Cooperating for a more egalitarian society: Complexity theory to evaluate gender equity

Abstract

Evaluating gender equity involves the assessment of the equality of opportunities and the equality of outcomes that public policies seek to attain for women and men. It focuses on how and to what extent both genders cooperate to expand access to paid work and control over material resources while sharing care and reproductive responsibilities. Drawing on complexity theory, this article puts forward a theoretical framework to identify cooperative behaviors within the household and the workplace as well as within broader socioeconomic, political and institutional domains.

Full text

Evaluation
2015, Vol. 21(1) 32 –46
© The Author(s) 2015
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/1356389014564856
evi.sagepub.com
Cooperating for a more
egalitarian society:
Complexity theory to
evaluate gender equity
Mita Marra
University of Salerno, Italy
Abstract
Evaluating gender equity involves the assessment of the equality of opportunities and the equality
of outcomes that public policies seek to attain for women and men. It focuses on how and to
what extent both genders cooperate to expand access to paid work and control over material
resources while sharing care and reproductive responsibilities. Drawing on complexity theory,
this article puts forward a theoretical framework to identify cooperative behaviors within the
household and the workplace as well as within broader socioeconomic, political and institutional
domains.
Keywords
complexity theory, feminist evaluation, gender cooperation, gender equality policies,
institutional feminism
Introduction
To provide effective solutions to the deeply embedded structural inequalities that characterize
the 21st century, we need to understand the ways in which race, class, gender, sexuality, nation
and other systems of inequality intersect. These systems are complex, pervasive, persistent,
mutually constituted, and socially constructed power relationships (Dill and Zambrana, 2009;
Hess-Biber, 2012). When social inequalities are viewed as power relationships, evaluation
needs to identify what and how dominant groups benefit from denying others access to mate-
rial and social resources, such as adequate child care, paid work opportunities, or political
representation. In these circumstances, evaluation is required to look in depth at all those
Corresponding author:
Mita Marra, Department of Political, Social and Media Science, University of Salerno, Via Giovanni Paolo II 132, 84084
Fisciano (SA), Italy.
Email: mimarra@unisa.it
564856EVI0010.1177/1356389014564856EvaluationMarra: Complexity theory to evaluate gender equity
research-article2014
Article
by Mita Marra on January 14, 2015evi.sagepub.comDownloaded from

Marra: Complexity theory to evaluate gender equity 33
measures that might alter the balance of power – i.e. a living wage; shifts in workplace control;
universal, affordable quality child care; accessible public transportation; safe, affordable
housing; equal access to education and universal prevention-focused health care; and greater
opportunities to reach higher level decision making within firms and political institutions.
Recently enacted policies for gender equality have sought to combine macro-level actions
with a localized and complex understanding of people’s lives – eschewing both an expert-
centered approach that fails to engage people and a grassroots approach that fails to engage
broader structural environment for macro-level policies (Naples, 2003). Feminist scholarship
and evaluation have influenced the design and implementation of these policies, while advanc-
ing the understanding of gender social inequality and its intersections. However, they have
done so largely outside the mainstream of public funding and outside legitimized avenues of
so-called ‘state feminism’ to shape social awareness and political debate for social justice
(Marra, 2014; Mazur, 2002; see also http://libarts.wsu.edu/pppa/rngs/index.html). The com-
plex nexus of systemic inequalities – within which gender power relationships are adaptive,
interdependent, embedded and emerging over time – have yet to be explicitly taken into
account and publicly discussed. This is essential if we are to avoid politically correct rhetoric
failing to influence individual and social behavior.
To incorporate an explicit analysis of race, class and gender as power relationships, evalu-
ation is asked to identify the linkages between the mix of conflict, power and cooperation that
pervade social relationships (Reynolds, 2014). Drawing on complexity theory, I call evalua-
tors’ attention specifically to gender cooperation in the pursuit of a more egalitarian society
(Gornick and Meyer, 2008; Seabright, 2012). I combine complexity theory (Byrne, 2013;
Forss et al., 2011; Rogers, 2008; Westhorp, 2012) with a realist evaluation approach (Pawson,
2006; Pawson and Tilly, 1997) to explore gender cooperative behaviors within the household
and the workplace as well as within broader socioeconomic, political and institutional domains.
Analyzing gender cooperation in different realms will help to characterize the multidimen-
sional notion of social equity, drawing attention to ethical criteria and standards that can be
shared within the evaluation community (Forss and Marra, 2014).
The aim of this article is twofold. From a theoretical perspective, I posit that complexity
theory allows us to explore some of the most daunting and elusive dimensions of gender
power imbalances or the unequal distribution of time between care and paid work across dif-
ferent domains and units of analysis. From a normative perspective, I endorse a view of evalu-
ation as contributing to social equity through the acknowledgement of those family,
organizational, institutional, and socioeconomic patterns; that may move our society towards
greater gender equality.
In shifting the focus towards gender cooperation, I contend that the issue is not developing
new methods for evaluation, but rather conceptually reconsidering how to assess gender
equity. The challenge is selecting a variety of methods among the many at our disposal – many
of which blur the distinctions between quantitative vs. qualitative research, theory-based ver-
sus counterfactual analyses. Yet, our selection of methods is complicated by the very power
dynamics and value conflicts that shape the broader terrain of knowledge production (Maynes
et al., 2008 in Weber and Castellow, 2012).
With this caveat in mind, in the first section, I discuss the reason for focusing on coopera-
tion, as a key dimension of gender equity. In section two, I identify those tacit and intentional
change mechanisms that can trigger cooperation between men and women across different
realms within our society. In section three, I lay out those complexity dimensions that may
by Mita Marra on January 14, 2015evi.sagepub.comDownloaded from

34 Evaluation 21(1)
help us examine cooperation mechanisms in the analysis of the micro interactions within the
household, as well as the meso and macro interdependencies unfolding within organizations,
institutions and the society as a whole. Finally, I present some concluding remarks discussing
ethical criteria and standards for the evaluation of gender equity.
Evaluating gender cooperation rather than state feminism
In Sen’s capabilities approach (1999), the goal of governments should be that of expanding the
real (that is, materially feasible) freedom to choose the kind of life one has reason to value
(Seguino, 2013). At the same time, Robyens (2003, 2007) identifies remaining questions:
 which capabilities matter for gender equity?
 what is the minimum set of capabilities that assure equal pay-offs, warranting men and
women the right to be different?
Nussbaum’s view on building capabilities conveys a deep conviction that new forms of
sociability can be engendered, in which individuals exhibit a strong commitment to the good
of others. The roots of this argument revolve around the Aristotelian notion that the good of
others is not just a constraint on a person’s pursuit of her own good; it is part of her good
(Comim and Nussbaum, 2014).
According to Hirschman (1958), the world image congenial to development is the so-
called group- and ego-focused idea of change – that is, the idea that ‘the individual can
advance at his own speed within an expanding economy’ without impeding the progress of
other people. This image of change ‘is likely to be adopted only after a considerable span of
experience has convincingly shown the possibility of such a development’ characterized by
all-around growth and mutual benefits (Hirschman, 1958: 23). Thus, cooperation between
men and women and the existence of relational ties of mutual support are constitutive fea-
tures of social structures and are reflected in mental representations of individuals, particu-
larly in their images of the world (Kahneman, 2003; North, 2005; Salvatori et al., 2009; see
later on).
Cooperation does not deny conflicting interests and behaviors; it requires explicit decisions
on how to distribute material resources and the attention to the social and psychological realm
that shapes people’s often tacitly held opportunity sets (Seguino, 2013). While recognizing
different gender identities and preferences, cooperation between men and women can create
capabilities and real freedoms, beyond a mere legalistic approach, accentuating procedural
freedoms – such as, for example, the right to vote or the right to property. Through coopera-
tion, power imbalances over economic and social resources and unequal sharing of care and
social reproduction can be addressed and redressed combining, for example, ample parental
time for children with gender-egalitarian divisions of labor (Gornick and Meyer, 2004, 2008).
Our attention shifts then from analyzing inequality as a static condition that constraints
women’s lives both in the private and public realm to unveil how men and women become full
partners in social interaction. Analysis moves to inquire how individuals, families and social
groups can improve their well-being, reconciling perceptions of efficiency and perceptions of
fairness. The idea is to grasp how an institutionalized pattern of socioeconomic relations and
cultural values – that prevent men and women from participating as peers – may increasingly
become more egalitarian and inclusive (Fraser, 2009).
by Mita Marra on January 14, 2015evi.sagepub.comDownloaded from

Marra: Complexity theory to evaluate gender equity 35
The methodology and epistemology of evaluating cooperation challenge the assumptions
of neutrality and objectivity that characterize rational-choice program design and traditional
impact evaluation. The theoretical paradigm of intersectionality (Davis, 2008; Dill and
Kohlman, 2012), emerging predominantly in social science and humanities, has fore-
grounded women who live at the intersections of social inequalities or privileges, due to
gender, race, class, sexualities, abilities, ethnicities, languages, and other systems of domi-
nation/subordination. Accordingly, ‘depending on the context, an individual may be an
oppressor, a member of an oppressed group or simultaneously oppressor and oppressed’
(Collins, 1990: 225).
The intersectional approach recognizes and values women’s multiple intelligences, diverse
ways of knowing, and frequently contradictory and sometimes silent and silenced voices, and
it presses to develop trust among researchers to initiate shared processes of action-reflection
(Dill and Kohlman, 2012). Intersectional research (Collins, 1990) has long contributed to this
cognitive and political agenda, seeking to validate the lives and stories of previously ignored
groups of people and to help empower communities and the people in them. The production
of knowledge to address real-life social issues and problems and the application and use of this
knowledge to solve problems of inequality have been fundamental to the intersectional project
of promoting social justice. For instance, the link between theory and social justice that enable
institutional transformation have been demonstrated quite effectively in critical race theorists’
studies (Crenshaw, 1992; Delgado, 2002), which include analyses of white privilege as struc-
tured throughout the legal system and other sociopolitical institutions, as well as the examina-
tion of civil rights law that operates through ‘race neutral’ principles to sustain white
dominance.
Linking theory, policy, and practice has been the principal focus also of critical legal theory
as epitomized in the work of Collins (1990) and Smith (1991), who discuss how power shapes
gender relations within and across racial, class, and sexual diversities. Conducting their study
at the interface of feminism and participatory and action research, these scholars drew upon
the critical feminist epistemologies wherein women’s work was problematized and women’s
agency tended toward activism and social change. They used a diverse range of research meth-
ods to facilitate distinct processes of knowledge construction, engagement with women, polit-
ical activity, and social change. Despite this research, the evidence on scaled-up change in
both advanced and developing countries is limited.
Within the policy evaluation field, the feminist approach has deeply examined the root
causes of social inequality to promote broader spaces of freedom for both women and men.
Feminist evaluation critically examines both formal and informal (stated and unstated) prac-
tices, policies, and activities embedded in a program context to explore and make explicit the
working of these inequities. The values and assumptions that drive programs – and the policies
that are implemented through them – are viewed as key factors influencing how girls and
women will be educated, receive healthcare, raise their families, experience the world, and
grow into old age (Sielbeck-Bowen et al., 2002). A feminist approach supports an active role
for the evaluator, who is committed to pursuing a more equitable social order (Podems, 2009).
The feminist evaluator acknowledges the differences in lifestyle choices of each individual
and seeks to overcome pervasive power asymmetries existing in organizational and political–
institutional contexts where he/she operates (Podems, 2009). For this reason, feminist evalua-
tion has been perceived as ideological or politicized (Patton, 2002; Podems, 2009) with little
influence on high-level decision making.
by Mita Marra on January 14, 2015evi.sagepub.comDownloaded from

36 Evaluation 21(1)
Although it is beyond the scope of this article to provide an assessment of both intersec-
tional and feminist research and evaluation approaches (Bustelo, 2003, 2011; Podems, 2009;
Sielbeck-Bowen et al., 2002; UNEG, 2011), it is noted here that the relevant scholarly and
evaluative work has neither drawn up comprehensive comparative analyses across different
political and institutional contexts, nor has it influenced, at the policy level, the design of
broad-based strategies capable of modifying traditional gendered power relations, roles and
stereotypes. While women’s policy agencies increasingly ally with activists to advance wom-
en’s interests in Western and Latin American democracies (Alvarez, 1999), there remains a
disconnect between state feminist theory and its practical translation within a political agenda,
which still lacks a clear and genuine vision of social justice.
Furthermore, within the evaluation community the adoption of gender-based and feminist
approaches has not been systematic and widespread, except for those impact assessments
which were explicitly required to examine gender issues. Mainstream evaluators may not
largely share a specific expertise in the field except within the networks of the so-called femo-
crats – those professionals, who specialize in gender studies usually working within interna-
tional organizations (Marra, 2014). There is a risk of developing ideological, rhetorical or
technocratic exercises with limited dissemination within policy circles. In these circumstances
the need for evaluation remains that of:
i) grasping and assessing the interplay of the social, economic, and political mechanisms
shaping the policy context at different decision making levels;
ii) developing an awareness of and dialogue about the strengths and limitations not only
of feminist and intersectional analyses, but also conventional positivist paradigms
addressing social inequalities; and
iii) last but not least, increasing the efficacy of efforts to strengthen research, evaluation
and public policy connections in ways that promote social justice.
If we recognize that power relations are a central dynamic in race, class, and gender struc-
tures, we need to demonstrate how power relations are co-constructed, maintained, and chal-
lenged. Explicating how these systems operate – the mechanisms that constitute and connect
them – becomes a critical challenge. Understanding, for instance, the persistence of gender-
based workplace discrimination in the presence of formal legal structures prohibiting it,
requires a shift of attention away from the motives for discrimination to its mechanisms. For
this purpose, as Weber and Castellow put it:
we should not ask why allocators (people in positions of power) in the workplace produce inequality
by making decisions that produce unequal outcomes for women (i.e. their motives) but instead ask
how their actions produce inequality – stated otherwise, what actions or processes produce the
inequalities (i.e. the mechanisms). (Weber and Castellow, 2012: 439)
Reversing the direction of analysis in this way allows us to ask what actions produce equi-
table outcomes, such as sharing care responsibilities within the household; unbiased assess-
ment and formalization of reward systems within the workplace; and access and capacity in
decision making within political institutions and private firms. Understanding the mechanisms
of cooperative relationships will enable us to identify different kinds of interventions, shifting
the balance of power away from the traditional paradigm with its focus on gendered roles and
power relationships (Weber and Castellow, 2012: 439).
by Mita Marra on January 14, 2015evi.sagepub.comDownloaded from

Marra: Complexity theory to evaluate gender equity 37
Exploring ‘what’ leads to cooperation rather than discrimination embraces a continuous
and iterative process or approach to research and evaluation. This work requires evaluators to
dig out those relationships that rely on and value strengths, capabilities, social capital, and
resiliency. It also asks us to be sufficiently challenging to engage in reflective critical practices
that problematize the matrices of power, privilege, and domination in social groups (Weber
and Castellow, 2012). Building on a realist evaluation approach, the next section examines the
social mechanisms of cooperation.
Unpacking multilevel mechanisms of gender cooperation
According to recent strands of cognitive and social psychology, the majority of social relation-
ships and interdependencies thrives both on systems of deliberate actions of individuals
including non-choices. In such systems, cooperation has a dual character: in some cases,
resulting from explicit and predetermined agreements; in others, cooperation emerges from
unexpected and unintentional behaviors. Thus some forms of cooperation are not the result of
planning or collective choice, but are ‘imagined’ schemes stemming from the minds and ideas
of individuals.
Through mental processes of imitation and identification (Kahneman, 2003; Salvatori et al.,
2009), an individual recognizes another as similar to himself/herself or having similar intentions.
This mirroring effect permits the creation of an interpersonal bond. The subject observes the
action of others, and then he/she maps it out as if they themselves were to do it. In this way,
shared meanings and practices are constructed that change the way in which individual decisions
are combined with socially interdependent choices – of production and social reproduction.
Thus, in creating and strengthening relationships of mutual support, whether intentional or not,
cooperation within households, firms, public and nonprofit organizations, and other political
spheres can influence the mindsets of individuals and exert an instrumental role. Such ideas have
the potential to transcend reality, support the emergence of new scenarios of equality and spread.
If credible and trusted these mental images will be seen as plausible and tacitly accepted, leading
to social change that can then trigger developmental results.
This possible chain of causes and effects is not just the latest finding of neurosciences. In
1977, anthropologist Cliffort Geertz and social scientist Albert Hirschman assumed and set
out to explore the variety of relationships linking social representation, social action and the
ways social problems are interpreted and addressed through public policies (Geertz and
Hirschman, 1977). Following this premise, it is possible to imagine that a culture built on a
worldview compatible with gender cooperation may emerge and evolve over time to suit the
needs of an environment of changing individual aspirations and new socioeconomic opportu-
nities (Abraham and Platteau, 2004). As gender-cooperative mental representations favor
gender-egalitarian preferences and action, so more egalitarian societies may reinforce and
diffuse gender-cooperative world views.
From an evaluation standpoint, these assumed links between social representation, social
action, and public policy deserve deeper analysis to understand the mechanisms and the condi-
tions that favor intended and unintended cooperative practices between men and women as
well as between social groups and institutions. Investigating the reactions of beneficiaries to
the opportunities offered by public policies requires the understanding of why and how these
opportunities relate to recipients’ expectations, aspirations, and perceptions of efficiency and
fairness via emotional, cognitive and economic mechanisms.
by Mita Marra on January 14, 2015evi.sagepub.comDownloaded from

38 Evaluation 21(1)
Evaluation can assess how various policy instruments – such as information, training, edu-
cation and moral suasion, as well as monetary incentives, services, infrastructure and rights
protection – can trigger intentional and unintentional emotional reactions, which can in turn
reinforce collaborative efforts (Marra, 2014).
While emotional mechanisms, such as motivation, imitation and empathy, act through men-
tal representations of cooperation and mutual support, cognitive and economic mechanisms
rely on information, knowledge, and rational calculus. Cognitive mechanisms can be condu-
cive to cooperation by strengthening, for instance, men’s awareness of women’s dignity as
human beings and the understanding of the needs for sharing child care and housework.
Economic mechanisms can be enacted through the (re)distribution of resources (including
income and time), which can modify beneficiaries’ costs and benefits associated with paid
work, family care and its outsourcing. In fact, recently introduced flexible working contractual
arrangements with a fast track for women can further equalize female and male paid positions
and earning capacity, hence changing the way women’s time and income are perceived. When
faced with the choice of who, within the couple, should leave a job to look after children, men
and women in ‘insecure’ positions may have to negotiate whose source of income to forego.
When both women’s and men’s earnings are unstable, the choice to forego women’s income
can no longer be taken for granted.
This suggests that the analytic framework has to be broadened to identify not only those
factors that play a role in decision-making but also to analyze how such factors are trans-
formed into subjective decisional criteria. Evaluation can then detect different mechanisms of
change which interact reciprocally and over time (Elster, 1998; Pawson, 2006; Pawson and
Tilly, 1997) across different contexts, reconstructing the interdependent choices of coopera-
tion between men and women as well as the multifaceted network of institutions that imple-
ment public policies.
Government agencies, nonprofit organizations, and businesses are not an undifferentiated
whole, subject to the same political and administrative constraints and likely to respond in
standardized and predictable ways. All actors and groups involved in the implementation of
public policies are exposed to various forms of cultural influence depending on their degree of
openness to the context in which they operate. They may respond to gender-sensitive policies
and needs in different ways depending upon social norms, and formal and informal institu-
tions, existing at a specific time.
When evaluating these processes, complexity grows enormously. For instance, it is daunt-
ing to assess support to families, in association with the planning and organization of urban
transportation and all other public services, as well as the programming of socioeconomic
investments addressing the needs of regions and localities. These factors derive from a variety
of interdependent socioeconomic and political systems, where it is crucial to examine the
availability of financial resources as well as physical and virtual infrastructures, the existing
production structures and commercial opportunities; and the openness of political processes
and participation.
By drawing attention to how such complex policy, socioeconomic and political systems
may adapt and innovate in dynamic and unpredictable situations, complexity science provides
a rich framework for evaluators to reconstruct different processes of change and coordination
occurring within rapidly evolving and uncertain environments. Complexity science urges
evaluators to identify and recognize connections across actors and structures emerging
between system participants. Evaluation can therefore utilize:
by Mita Marra on January 14, 2015evi.sagepub.comDownloaded from

Marra: Complexity theory to evaluate gender equity 39
 local knowledge and contribute to ongoing trial-and-error learning;
 shared meaning, in which use is enhanced through the distributed intelligence; the
networked capacities of all actors – what social networking refer to as social capital
and cooperation (Marra, 2011b). I next turn to discussing key dimensions of complex-
ity theory, which explain how cooperation mechanisms can emerge and consolidate
over time.
Cooperation within program and context complexities
In this section, I argue that cooperation (of lack of it) does not depend on a lack of oppression,
but rather on interlocking interactions between conflict, power, subordination, solidarity and
freedom. I develop this argument using five specific dimensions of complexity – namely,
adaptation, self-organization, interdependencies, emergence, and embeddedness. Interactions
may be the characteristics of the contexts in which programs are conceived and implemented,
or specific traits of the programs themselves. By exploring the interactions that can potentially
unfold between program features and context traits, we can then identify what cooperation
levers the program may activate (e.g. mechanisms) within the different contextual conditions
where outcomes (intended and unintended) are observed.
Adaptation refers to diverse agents who can learn, self-organize, and co-evolve with their
environment in non-linear ways. Order and progress can emerge naturally from the interac-
tions within complex adaptive systems; they do not need to be imposed centrally or from
outside. For instance, the construction of professional and service-user identities – e.g. ‘social
workers’ or ‘welfare recipients’ (who are usually women) in social programs – are subject to
system-wide processes where cooperative behaviors are reinforced, or attitudes are adapted.
Individuals are part of wider systems of professional training and validation, where collabora-
tive work organization may be the result of a significant capacity to sustain mutually benefi-
cial work routines, adapting to existing environmental characteristics and constraints.
The complexity lens of adaptation suggests that any intervention will always be locked into
different social, economic, political and institutional systems and we need to understand how
these systems facilitate or undercut different chains of causation. For instance, focusing on
how work–family relationships change over time to adapt to newly emerging needs may yield
valuable insights into the nuanced condition of social vulnerability affecting programs that
seek to mobilize beneficiaries and enhance their potential for personal empowerment and
professional growth (Albelda, 2011). Evaluating, for instance, how disadvantaged women
adapt to work assignments may lead to a more robust program theory of change, that charac-
terize gender equity in terms of cooperation mechanisms both within the household and the
workplace.
Different evaluation approaches can grasp cooperation as an adaptive pattern. Various
approaches lend themselves to reconstructing how interpersonal relationships and organiza-
tional structures adapt to environmental features and constraints through mutually supportive
interactions. These include:
 ‘utilization-focused evaluation’ (Patton, 2008);
 ‘fourth generation evaluation’ (Guba and Lincoln, 1994);
 ‘empowerment evaluation’ (Fetterman et al., 1996);
 ‘goal free evaluation’ (Scriven, 1995);
by Mita Marra on January 14, 2015evi.sagepub.comDownloaded from

40 Evaluation 21(1)
 participatory, transformative or culturally inclusive assessment (Mertens, 2008);
 social experiments (Carpenter, 2005; Duflo et al., 2013);
 organizational studies; and
 ethnographic accounts (Brisolara, 1998).
The dimension of self-organization helps us understand and constructively deal with sys-
tem variations in local contexts. Self-organization arises from local contextual circumstances
(Stacey, 2005). Patterns of how people interrelate are influenced by how they find it most
effective to complete tasks, given locally available resources and contexts. Patterns of self-
organization are powerful because they are rooted in what is required to accomplish tasks on
the ground (Lanham et al., 2013). We therefore need to acknowledge the limits of imposed
structures such as implementation designs, project plans, and formal organizational hierar-
chies. Even in the face of formal rules, procedures, and structures designed to control, self-
organization will continue based on needs that exist but that might not be recognized at higher
levels of a system.
The dimension of self-organization is particularly relevant in the observation of organiza-
tional processes and within households. Here leadership, team work and division of labor may
unfold through cooperative or conflicting patterns so that any attempt to change such patterns
may just fail because of self-organizing pressures. For instance, ingrained perceptions of gen-
dered roles highlight the flawed assumption that female and male roles are interchangeable
and that public policy can employ measures regardless of the different meanings men and
women attach to them. Thus, parental leave is assumed to be equally enjoyed by men and
women or outsourcing care is presumed to be a commonly accepted practice (Mansdotter
et al., 2007). By contrast, in those contexts where traditional culture persists, few men take
parental leave and women’s freedom to shed caring work onto non-family members lacks
legitimacy. For example women may feel guilty about time spent outside the household unless
grandparents help is assured. Domestic organization therefore rests, with women’s own capac-
ity to balance work and family which they may not perceive as a burden.
In such circumstances, ‘participatory’ methods of data collection and action-reflection pro-
cesses can have a transformative power, by paving the way for women to have a voice – both
as beneficiaries or stakeholders. This can be reinforced by the involvement of the evaluator,
who can raise awareness of actors’ subjective assumptions and broaden the space for partici-
pation in management and decision making within organizations and the household (Weber
and Castellow, 2012). Evaluation can then be an inclusive and evolutionary process, uncover-
ing latent or tacitly changing caring needs, challenging traditional power relations and stereo-
typical perceptions, opening up opportunities for empowerment and agency.
Interdependencies are the structures and processes through which people interact, exchange
information and interpret observations. Assessing interdependencies suggests focusing on
social norms and institutions that can affect actors operating in delimited contexts as well as
social groups and organizations in broader environments. Gender inequality – Seguino (2013)
posits – works as a ‘conversion factor,’ discounting the extent to which women can convert
income and other economic resources into capabilities and power. This is undergirded by an
ideology that justifies the unequal state of gender relations, socially and materially, which are
in turn embedded in a variety of institutions, including marriage laws, property laws, labor
markets, cultural and religious institutions (Seguino, 2013: 4). The greater the degree of gen-
der inequality at the macro level (i.e. women’s greater concentration in low-wage insecure
by Mita Marra on January 14, 2015evi.sagepub.comDownloaded from

Marra: Complexity theory to evaluate gender equity 41
jobs or lack of jobs compared to men), the less bargaining power all women have to differing
degrees within the household, (Seguino, 2013: 13).
Interdependencies show that macroeconomics influences the extent to which cooperation
can unfold within the household by affecting women’s bargaining power and outside options.
Different evaluation approaches can effectively uncover how power, conflict, and cooperation
is reinforced or weakened through political, institutional and organizational processes. These
approaches range from theory-based and realist evaluations to structural models and multi-
variate statistical methods, from social network analysis to time use studies.
Emergence refers to the ideas that existing components in a system will often combine to
produce new components, thus, continually changing the composition of a system. In social
science, this phenomenon is rooted in the way we (as a group or a society) interpret situations.
Collective interpretations can have causal power and become a strong driver of change. Recent
studies underscore that people adjust to poverty and deprivation by learning to suppress their
wants, hopes and aspirations (Nussbaum and Sen, 1993). Human goals – in the form of aspira-
tions for future education and employment – are conditioned by pre-existing inequality.
Accordingly, the evaluation of disadvantage or inequality based on the attainment of goals
will run into difficulties since the aspirations expressed by those disadvantaged are constrained
by their previous experience. The more disadvantaged have more modest goals, and are more
likely to attain them. Thus, a ranking of advantage and disadvantage based on goals – for
example, employment aspirations – produces the counter-intuitive result that girls at the top of
the scale are those who were bullied at school, received little support from their parents and
were poorly educated, while those at the bottom of the scale are disproportionately from a high
social class, well-supported and well-educated background (Burchardt, 2009).
Sometimes what emerges is continuity: we choose to stick with institutions and systems
that we collectively favor. In North America, for instance, affirmative action endorses a view
of social cooperation predominantly relying on the promotion of socially disadvantaged
groups, and specifically women, into traditional male occupations. Quotas within boards of
directors of listed companies, for example, foster female professional advancement across dif-
ferent class, race and ethnic groups towards higher social positions (Alber, 2010; Orloff, 2002;
Shalev, 2008). By contrast, within continental Europe, social cooperation has mostly relied on
monetary transfers and welfare state assistance to ‘de-commodify’, in Esping-Andersen’s
words (1990), caring and family responsibilities. Differences remain between the welfare
regimes and policy approaches to gender equality in the North-American and European social
protection systems (Alber, 2010; Orloff, 2002; Shalev, 2008).
Emergence may also lead to change and produce new institutions and systems. Throughout
Western democracies, current gender-sensitive policies build upon the new paradigm of ‘social
investment’ to balance work and family (Knijn and Smit, 2009). Recently enacted policies for
gender mainstreaming increasingly integrate different instruments and approaches to create
conditions for sharing family and professional responsibilities.
At other times, emergence generates unintentional and unwanted changes to institutions
and systems. For example, early intervention programs target families with multiple demo-
graphic risks factors (i.e. poverty, single parents, young parents, etc.). This targeting may,
indeed, be associated with negative social judgments and an increased sense of threat from
social services, which in turn may lead to withdrawal from social and community net-
works, increased isolation and decreased social capital (unintended consequences)
(Westhorp, 2013).
by Mita Marra on January 14, 2015evi.sagepub.comDownloaded from

42 Evaluation 21(1)
The phenomenon of emergence tells us that we have imperfect control over expected
or desired outcomes. The reason for this is that programs not only work to change behav-
ior but they may also change the conditions that make the program work in the first place.
For instance, family mediation interventions may trigger cooperative interactions within
the couple, which can have a transformative power not only because these interactions
overcome the traditional division of care and domestic work, but also because these con-
tribute to a new understanding of parenting, affection and responsibility towards minors
and dependents.
Embeddedness stresses the role of personal relations and ‘networks’) of such relations in
generating trust and discouraging bad behavior. The key premise is that individuals choose to
cooperate in anticipation of the likely choices of others, and this means that their concrete
social relations are critical to their actions. This premise leads to questions such as: How fre-
quently do actors interact? Has there been a history of successful cooperation among these
actors in the past? Are there rivalries among actors that might work to reduce trust?
From this perspective, gender stratification is a crucial issue to be assessed within a frame-
work that links the micro- and macro-levels of analysis. As previously emphasized, the more
power women have at macro-levels of social organization (in the workplace, in the larger
economy, and in political spheres), the greater their ability to control a proportionate share of
their output at the household level (Seguino, 2013). Yet, circumstances where interactions
occur are all situational, dependent on the history of locations and social relations. These
contextual conditions imply that very similar individuals, confronting very similar circum-
stances of choice, may arrive at very different patterns of social action depending on their
histories of interaction with each other. So, for instance, we cannot assume that because of
provision of maternal leave benefits, all mothers will go back to work. Nor can we assume
that due to care services opportunities, low skilled women will automatically be employed in
social work. Depending on their specific background, these individuals and groups may be
gradually induced to modify their expectations and make an effort to pursue change within
their own family as well as within the workplace. For highly educated women, more person-
alized assistance coupled with commercialized solutions and flexible work arrangements
could ease covert familial conflicts in favor of fairer cooperation. Still, such power shifting
processes may vary greatly and generate very different solutions to rearrange work–family
priorities.
To recap: the dimensions discussed thus far offer theoretical insights and methodological
solutions to grasp and assess key gender equity factors related to both program features and
context-related traits. Identifying these complex dimensions in different domains in relation to
different public policies will help us understand how interventions interact with contextual
conditions; and where social mechanisms of gender cooperation may be activated and rein-
forced over time.
Raising ethical standards?
The multidimensionality of social equity suggests shifting program design and policy evalua-
tion towards a complex notion of intervention that takes account of individuals, their ideas,
institutions and their ingrained culture and stereotypes, as well as physical and virtual infra-
structures that facilitate or discourage change. From this perspective, power relations, percep-
tions of gendered roles and the ways in which men and women cooperate in the choice of work
by Mita Marra on January 14, 2015evi.sagepub.comDownloaded from

Marra: Complexity theory to evaluate gender equity 43
and care can be adaptive and self-organizing patterns, embedded within social relations and
networks. Cooperative behaviors emerging within the households and workplaces are micro-
dimensions of social equity that if explored may open up wider freedom and opportunity-
space for men and women. From the other end of the spectrum socioeconomic and human
development of a region or country are meso and macro dimensions of social equity that may
hinder or accelerate the realization of individual aspirations for professional development and
family well-being as well as democratic social, political and institutional processes over time.
These factors interact in complex ways and affect all actors, with perceptions, aspirations,
cultures and values. It is crucial, therefore, to investigate the emotional, cognitive, and eco-
nomic mechanisms that generate, sustain, and propagate social cooperation – and specifically
cooperation between men and women – over time and through different spaces.
Evaluators who engage with this work often do so out of strong commitments to diversity,
multiculturalism, and human rights, combined with a desire to create a more equitable society
that recognizes, validates, and values difference (Dill and Kohlman, 2012). The social justice
agenda of these evaluations is crucial if it is to lead to theory and practice able to analyze
inequalities of power and privilege. Consequently, the complexity paradigm is of interest to
evaluators both inside and outside academia who share the values that underpin this research
framework.
This cognitive and political agenda raises the question of whether gender sensitive ethical
principles need to be explicitly shared within the evaluation community. Ethical discussions
usually remain detached from a discussion of the evaluation process; and some evaluators
consider this aspect of evaluation an afterthought. However one could argue that such princi-
ples should be added to existing ethical frameworks such as those of the American Evaluation
Association (AEA, 1995).
The ethical standpoint of the evaluator is a critically important aspect of ensuring that the
evaluation process and an evaluator’s findings are trustworthy and valid. Recently the United
National Evaluation Group’s (UNEG) Handbook ‘Integrating Human Rights and Gender
Equality in Evaluation – Towards UNEG Guidance’ provides step by step guidance on how to
integrate these dimensions throughout an evaluation process, helping, in this way not only
evaluators, but also evaluation managers, and program managers both within and outside the
UN system. Along with ‘inclusion’ and ‘participation,’ the Handbook explicitly articulates the
key principle of ‘fair power relations’ as follows:
. . . when evaluators assess the degree to which power relations changed as a result of an intervention,
they must have a full understanding of the context, and conduct the evaluation in a way that supports
the empowerment of disadvantaged groups, e.g. women’s empowerment where women are the
disadvantaged gender within a given context. In addition, evaluators should be aware of their own
position of power, which can influence the responses to queries through their interactions with
stakeholders. (UNEG, 2011)
Although such gender-sensitive guidelines may raise awareness and increase compliance
over time, I am convinced that what really makes a difference is ongoing debate within the
evaluation community and society as a whole, a debate that will be strengthened if key dimen-
sions of complexity thinking are incorporated into evaluators’ practice of gender equality and
cooperation. In this way evaluation can also contribute to overcoming ingrained preconceptions,
social norms and stereotypes that stand in the way of greater gender cooperation in households,
communities, workplaces and more widely across societal institutions.
by Mita Marra on January 14, 2015evi.sagepub.comDownloaded from

44 Evaluation 21(1)
Acknowledgement
This article builds on and further develops arguments first discussed in Marra (2014).
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for
profit sectors.
References
Abraham A and Platteau JP (2004) Participatory development: where culture creeps in. In: Vijayendra
Rao V and Walton M (eds), Culture and Public Action. Stanford, CA: University Press Stanford.
Albelda R (2011) Time binds: US anti-poverty policies, poverty and single mother’s wellbeing. Feminist
Economics 17(4): 189–214.
Alber J (2010) What the European and American welfare states have in common and where they differ.
Journal of European Social Policy 20(2): 102–24.
Alvarez S (1999) Advocating feminism: the Latin American feminist NGS boom. International Feminist
Journal of Politics 1(2): 181–209.
Brisolara S (1998) The history of participatory evaluation and current debates in the field. New
Directions for Evaluation 80: 25–41.
Brisolara S and Seigart D (2012) Feminist evaluation research. In: Hesse-Bibier SN (ed.), The Handbook
of Feminist Research. Theory and Praxis. Thousand Oaks, CA: SAGE.
Burchardt T (2009) Agency goals, adaptation and capability sets. Journal of Human Development and
Capabilities 10(1): 3–20.
Bustelo M (2003) Evaluation of gender mainstreaming: ideas from a meta-evaluation study. Evaluation
9(4): 383–403.
Bustelo M (2011) Last but not least: gender-sensitive evaluation as a forgotten piece of the policymak-
ing process. Unpublished manuscript.
Byrne D (2013) Evaluating complex social interventions in a complex world. Evaluation 19(3):
217–28.
Carpenter J, Harrison G and List JA (eds) (2005) Field Experiments in Economics. Stamford, CT: JAI
Press.
Collins PH (1990) Black Feminist Thought Knowledge, Consciousness, and the Politics of Empowerment.
Boston, MA: Unwin Hyman.
Comim F and Nussbaum M (eds) (2014) Capabilities, Gender, Equality: Towards Fundamental
Entitlements. Cambridge: Cambridge University Press.
Crenshaw K (1992) Whose story is it, anyway? Feminist and antiracist appropriations of Anita Hill. In:
Morrison T (ed.), Race-ing Justice, Engendering Power: Essays on Anita Hill, Clarence Thomas,
and the Construction of Social Reality. New York: Pantheon Books.
Crenshaw K, Gotanda N, Peller G and Thomas K (eds) (1995) Critical Race Theory: The Key Writings
that Formed the Movement. New York: The New Press.
Davis K (2008) Intersectionality as buzzword: a sociology of science perspective on what makes a femi-
nist theory successful. Feminist Theory 9(1): 67–85.
Delgado D (2002) Critical race theory, latino critical theory, and critical raced-gendered epistemologies:
recognizing students of color as holders and creators of knowledge. Qualitative Inquiry 8(1): 105–26.
Dill B and Zambrana RE (2009) Emerging Intersections: Race, Class, and Gender in Theory, Policy,
and Practice. New Brunswick, NJ: Rutgers University Press.
Dill B and Kohlman MH (2012) Intersectionality: a transformative pradigm in feminist theory and
social justice. In: Hesse-Biber SN (ed.), The Handbook of Feminist Research. Theory and Praxis.
Thousand Oaks, CA: SAGE.
Duflo E, Banerjee A and Kinnan C (2013) The miracle of microfinance: evidence from a randomized
evaluation. Working Paper, MIT, MA.
by Mita Marra on January 14, 2015evi.sagepub.comDownloaded from

Marra: Complexity theory to evaluate gender equity 45
Elster J (1998) A plea for mechanisms. In: Hedström P and Swedberg R (eds), Social Mechanisms: An
Analytical Approach to Social Theory. New York: Cambridge University Press, 45–73.
Fetterman D, Kaftarian SJ and Wandersman A (1996) Empowerment Evaluation: Knowledge and Tools
for Self-Assessment and Accountability. London: SAGE.
Forss K and Marra M (eds) (2014) Speaking Justice to Power. Ethical and Methodological Challenges
for Evaluators. New Brunswick, NJ: Transaction Publishers.
Forss K, Marra M and Schwartz R (eds) (2011) Evaluating the Complex. Beyond Attribution and
Contribution. New Brunswick, NJ: Transaction Publishers.
Fraser N (2009) From redistribution to recognition? Dilemmas of justice in a ‘post-socialist’ age. In:
Phillips A (ed.), Feminism and Politics. Oxford: Oxford University Press.
Geertz C and Hirschman AO (1977) Our view of social science. Princeton Seeley G. Mudd Library,
box 4, folder 12.
Gornick JC and Meyers MK (2004) Welfare regimes in relation to paid work and care. In: Zollinger
Giele J and Holst E (eds), Changing Life Patterns in Western Industrial Societies. Netherlands:
Elsevier Science Press, 45–67.
Gornick JC and Meyers MK (2008) Creating gender egalitarian societies: an agenda for reform. Politics
and Society 36(3): 313–49.
Guba EG and Lincoln YS (1994) Fourth Generation Evaluation. London: SAGE.
Hirschman AO (1958) The Strategy of Economic Development. New Haven, CT: Yale University Press.
Kahneman D (2003) A perspective on judgment and choice: mapping bounded rationality. American
Psychologist 58, 697–720.
Knijn T and Smit A (2009) Investing, facilitating, or individualizing the reconciliation of work and fam-
ily life: three paradigms and ambivalent policies. Social Politics 16(4): 484–518.
Lanham HJ, Leykum LK, Taylor BS, et al. (2013) How complexity science can inform scale-up and
spread in health care: understanding the role of self-organization in variation across local contexts.
Social Science & Medicine 93: 194–202.
Mansdotter A, Lindblom L and Winkvist A (2007) Parental leave in Sweden: cost savings and health
gains. Health Policy 82: 102–15.
Marra M (2011a) Micro, meso, and macro dimensions of change: a new agenda for the evalutation of
structural policies. In: Forss K, Marra M and Schwartz R (eds), Evaluating the Complex. Beyond
Attribution and Contribution. New Brunswick, NJ: Transaction Publishers.
Marra M (2011b) Some insights from complexity science for the evaluation of complex policies.
In: Forss K, Marra M and Schwartz R (eds), Evaluating the Complex. Beyond Attribution and
Contribution. New Brunswick, NJ: Transaction Publishers.
Marra M (2012) The missing links of the European gender mainstreaming approach. assessing work-family
reconciliation policies in the Italian Mezzogiorno. European Journal of Women’s Studies 19(3): 349–70.
Marra M (2014) Beyond feminist and gender-based approaches: evaluating cooperation between men
and women for gender equity. In: Forss K and Marra M (eds), Speaking Justice to Power. Ethical
and Methodological Challenges for Evaluators. New Brunswick, NJ: Transaction Publishers.
Maynes MJ, Pierce JL and Laslett B (2008) Telling Stories: The Use of Personal Narratives in the
Social Sciences and History. Ithaca, NY: Cornell University Press.
Mazur A (2002) Theorizing Feminist Policy. Oxford: Oxford University Press.
Mertens D (2007) Transformative paradigm: mixed methods and social justice. Journal of Mixed
Methods Research 1(3): 212–25.
Naples N (2003) Feminism and Method: Ethnography, Discourse Analysis, and Activist Research. New
York: Taylor & Francis.
Nussbaum MC (2000) Women and Human Development in the Capability Approach. Cambridge:
Cambridge University Press.
Orloff AS (2002) Women’s employment and welfare regimes globalization, export orientation and social
policy in Europe and North America. Social Policy and Development Program Paper Number 12,
June 2002, United Nations Research Institute for Social Development.
Patton MQ (2008) Utilization-focused Evaluation, 4th edn. Thousand Oaks, CA: SAGE.
by Mita Marra on January 14, 2015evi.sagepub.comDownloaded from

46 Evaluation 21(1)
Patton MQ (2010) Developmental Evaluation: Applying Complexity Concepts to Enhance Innovation
and Use. New York: Guilford Press.
Pawson R and Tilly N (1997) Realist Evaluation. London: SAGE.
Pawson R (2006) Evidence-based Policy: A Realist Perspective. London: SAGE.
Podems DR (2010) Feminist evaluation and gender approaches: there’s a difference? Journal of
Multidisciplinary Evaluation 6(14): 1–17.
Reynolds M (2014) Equity-focused developmental evaluation using critical systems thinking. Evaluation
20(1): 75–95.
Robeyns I (2003) Sen’s capability approach and gender inequality: selecting relevant capabilities.
Feminist Economics 9(2/3): 61–92.
Robeyns I (2007) When will society be gender just? In: Brown J (ed.), The Future of Gender. Cambridge:
Cambridge University Press.
Rogers P (2008) Using program theory to evaluate complicated and complex aspects of interventions.
Evaluation 14(1): 29–48.
Salvatore S, Forges Davanzati G, Potì S and Ruggeri R (2009) Mainstream economics and sense-mak-
ing. Integrative Psychological & Behavioral Science 43(2): 158–77.
Scriven M (1995) The logic of evaluation and evaluation practice. In: Fournier DM (eds), Reasoning in
Evaluation: Inferential Links and Leaps, New Developments in Evaluation, n. 68. San Francisco,
CA: Jossey Bass.
Seabright P (2012) The War of the Sexes: How Conflict and Cooperation Have Shaped Men and Women
from Prehistory to the Present. Princeton, NJ: Princeton University Press.
Seguino S (2013) Toward gender justice: confronting stratification and unequal power. Generos,
Multidisciplinary Journal of Gender Studies 2(1): 1–36.
Sen A (1999) Development as Freedom. New York: Albert A Knopf.
Shalev M (2008) Class divisions among women. Politics and Society 36(3): 421–44.
Sielbeck-Bowen KA, Brisolara S, Seigart D, et al. (2002) Exploring feminist evaluation: the ground
from which we rise. New Directions for Evaluation 96: 3–8.
Smith DE (1991) The Conceptual Practices of Power: A Feminist Sociology of Knowledge. Boston,
MA: Northeastern University Press.
Stacey RD (ed.) (2005) Experiencing Emergence in Organizations: Local Interactions and the
Emergence of Global Patterns. London: Routledge.
United Nations Evaluation Group (UNEG) (2011) Integrating Human Rights and Gender Equality in
Evaluation. Towards UNEG Guidance. New York: UNEG Human Rights and Gender Equality
Task Force.
Verloo M (2005) Mainstreaming gender equality in Europe: a critical frame analysis approach. The
Greek Review of Social Research 117(B): 11–34.
Weber L and Castellow J (2012) Feminist research and activism to promote health equity. In: Hesse-
Biber SN (ed.), The Handbook of Feminist Research. Theory and Praxis. Thousand Oaks, CA:
SAGE.
Weiss C (1997) Theory-based evaluation: past, present, and future. Progress and Future Directions in
Evaluation: Perspectives on Theory, Practice, and Methods 76: 41–55.
Westhorp G (2012) Using complexity-consistent theory for evaluating complex systems. Evaluation
18(4): 405e20.
Westhorp G (2013) Developing complexity-consistent theory in a realist investigation. Evaluation
19(4): 364–82.
Mita Marra is tenured Assistant Professor of Public Policy at the University of Salerno, Italy. She
received a MA in International Relations in 1998 from the Johns Hopkins University and a PhD in
Public Policy in 2003 from the George Washington University. She is the president of the Italian
Evaluation Association.
by Mita Marra on January 14, 2015evi.sagepub.comDownloaded from