Social Norms and Sanitation Practices in India
Introduction
Problem Statement
This research is part of a broader three-year project designed to identify the social networks and norms affecting individuals’ open defecation behaviours.
This research is part of a broader three-year project designed to identify the social networks and norms affecting individuals’ open defecation behaviours.
My role
Drive data analysis to identify key networks effects and other actionable insights for the purpose of designing behavioural interventions.
Drive data analysis to identify key networks effects and other actionable insights for the purpose of designing behavioural interventions.
Project Lead
Dr. Cristina Bicchieri
University of Pennsylvania's PennSoNG
Dr. Cristina Bicchieri
University of Pennsylvania's PennSoNG
Funded by
Bill and Melinda Gates Foundation
Bill and Melinda Gates Foundation
Project Outcome
Thesis of consolidated research findings
Thesis of consolidated research findings
Duration
3 months
3 months
Survey Scope
Purpose
1. Understand the social motivators that affect latrine use in rural and urban India
2. Identify demographic patterns, behaviors and endorsements in one's social networks
2. Identify demographic patterns, behaviors and endorsements in one's social networks
Sample Size
This survey reached a population between 16-65 years of age sampled equally from the two states in India -
1. Bihar (n=1702)
2. Tamil Nadu (n=1668)
1. Bihar (n=1702)
2. Tamil Nadu (n=1668)
Key Survey Blocks
1. Basic socioeconomic characteristics:
Demographics, economic status, and caste
Demographics, economic status, and caste
2. Household information
Ages, genders, health status, and migration history of household members
Ages, genders, health status, and migration history of household members
3. Egocentric network data about individuals in the one's reference network
4. Toilet ownership and use information and beliefs about the advantages and disadvantages of open defecation
Demographics
Looking at the distribution by gender, women make up of 55.1% of the respondents in Bihar and 47.1% of the respondents in Tamil Nadu.
A majority of the respondents identified as Hindus – 73.6% in Bihar and 85.4% in Tamil Nadu. 26% of the population in Bihar and 7.9% of those in Tamil Nadu identified as Muslim. In
Tamil Nadu, 6.5% identified as Christian.
Tamil Nadu, 6.5% identified as Christian.
Latrine Ownership and Usage
43.7% of respondents in Bihar and 32.3% of those in Tamil Nadu stated that the last time they had to defecate, they openly defecated
A little over half the respondent population in Bihar (54.3%) and Tamil Nadu (59.4%) are the sole owners of latrines or share them with other households.
When respondents were asked about access to a community toilet, 88.9% of Bihar and 49.8% of Tamil Nadu lacks access to one.
When respondents were asked why they openly defecate, 93% in Bihar and 60% in Tamil Nadu stated the lack of access to a latrine was the main reason.
This high correlation between latrine ownership and usage suggests that one potential strategy to reduce open defecation could be to formulate interventions focused on latrine construction.
Social Network Analysis
Place-Based Networks
Here the relationship between open defecation rates and the frequency with which respondents visit certain locations was studied.
Of the 10 places where this behaviour was measured temples, mosques, movie theatres, funerals, and street plays had a statistically significant influence on open defecation behaviour.
The most significant relationship was measured at temples.
Method
Logistical Regression (since the behaviour had only two possible outcomes i.e. open defecation or not.)
Logistical Regression (since the behaviour had only two possible outcomes i.e. open defecation or not.)
Result
Statistically significant (p-value < 0.001)
Statistically significant (p-value < 0.001)
This analysis proved that respondents were 39.5% less likely to openly defecate if they visited the temple once.
We further expanded the model to measure added effects of gender.
Result Statistically significant (p-value < 0.01)
For one visit to the temple, we expect to see a 38.4% increase in the probability of a man not openly defecating. For a woman who visits the temple once, there is a 40.8% increase in the probability of her not openly defecating.
Insight
While temples are a location that can be leveraged for interventions, there is no added benefit in having separate interventions for men and women.
While temples are a location that can be leveraged for interventions, there is no added benefit in having separate interventions for men and women.
Social Network Interactions
Next, we measure who the respondents spoke to most frequently in their reference network.
Of the 7 relations whose interactions with the respondent were measured, the village headman and landlord had a statistically significant influence on open defecation behaviour. The most significant relationship was with the village headman. The gender effects were also studied.
When considering visits to the village headman once a month, men were 41.2% less likely to openly defecate and women were 26.9% less likely to openly defecate.
Measuring Mass Media Effects
We measured whether people had ever seen advertisements promoting latrines across various mass media channels and 57% said they had.
Our regression analysis showed us that people are 29.8% less likely to openly defecate if they saw an advertisement across mass media.
Result
Statistically significant (p-value < 0.001)
Statistically significant (p-value < 0.001)
Of those who said they have seen advertisements promoting latrines, 90% of the respondents had not heard radio advertisements and 75.6% had not seen advertisements in the newspaper.
When we look at television advertisements, 92.1% of the respondents had seen television advertisements related to latrines. An individual who has seen a television advertisement was 28.5% less likely to openly defecate.
Insight
This analysis shows us that television and billboard advertisements would be the most efficient medium to use for behaviour change interventions.
Insight
This analysis shows us that television and billboard advertisements would be the most efficient medium to use for behaviour change interventions.
Identifying Trendsetters
Based on social norms research, trendsetters are those who:
- are highly central to a social network or have a high status in society
- are more likely to abandon existing social norms
- are highly central to a social network or have a high status in society
- are more likely to abandon existing social norms
Next, we conducted a qualitative text analysis on the unstructured responses in the survey. We see that the most influential people fell into three categories – politicians, cricketers and actors. The most respected person by far is Prime Minister Narendra Modi, who is by far the second most followed politician in the world on social media.
Based on this analysis, we can conclude that using trendsetters for future behavioural interventions to curb open defection could be highly effective.
