Social norms

In recent years the development community has witnessed an upsurge of interest in the role that social norms may play in perpetuating a host of harmful practices, especially practices affecting women and girls.  There has long been interest in how deeply held beliefs, attitudes, and norms can justify male dominance and reinforce behaviour and institutions that discriminate again women.  Despite this recognition, there has been little conceptual clarity about the distinctions between these various constructs or how they relate to actual practices such as female genital cutting, domestic violence, or early marriage.

Indeed, activists, program planners, and donors have tended to refer loosely to the need to address “gender norms,” without making distinctions between whether the construct of interest is an attitude, a norm, a belief or behaviour. This undisciplined approach misses an important opportunity to use theory-based distinctions to shape programme planning and evaluation.  As discussed below, both the content of an intervention and who should be involved is greatly affected by the nature of the construct – whether it is an individually held belief or attitude or whether it is a social norm.  This reading pack is designed to help practitioners begin to understand these distinctions and thereby capitalise on evolving theory on how to transform and measure shifts in social norms.

Defining terms

The notion of “norms” is multi-disciplinary. A wide range of academic disciplines, including sociology, behavioural economics, philosophy, and social psychology use slightly different terms, however they largely converge around certain key insights related to norms:

  • A norm is a social construct. It exists as a collectively shared belief about what others do (what is typical) and what is expected of what others do within the group (what is appropriate). Social norms are generally maintained by social approval and/or disapproval. There are several more formal and complicated definitions of norms, but for our purposes, the above is sufficient.
  • An attitude is an individual construct. It is an individually held belief that has an evaluative component—suggesting that something is good, bad, exciting, boring, sacrilegious, disgusting, etc.  “Children should be seen and not heard” is an example of an attitude.
  • Individuals may also hold factual beliefs about reality and the physical world that may or may not be true. For example, I may believe that if I don’t bury the placenta of my newborn, bad spirits will sicken my child.  Sometimes all that is required to change behaviour is to correct factually inaccurate information. However, if all of my peers believe that you must bury the placenta or you are not a good mother, then I may continue the practice in order to receive their approval, even if I no longer believe in its value. Those whose opinions are important to me are called my “reference group.”

To shift social norms, interventions must create new beliefs within an individual’s reference group so that the collective expectations of the people important to them allow new behaviours to emerge. Evidence suggests that when norms are at play, shifting knowledge or individual attitudes is often not enough to shift behaviour.  That is because norms are generally enforced through either positive or negative sanctions.  People conform to group expectations out of the human need for social approval and belonging.  If individuals depart from a norm, they frequently loose social approval and may be ostracised, gossiped about, or sanctioned in some other way.

The table below offers one simple way to conceptualise different types of beliefs.  The most important distinction is between social and non-social beliefs: social beliefs are known as “interdependent” and are a property of a group, non-social beliefs are the property of individuals.

Social norms 1

Moral beliefs motivate behaviour regardless of what others may think.  For example, if I believe that the death penalty is morally wrong, I may protest the practice even if most of my family and other close friends support it.  Sometimes moral beliefs can acquire a normative dimension when people conform to a religious practice not because of their own internal belief, but because they fear that they may lose standing in their faith community if they did not conform to group expectations.

Problem diagnosis

A key challenge in transforming harmful behaviours is to diagnose the matrix of linked factors that keep the practice in place.  Most practices are held in place by a mix of structural, social, and individual level factors.  When social norms are operative, they can act as a “brake” on social change, which is why changes in knowledge and attitudes do not always result in changes in behaviour.   Sometimes behaviours are held in place by a matrix of interlocking norms.  For example, even in settings where wife beating itself is not a norm (in that people are not sanctioned if they don’t beat their wives), other norms – family privacy, male authority in the family and female obedience – help keep the practice in place.

In other instances, practices are driven almost entirely by structural or material factors rather than norms.  Here, focusing on social norms would largely be ineffective. For example, if poor parents marry their children young because they want one less mouth to feed, focusing on early marriage as a social norm may be misguided.  If, however, a main driver of early marriage is fear that unmarried girls may become sexually active and bring shame upon the family, then focusing on the collective expectations of virginity and family honour would be essential.

Agency, the ability to make decisions about one’s own life and act on those outcomes, is another factor that can be critical at an individual level. For example, many empowerment programmes focus on building women and girls’ agency, their capacity to aspire and their self-efficacy as a means to strengthen their ability to resist oppressive social structures. However, agency can be constrained by fear of retribution or violence, structured inequalities, or lack of skills and self-confidence to translate desires into action.  It can also be inhibited by a limited vision of what is possible.  Many women and girls have internalised social norms and messages to such an extent that they can’t imagine the world being any other way.

Thus, most social change strategies must target factors operating at multiple levels. Several of the readings in the package provide further guidance on how to assess these different factors and the degree to which they may interact to perpetuate harmful practices as illustrated in Figure 1.

Social norms 2

A note on terminology

The accompanying articles use a variety of different terms to refer to largely similar concepts. While almost all fields agree that a social norm has to do with beliefs about others, within some reference group and maintained by social approval and disapproval, most theorists make distinctions between beliefs about the prevalence of a practice within a reference group and shared expectations within that group about how one should behave.  A simple way to think about this is to consider what is “typical” compared to what is considered “appropriate”.  The table below summarises terms that are roughly equivalent in meaning, with each colour indicating the pair of terms used by different disciplines:

Terminology What it means
Descriptive norm ≈
Collective behavioural norm ≈
Empirical expectation
What is typical in one’s reference network
Injunctive norm ≈
Collective attitudinal norm ≈
Normative expectation
What is appropriate in one’s reference network

Social norm theory distinguishes between what is typical (beliefs about what others do) and what is appropriate (beliefs about what others think one should do) because both can be important for shaping behaviour.  For example, beliefs about what others do, can influence perceptions of what is appropriate or expected.  For example, research has shown that students on US college campuses believe that binge drinking is actually far more common than it really is.   This false “empirical expectation” creates subtle social pressure to conform to what is perceived as “normal” behaviour.  Some interventions have successfully reduced harmful drinking by publicising (making visible) the high proportion of students who do not engage in binge drinking. 

Likewise, while shifts in gender-related behaviour do not always lead to shifts in gender-related norms, seeing more people act in a new way can challenge people’s sense of what men and women usually do; as it becomes apparent that a large number of people are now acting in a new way (for example, women entering the workforce), changes in gender ideologies and the emergence of new normative expectations of how society should be organised emerge.

Insights from research and evidence-based practice

As you read the recommended literature, keep in mind the following insights that have emerged from research and evidence-based practice:

  • Building a new norm can often be easier and more strategic than attempting to dismantle a harmful one. For example, rather than directly challenging the norm that binge drinking is a way to prove your mettle, public health workers in the United States cultivated a new norm: ‘Friends don’t let friends drive drunk’.
  • Clarifying values and consensus-building through a deliberative process appears critical to the success of many norms-based approaches. Most successful interventions include processes to promote critical reflection, drawing on experiential learning and popular education techniques.
  • Breaking one norm can make it easier to shift associated norms. For example, breaking down the norm that men alone have responsibility to provide economically for the family, can smooth the way for shifts in other gender-related role expectations.
  • Considering opportunities to build strategies around “meta norms” that drive multiple behaviours. By challenging the norm that violence is an appropriate form of discipline, programmes can simultaneously attack a key rationale that undergirds corporal punishment in schools, harsh punishment at home, and wife beating in some settings.
  • Identifying those individuals or reference groups whose opinions matter most can help effect change. Defining the reference group, defines who must be targeted by the intervention.  If girls have traditionally not played sports because their families fear their reputation will be sullied, seeking to empower girls through sports will likely fail unless the project engages with family and community members as well as girls.
  • Communicating change (through testimony, organised diffusion and pledges) as it begins to norms and behaviours begin to shift in an initial core group is important.

Key readings

Reading 1: Marcus, R. & Harper, C. (2014). Gender justice and social norms – processes of change for adolescent girls. London, UK: ODI.

This paper is a useful introduction to how strategies for change may be most effective when they promote norm change in multiple spheres. It provides a framework that conceptualises the forces that maintain discriminatory gender norms against adolescent girls at the individual, community and structural level. The framework integrates how norms are experienced, the broader structural forces that sustain these norms and the social psychological processes by which gender norms change in order to understand the processes of change.

Reading 2:  Alexander-Scott M; Bell E.; Holden J. (2015). DFID Guidance Note: Shifting norms to tackle violence against women and girls (VAWG). London, UK: VAWG Helpdesk.

This guidance note introduces the major concepts of social norm theory and applies them to the specific challenge of preventing violence against women and girls.  It demonstrates how key concepts can be incorporated into programme design decisions.

Reading 3: Bicchieri, C.; Jiang, T.; & Lindermans, J. W. (2014). A social norms perspective on child marriage: The general framework. Commissioned and to be Published by UNICEF.

This paper describes a theoretical framework that integrates different explanations of child marriage. It distinguishes social norms from other belief systems and provides insights about how individuals make decisions, backed by Bicchieri’s theory of social norms. These insights guide the development of measurement tools for child marriage M&E.

Reading 4: Moneti, F. & Mackie, G. (2013). The general considerations in measuring social norms. Presentation based on longer article.

To date, most researchers who have attempted to measure “gender norms” have instead collected information on beliefs and attitudes. This presentation outlines a number of general considerations and measurement strategies to help with capturing norms and their change on the following three elements: empirical and normative expectations (beliefs about beliefs); consequences of action (e.g. perceived rewards or sanctions); reference group (which maintains compliance to normative expectations through application of informal sanctions).

For a more elaborate and critical discussion on the measurement of norms, see the full article: Mackie, G. et al. (2015). What are social norms? How are they measured?  California, USA: UNICEF/Centre on Global Justice, University of San Diego.

Page numbers below indicate the different measurement elements outlined in the presentation:

  1. Questions to identify reference network (pg. 45-48 50-53, & 64-65)
  2. Questions to identify what is typical & appropriate (pg. 54-56)
  3. Identifying social norms in qualitative research (pg. 57-60)
  4. Cues from DHS & MICS data suggesting the presence of a social norm (pg. 59-63)
  5. Questions that can be incorporated into surveys to gather data on ‘belief about others’ and how to interpret this data (pg. 49-50)
  6. Methodological implications, including social desirability bias (pg. 47)

Reading 5: Paluck, E.L., & Ball, L. (2010). Social norms marketing aimed at gender based violence: A literature review and critical assessment. New York: International Rescue Committee.

This paper reviews three case studies, engaging in a critical analysis of their design and evaluation in order to generate a list of considerations to guide future norms inventions aimed at GBV, and more widely.  It argues that it is more effective to target injunctive vs. descriptive norms, and provides examples of the risk of circulating and further entrenching negative descriptive norms. It also discusses the inadequacy of the GEM scale as a measure of social norms.

Questions to guide the readings

  1. Social norms are often held in place by a number of factors. Using an example of a harmful practice from your setting, map the relationship between structural factors, social norms and individual attitudes and agency in the context of this behaviour. What strategy will be most effective for promoting norm change related to the behaviour?
  1. What are the social vs. non-social explanations that account for child marriage? Why is this distinction crucial in the design of programmes aimed at ending the practice?
  1. Why is the investigation of non-compliance crucial for discerning norms? What would be a possible strategy to investigate sanctioning mechanisms? (Note: if a norm is effective, then one would rarely observe what happens in the context of non-compliance).
  1. Why is the evidence for the effectiveness of social norms marketing programmes aimed at GBV mixed? What are some of the methodological challenges in evaluating such programmes and how can they be minimised?
  1. Can you think of an example from your setting where a campaign slogan may have inadvertently reinforced harmful descriptive norms?

Watch Lori explore social norms further in our webinar discussion in February 2016.


Send us feedback on this reading pack

Suggested citation

Heise, L. & Manji K. (2016). Social Norms. GSDRC Professional Development Reading Pack no. 31.  Birmingham, UK: University of Birmingham.