important wouldbe sustaining these improvements over time.The South Africa case serves to reinforce the

importance of maintaining an eye on advocacygoals over the long run and remaining vigilant,even after policy-level victories, to keep onstrengthening the work in each of the outcomecategories so that impacts are monitored overtime, and strategies tomaintain and improve theseare re-oriented if and when contexts change.

ConclusionWhile policy change itself is easy to monitor,the complexity and unpredictability of imple-

menting policy change raises questions abouthow to plan and strategize for such changes,and how to monitor whether any progress isbeing made. The theory of change presented inthis article offers donors, advocacy organiza-tions, and evaluators a way of conceptualizingthe process of change and a range of outcomesthat can be assessed in an ongoing way. Theyneed to monitor whether existing organizationaldevelopment, mobilisation of a base and alliesand advocacy strategies are ensuring that orga-nizations are able to effectively shift publicopinion and ready to engage policymakers andimplementers about a problem and potentialpolicy or implementation options as opportu-nities arise. Hence the focus of both strategicplanning and evaluation is on the “steps thatlay the groundwork”,8 shifting the policy envi-ronment, and thus contributing to the achieve-ment of policy change and implementation.Monitoring of impacts, too – whether shifts insocial norms or in population-level indicators –while frequently too distant to measure withinthe time-frame of annual grants, or even five-or ten-year periods, can be helpful in identify-ing where policy victories do not appear to beresulting in effective implementation.The primary lesson illustrated by the South

Africa case study is that social justice goals arefrequently complex and require long-term invest-ment and ongoing evaluation. In South Africa,demands to address AIDS treatment took overpublic and political space in ways that ignoredthe importance of sexual and reproductive rightsin their own right and in preventing the spread ofHIV. This drew attention away from these issuesdespite the slew of new laws that required imple-mentation to protect public health. Would a morereflective process along the way have preventedthese losses? From the perspective of donors,evaluators, and, above all, those hoping to revi-talize this movement, there are critical lessons inrelation to all of the outcome categories. Mostparticularly, the failure of the reproductive rightsmovement to rethink its theory of change asthe context changed provides lessons for socialjustice advocates in many contexts.The approach to policy advocacy described in

this article is in contrast with the traditionalview of policy processes as linear.64 Rather, itrecognizes that those engaged in policy advo-cacy processes – whether as donors, advocates,

B Klugman / Reproductive Health Matters 2011;19(38):146–162

158

or evaluators – need to monitor changes incapacity in all outcome categories and create acontinuing process of analysis and learning forthe advocacy initiative, assessing the appropri-ateness of the theory of change in relation tochanges in the political and organizational con-text. If the only focus of evaluation were thepolicy outcome, there would be no reason forassessing readiness, and, given that a policyvictory may take years, even decades to achieve,and yet more to implement, the work alongthe way would be devalued – just as the needfor ongoing post-policy advocacy would beforgotten. Moreover, failure to recognize thecomplexity of social change, particularly withhighly contested issues, allows donors to imaginethat short-term grant making should deliver andsustain policy victories – something for whichthere is no evidence. The base and allies needto be sustained, renewed and expanded overtime, problem definitions and policy or imple-

mentation options continually re-assessed, mes-sages reshaped, and advocacy sustained. Inaddition, the perspectives, interests, participationand agency of those most affected by the issuehave to be kept at the heart of these processes.

AcknowledgementsThis paper was originally published in theFoundation Review 2011;2(3):94–107 and isreprinted here with their kind permission. TheFoundation Review retains the copyright. Thisis a revised and updated version. I was Directorof the Women's Health Project from 1991 to2000. I want to thank the Ford Foundation forgiving me two months' study leave in 2009,during which I investigated these issues. I wouldalso like to thank Marion Stevens for providinghelpful input on an earlier version of the paper,and Gail Andrews and Stefanie Röhrs for theirextremely helpful comments on this version.

References1. Fraser N. Justice Interruptus:

Critical Reflections on the“Postsocialist” Condition.New York: Routledge; 1997.

2. Fraser N. Social Justice in theKnowledge Society:Redistribution, Recognition,and Participation. Beitrag ZumKongress “Gut zu Wissen.”Heinrich-Boll-Stiftung.26 October 2006. At: <http://wissensgesellschaft.org/themen/orientierung/socialjustice.pdf>.Accessed 17 August 2009.

3. Gruskin S, Ferguson L. Usingindicators to determine thecontribution of human rights topublic health efforts: Why?What? and How? Bulletin ofWorld Health Organization2009;87:714–19.

4. Christie C, Alkin M. Theuser-oriented evaluator's rolein formulating a programtheory: using a theory- drivenapproach. American Journalof Evaluation 2003;24:3.

5. Mackinnon A, Amott N.Mapping a change: using atheory of change to guideplanning and evaluation.

GrantCraft 2006;15. At:<www.grantcraft.org/index.cfm?pageId=808>. Accessed14 August 2009.

6. Kingdon J. Agendas, Alternatives,and Public Policies. New York:Longman; 2002.

7. Coffman J. Advocacyevaluation trends and practiceadvice. Advocacy ImpactEvaluation Workshop. Seattle:Gates Foundation. MarcLindenberg Center, Universityof Washington, EvansSchool of Public Affairs;2 December 2007.

8. Guthrie K, Louie J, David T,Foster C. The challenges ofassessing policy and advocacyactivities: strategies for aprospective evaluationapproach. CaliforniaEndowment, 2005. At:<www.calendow.org/reference/publications/pdf/npolicy/51565_CEAdvocacyBook8FINAL.pdf>.Accessed 15 September 2009.

9. Leat D. Theories of SocialChange. International Networkon Strategic Philanthropy.

Paper 4. Washington, DC:Bertelsmann Stiftung; 2005.At: <www.peecworks.org/PEEC/PEEC_Inst/0021E1B7-007EA7AB.0/Leat%202005%20Theories_of_change.pdf>.Accessed 15 September 2009.

10. Puttenney D. Measuring socialchange investments. Women'sFunding Network, 2002. At:<www.wfnet.org>. Accessed16 August 2009.

11. Klugman B. Module 5: Policy.In: Ravindran S, editor.Transforming Health Systems:Gender and Rights inReproductive Health: Atraining curriculum for healthprogramme managers.WHO/RHR/01.29. Geneva:World Health Organization;2002. p.299–380. At:<www.who.int/reproductive-health/publications/transforming_healthsystems_gender/index.html>. Accessed15 September 2009.

12. Klugman B. Empoweringwomen through the policyprocess: the making of healthpolicy in South Africa. In:

B Klugman / Reproductive Health Matters 2011;19(38):146–162

159

Presser H, Sen G, editors.Women's Empowerment andDemographic Processes: MovingBeyond Cairo. Oxford:Oxford University Press; 2000.p.95–118.

13. Pastor M, Ortiz R. Makingchange: how social movementswork and how to support them.Los Angeles: Program forEnvironmental and RegionalEquity, University of SouthernCalifornia; 2009. At: <www.calendow.org/uploadedFiles/Publications/Policy/General/Making%20Change%20-%20March%202009.pdf>.Accessed 18 September 2009.

14. Korwin L. The Catalyst FundTheory of Change. SanFrancisco: Korwin Consultingfor the Tides Foundation; 2009.

15. Reisman J, Gienapp A,Stachowiak S. A Guide toMeasuring Advocacy andPolicy. Seattle: OrganizationalResearch Services; 2007.

16. Albertyn C, Meer S. Citizens ormothers? The marginalizationof women's reproductive rightsin the struggle for access tohealth care for HIV-positivepregnant women in SouthAfrica. In: Mukhopadhyay M,Meer S, editors. Gender Rightsand Development: A GlobalSource Book, Gender Societyand Development Series.Amsterdam: Royal TropicalInstitute; 2008. p.27–55. At:<www.kit.nl/net/KIT_Publicaties_output/ShowFile2.aspx?e=1456>. Accessed9 December 2010.

17. Klugman B, Stevens M, ArendsK. Developing women's healthpolicy in South Africa from thegrassroots. Reproductive HealthMatters 1995;3(6):122–31.

18. Marcus G, Budlender S. Astrategic evaluation of publicinterest litigation in SouthAfrica. Johannesburg: AtlanticPhilanthropies; 2008.

19. Robins D. From revolution torights in South Africa: socialmovements, NGOs and popular

politics after apartheid.Pietermaritzburg: University ofKwaZulu-Natal Press; 2009.

20. Klugman B, Mokoetle K. Reporton civil society's engagementwith sexual and reproductivehealth and rights andopportunities for identifying anIPPF affiliate in South Africa.International PlannedParenthood Federation, AfricaRegion. 2010. Unpublished.

21. Department of Health. Savingmothers. Second report onconfidential enquiries intomaternal deaths in South Africa1999–2001. Pretoria:Department of Health; 2003.

22. Jewkes R, Rees H, Dickson K,et al. The impact of age on theepidemiology of incompleteabortion in South Africa afterlegislative change. BJOG 2005;112:355–59.

23. Republic of South Africa. WhitePaper on Population Policy.Pretoria: Ministry for Welfareand Population Development;April 1988. Notice 1930 of1998. Government Gazette399(19230). 7 September 1998.

24. Department of Health.National Contraception PolicyGuidelines – Within aReproductive HealthFramework. Pretoria:Department of Health; 2001.

25. Department of Health. NationalGuidelines for Cervical CancerScreening Programme. Pretoria:Department of Health; 2000.

26. National Committee on aConfidential Inquiry intoMaternal Deaths in SouthAfrica. Saving Mothers2005–2007: Fourth Reporton Confidential Inquiriesinto Maternal Deaths in SouthAfrica – Expanded ExecutiveSummary. Pretoria: NationalDepartment of Health; 2007. At:<www.doh.gov.za/docs/reports/2007/savingmothers.pdf>.

27. Bradshaw D, Chopra M, KerberK, et al. Every Death Counts.Pretoria: South African NationalDepartment of Health; 2008.

28. Anderson B, Phillips H. Adultmortality (age 15–64) based ondeath notification and data inSouth Africa: 1997–2004.Pretoria: Statistics South Africa;2005. At: <www.statssa.gov.za/publications/Report-03-09-05/Report-03-09-052004.pdf>. 2006.

29. Khan M, Pillay T, Moodley J,et al. Maternal mortalityassociated with tuberculosis–HIV-1 co-infection in Durban,South Africa. AIDS 2001;15:1857–63.

30. National Department ofHealth. Annual Report2009/10. 2010:3. Pretoria.

31. Margaret Hoffman, Professor,Women's Health Research Unit,School of Public Health andFamily Medicine HealthSciences Faculty, University ofCape Town. Personalcommunication, March 2011.

32. Devlin-Foltz D. Impactevaluation tools and methods:learning to live withimprecision and like it (mostly).Advocacy Impact EvaluationWorkshop. Seattle: Bill &Melinda Gates Foundation;2 December 2007.

33. Letts C, Ryan W, Grossman A.High Performance NonprofitOrganizations: ManagingUpstream for Greater Impact.New York: Wiley; 1999.

34. Martineau J, Hannum K.Handbook of LeadershipDevelopment Evaluation. SanFrancisco: Jossey-Bass/Centerfor Creative Leadership; 2007.

35. Ospina S, Schall E. Leadership(re)constructed: how lensmatters.Washington, DC: APPAMResearch Conference; 2001.

36. Heifetz R, Laurie D. The workof leadership. HarvardBusiness Review 1997;January–February:124–34.

37. Rees H, Katzenellenbogen J,Shabodien R, et al. Theepidemiology of incompleteabortion in South Africa. SouthAfrican Medical Journal 1997;87:432–37.

B Klugman / Reproductive Health Matters 2011;19(38):146–162

160

38. Jewkes R, Brown H,Dickson-Tetteh K, et al.Prevalence of morbidityassociated with abortionbefore and after legalisation inSouth Africa. BMJ 2002;324:1252–53.

39. Klugman B, Varkey S. Frompolicy development to policyimplementation: the SouthAfrican Choice on Terminationof Pregnancy Act. In:Klugman B, Budlender D,editors. Advocating forAbortion Access: ElevenCountry Studies. JohannesburgInitiative, Women's HealthProject, School of Public Health,University of Witwatersrand;2001. p.251–82.

40. Friedman S, Mottiar S. A moralto the tale: the Treatment ActionCampaign and the politics ofHIV/AIDS. Center for CivilSociety, University ofKwaZulu-Natal; 2004.

41. Boaz A, Fitzpatrick S, Shaw B.Assessing the impact of researchon policy: a review of theliterature for a project onbridging research and policythrough outcome evaluation.Policy Studies Institute. London:Kings College; 2008. At:<www.psi.org.uk/pdf/2008/bridgingproject_report_with_appendices.pdf>. Accessed20 September 2009.

42. Young J. Impact of research onpolicy and practice. 2008. At:<www.capacity.org/en/content/pdf/4877>. Accessed 15September 2009.

43. Varkey S, Fonn S, Ketlhapile M.The role of advocacy inimplementing the South Africanabortion law. ReproductiveHealth Matters 2000;8(16):103–11.

44. Mitchell E, Trueman K, GabrielM, et al. Building alliances fromambivalence: evaluation ofabortion values clarificationworkshops with stakeholders inSouth Africa. African Journal ofReproductive Health 2005;9(3):89–99.

45. Sutton R. The Policy Process:An Overview. London: OverseasDevelopment Institute; 1999.

46. Trostle J, Bronfman M, LangerA. How do researchers influencedecision-makers? Case studiesof Mexican policies. HealthPolicy and Planning 1999;14(2):103–14.

47. Allan C, McAdam D, Pellow D.What is the role of civil societyin social change? SuccessfulSocial Movements. 2010. At:<http://chrisallan.info/wp-content/uploads/2010/04/Supporting-successful-movements3.pdf>. Accessed26 September 2010.

48. Jacobs S, Johnson K. Media,social movements and the state:competing images of HIV/AIDSin South Africa. African StudiesQuarterly 2007;9(4). At:<www.africa.ufl.edu/asq/v9/v9i4a8.htm>.

49. Mbali M. A long illness: towardsa history of NGO, governmentand medical discourse aroundAIDS policy-making in SouthAfrica. Submitted in PartialFulfilment of an HonoursDegree in History. Durban:University of Kwa-Zulu/Natal;December 2001.

50. Stevens M. From HIVprevention to reproductivehealth choices: HIV/AIDStreatment guidelines for womenof reproductive age. AfricanJournal of AIDS Research 2008;7(3):353–59.

51. Eyakuze C, Jones D, Starrs A,et al. From PMTCT toa more comprehensive AIDSresponse for women: amuch-needed shift. DevelopingWorld Bioethics 2008;8(1):33–42.

52. Schneider H. On the fault-line:the politics of AIDS policy incontemporary South Africa.African Studies 2001;61:1.

53. Fassin D, Schneider H. Thepolitics of AIDS in South Africa:beyond the controversies.BMJ 2003;326:495.

54. Westley F, Zimmerman B,

Patton M. Getting to Maybe:How the World is Changed.Toronto: Vintage Canada; 2007.

55. Shaw R. The Activist'sHandbook: A Primer. Berkeley:University of CaliforniaPress; 2001.

56. Chapman J. Monitoring andevaluating advocacy. PLA Notes2002;43:48–52. At: <www.planotes.org/documents/plan_04316.pdf>. Accessed15 September 2009.

57. Helzner J. Guidelines onmonitoring and evaluation:considerations for projectdesign, implementation, andreporting. Chicago: John Dand Catherine T MacArthurFoundation; 2006.

58. Newton K. May the weakforce be with you: the powerof the mass media in modernpolitics. European Journalof Political Research 2006;45:209–34.

59. Souter D. The challenge ofassessing the impact ofinformation andcommunications ondevelopment. Buildingcommunication opportunitiesbriefing. At: <www.bcoalliance.org/system/files/BCO_Synthesis_web_EN.pdf>.Accessed 7 September 2009.

60. Mitchell E, Trueman K, GabrielM, et al. Building alliances fromambivalence: evaluation ofabortion values clarificationworkshops with stakeholdersin South Africa. AfricanJournal of Reproductive Health2005;9(3):89–99.

61. Jewkes R, Nduna M, Levin J,et al. Evaluation of SteppingStones: a gender transformationHIV prevention intervention.Policy Brief. Pretoria:Medical Research Council;March 2007. At: <www.mrc.ac.za/policybriefs/steppingstones.pdf>.

62. Onyango-Ouma W, Laisser R,Mblilma M, et al. Anevaluation of Health Workersfor Change in seven settings:

B Klugman / Reproductive Health Matters 2011;19(38):146–162

161

a useful management andhealth system developmenttool. Health Policy andPlanning 2001;16(Suppl.1):24–32.

63. Mitchell E, Trueman K,

Gabriel M, et al. Acceleratingthe pace of progress in SouthAfrica: an evaluation of theimpact of values clarificationworkshops on terminationof pregnancy access in

Limpopo Province.Johannesburg: Ipas; 2004.

64. Grindle M, Thomas J. PublicChoices and Policy Change.Baltimore: Johns HopkinsUniversity Press; 1991.

RésuméCet article propose un cadre pour la théorie duchangement dans le plaidoyer en faveur de lajustice sociale. Il décrit de vastes catégories derésultats selon lesquelles les activistes, lesdonateurs et les évaluateurs peuvent jaugerde manière suivie les progrès (ou le manquede progrès) : les changements des capacitésd'organisation, la base de soutien, les alliances,les données et les analyses dans une perspectivede justice sociale, la définition des problèmes etles options politiques potentielles, la visibilité,les normes publiques et les impacts au niveaude la population. En utilisant ces critères pourl'évaluation, les activistes et les donateurspeuvent tirer des enseignements et revoir leursstratégies pour suivre l'évolution du contextepolitique et/ou des acteurs. L'article présenteune étude de cas comparant des facteurs quiont facilité des gains politiques pour les droitsgénésiques pendant la transition de l'apartheidà la démocratie en Afrique du Sud et desfacteurs qui ont miné leur application à la finde l'apartheid. Il avance qu'après les victoiresjuridiques et politiques, l'incapacité à conserverdes organizations fortes et à repenser enpermanence les stratégies a contribué à uneperte de la priorité gouvernementale et desressources accordées à l'application de politiquesnouvelles. Par conséquent, l'évaluation del'efficacité uniquement par un changementpolitique réel ne permet pas un apprentissageconstant pour assurer des stratégies adaptées.Elle ne tient pas non plus compte du risque deperte des gains politiques qui requiert d'êtrevigilant dans le plaidoyer et le suivi de la miseen œuvre. Il en découle que le financement etl'organisation du plaidoyer doivent rarementêtre des activités à court terme. L'article suggèreaussi que l'instauration et l'entretien de lacapacité organisationnelle et du leadership sontaussi importants que toute autre catégorie derésultat pour parvenir au succès.

ResumenEn este artículo se ofrece un marco de teoría decambio para abogar por la justicia social. Sedescriben categorías generales de resultadosmediante los cuales activistas, donantes yevaluadores pueden evaluar los avances (o faltade estos) de manera continua: cambios en lacapacidad organizacional, base de apoyo,alianzas, datos y análisis desde la perspectivade justicia social, definición del problema yposibles opciones de políticas, visibilidad,normas públicas e impactos en los niveles depoblación. Al utilizar estos resultados para laevaluación, los activistas y donantes puedenaprender de sus estrategias y reformularlassegún vayan cambiando los actores y/o elcontexto político. Se expone un estudio decaso en el cual se comparan los factores quefacilitaron las victorias de políticas a favor delos derechos reproductivos durante la transicióndel apartheid a la democracia en Sudáfrica y losfactores que debilitaron su implementación enel período post-apartheid. Se argumenta que,tras las victorias legislativas y políticas, el nomantener organizaciones sólidas y reformularlas estrategias continuamente contribuyó a lapérdida de enfoque del gobierno en la aplicaciónde nuevas políticas y en los recursos para éstas.Implícitamente, evaluar la eficacia exclusivamentemediante un cambio de política no permite elaprendizaje continuo para garantizar estrategiasadecuadas. Además, no se reconoce que tras logrardeterminada política, ésta puede ser revocada yque su implementación requiere monitoreoatento y esfuerzos de promoción y defensa(advocacy). Esto significa que el financiamientoy la coordinación de actividades de promoción ydefensa rara vez se deben realizar como unapropuesta de corto plazo. Indica también queel desarrollo y mantenimiento de la capacidadde organización y liderazgo es tan importantecomo cualquiera de las demás categorías deresultados para tener éxito.

B Klugman / Reproductive Health Matters 2011;19(38):146–162

162