Adapting and Validating the G-NORM (Gender Norms Scale) in Nepal: An Examination of How Gender Norms Are Associated with Agency and Reproductive Health Outcomes

"Findings offer an improved measure of gender norms in Nepal and provide evidence that gender norms are critical for agency and reproductive health outcomes."

As gender equality gains recognition as a public health priority, more studies seek to measure gender norms as an important outcome, as well as a barrier to or facilitator of behaviour change. This study extends existing gender norms measurement tools to adapt and validate the G-NORM, a scale originally tested in rural India that measures perceptions of community-level gender norms (and includes both descriptive and injunctive norms as two subscales) in rural Nepal. It also demonstrates construct validity by examining the scale's association with several sexual and reproductive health (SRH)-related constructs..

The research is grounded in the Theory of Normative Social Behavior (TNSB), which holds that our basic human desires to belong and to connect lead us to care a great deal about what we perceive people in our community are doing (descriptive norms - e.g., the belief that most parents in the community pay a dowry for their daughter's marriage) and what we believe they think we should be doing (injunctive norms - e.g., the belief that that parents should pay a dowry when their daughter is married). Descriptive norms serve to identify perceptions of behaviour, whereas injunctive norms serve to motivate behaviors to seek social rewards or avoid sanctions.

The data used in this paper to adapt and validate the G-NORM scale come from the last wave of a longitudinal study examining nutrition and gender empowerment among newly married women in the Nawalparasi district in Nepal. The status of women, including household decision-making, is lower in this area. At baseline, women were eligible if they had been married within the last four months, were 18-25 years old, and were currently co-residing with their mother-in-law. The sample for this study represents 93% of the original sample; at the time the G-NORM was asked, 86% were still living with their mother-in-law.

To ensure that the prior items from India and the new items developed were relevant to the Nawalparasi district, the researchers first analysed qualitative data collected from the community in the same district. They found that many of the inequitable gender practices they identified in India were also relevant in Nepal, which was not surprising given their geographic proximity and sociocultural similarity. They also identified additional inequitable norms that were linked to contraceptive autonomy around the use of family planning methods, decision-making about when to have the first child, and decision-making around the total number of children to have. Other inequitable gender practices were parents choosing whom their daughter would marry (an arranged marriage) and parents paying a dowry when their daughter gets married.

To adapt and validate G-NORM, the researchers translated all items from English to Nepali and checked the translation with multiple researchers. They initially included a total of 28 items: all 18 items from the original G-NORM and 10 new items developed for this context based on analysis of qualitative data from the Nepali study communities, review of the literature, and expert input. After cognitive interviewing with draft questionnaire items, they determined the dimensionality of the scale and identified and removed poorly performing items - leaving them with a 20-item scale (10 items for each subscale). They validated G-NORM by applying confirmatory factor analysis and examining associations with outcomes hypothesised to be associated (e.g., decision-making around family planning and intent to use family planning in the future).

Like the original G-NORM, the analyses showed that descriptive norms and injunctive norms comprise two distinct scales that fit the data well and had Cronbach alphas of 0.92 and 0.89. More equitable descriptive gender norms were associated with higher decision-making scores, increased odds of intending to use family planning, disagreeing that it is wrong to use family planning, and older ideal age at marriage. Injunctive gender norms were only associated with disagreeing that it is wrong to use family planning.

As in in India, the researchers "found that women reported more equitable injunctive norms (perceptions around gender expectations) compared to descriptive norms (perceptions about actual behavior). This illustrates that perceived expectations may be changing faster than perceived behaviors around gender norms. This has implications for interventions trying to change gender norms as it is important to measure both perceived expectations and perceived behaviors." The scale with separate subdomains - descriptive norms and injunctive norms - "allows researchers to do just that".

On a practical level (interventions), the researchers note that "changing descriptive norms may be easier (by communicating, e.g., that many others are engaging in the focal behavior) than changing injunctive norms (which requires communicating about perceived levels of support), but which of these changes is more instrumental in driving behaviors remains unclear." It is also unknown at this time "which of the two norms decay (or strengthen) more quickly. This is a ripe area for future work."

In conclusion: "This improved measurement tool could be valuable to interventionists working to reduce gender inequalities in Nepal or on other sexual and reproductive health programs known to be associated with gender norms (e.g., contraceptive agency, attitudes about contraception, etc.)....Given that global crises, such as the COVID-19 pandemic, threaten gender equality, entrench restrictive gender norms, and increase the risk of harmful practices, particularly in Nepal, this improved measure is needed....It is important, now more than ever, to properly measure how and which gender norms are changing."