Social norms, an emerging field in development, is an area where a lot of measures developed for WEIRD contexts are applied globally, potentially damaging the accuracy of results. We partnered with the University of San Diego to build the Strength of Social Gender Norms (SSGN) scale — a social norm measure about women working outside the home. We built this measure from the ground up in three phases. They were:
- Development of item content, including reviews by field experts and cognitive interviews with young adolescents.
- Creating the scale using item reduction and factor extraction
- Assessing the scale properties for internal consistency, test-retest reliability, construct validity and known-groups validity.
We tested the scale with 399 adolescents aged 14–17 years enrolled in secondary schools across urban-rural regions of Uttar Pradesh, India. Our ability to validate this measurement with a hard-to-reach, understudied sample — a neglected one even in norms and psychology research — represents an undeniable merit of the Strength of Social Gender Norms (SSGN). Our domain-specific measure has response patterns that were adapted to reflect positive (‘their family will be celebrated’), neutral (e.g. no one will say anything), and negative (‘their family will be shunned’) responses by the community at large that are typically heard or used in the cultural setting in India.
Ideally, the set of items a measure includes must capture as much of the system of social norms that bear on whether or not women can work outside the home. In our case, the literature recognises that it’s not only the norm of working outside the home that presents a barrier. Women in India and other developing countries, are also confronted with the expectation of being bound to household work, felt inappropriateness of daughters financially supporting their parents, and awareness of the unsuitability of interacting with men outside of one’s own household, amongst other social rules that indirectly bear on the practice of being able to earn an independent livelihood. By only measuring social perceptions about working outside the home, we would neglect crucial factors that may interfere with efforts to strengthen the social desirability for women to work. Our scale thus reoriented the focus from a single item to a broader set of indicators that captured these indirect social norms.
How we measure a construct, in behavioral science, determines which evidence is used to understand what the bottlenecks are and the interventions we ultimately put forth as recommendations to practitioners. To advance development outcomes in the global south, we must pay closer attention to how we choose to measure the pathway we rely on to change perceptions, attitudes or behavior. When a valid and reliable tool that is fit for this purpose and suited to the context that we deliver interventions in is not available, we must spend time and resources on measurement development.
Read the full project working paper here.