Postscript #14: The Importance of Doing Micro Data Collection

In Postscript, we reflect on the editorial choices we make and abandon. Plus all things behind the scenes. This week: anecdotes at scale

In November 2025, journalist Sarasvati Thuppadolla approached us with an idea – to collect data on how caste concretely shapes the working lives of ASHA workers. During her field reporting for BehanBox across Gujarat, Bihar and Uttar Pradesh, she encountered searing accounts of caste-based discrimination – within communities, among co-workers, and in interactions with officials. It stemmed from her concern that she did not want these experiences to be dismissed with the familiar refrain that “caste is everywhere, especially in rural India”. She wanted to document, with precision, how caste discrimination materially affects ASHA workers’ labour rights, wages, dignity, mobility and access to work itself, especially when so much of the ASHA movement has centred questions of labour and recognition.

We discussed a sample size of 100 Dalit ASHA workers across states but eventually settled for 50 to keep it wieldy. The final sample was 52 workers across six states. What we built was a micro survey. Of this came our three part series on ‘Caste and Care Work’.

In this edition of Postscript, we want to talk about the importance of doing micro data work. Far from being a lesser form of research, it is often an honest one, born of conviction: that something ought to be recorded, that a silence is worth breaking even without the resources.

The Politics of Small Data

At BehanBox, we believe in data. But we also have a feminist and decolonial approach to it – the politics of who collects, who is represented, and who holds stewardship over it.

A ‘micro survey’ may be taken to mean ‘limited’ or ‘preliminary’ or even ‘statistically insignificant’.We understand it differently and offer an alternate frame for it: a micro survey is a tightly scoped instrument designed to listen closely to people, understand nuance and offer a more localised and granular truth that large datasets are not designed to measure.

National level data tell us how many ASHA workers exist and even those have failed to keep a record of and publicly disclose data disaggregated by caste. What they thus cannot tell us is what it feels like to be an ASHA worker from a Scheduled Caste community, expected to provide care in a village where caste still governs everything.

Consider what we actually needed to understand: How many times have ASHA workers witnessed caste discrimination in their day-to-day work? What do these indignities, big and small, look like? How do their peers respond when it happens? How do institutions including unions make space for discussion and redressal? What is the economic and social cost of caste? How do social and political networks shape appointments and opportunities?

These aren’t questions a check-box dataset can answer. These need precision and care in recording, which can then be turned into tangible evidence that can shift policy and narrative, and in the long run offer transformative justice.

This form of data collection requires patience and persistence. Sarasvati spent close to four months finding enough Dalit ASHA workers who were willing to speak in the first place and eventually indulge her in her inquiry. Each conversation lasted anywhere between 60 to 90 minutes,;. It was not easy for ASHA workers to tap into their memories to recount their indignities, which was in equal measures hurtful and necessary to name. A micro survey in feminist hands, then, beyond methodology, is deeply concerned with ethics and empathy.

Listening became the central principle around which the research and instruments were built. With this study, and another one in 2020 where we surveyed 55 ASHA workers’ experiences during the pandemic, we inverted the conventional survey process. Instead of beginning with a fixed questionnaire, we began with long, detailed conversations with ASHA workers about their lives and work. Those conversations then informed the questions themselves. This approach allowed us to capture quantifiable data alongside the overlooked texture, nuance and granularity of lived experience, giving the research some degree of scale but richer, unmatched depth.

There was a subversiveness to collecting this data. Micro-data is inventive, akin to a statistical pushback against the structural limits of data systems obsessed with codification, standardisation and patterns. Think data in the form of a daily log tracking an ASHA worker’s time, a heat ledger documenting climate-linked income losses among working-class communities, or a time-use survey that asks not just how women spend their leisure, but whether leisure exists as a protected category of time for them at all.

Intellectual honesty requires us to be equally clear about the limits. Micro data cannot quantify national prevalence or produce numbers suited to policy interventions and budget allocations. Nevertheless, it has its uses.

Lakshmi Kaurav, an ASHA facilitator in Madhya Pradesh, told us that officials routinely dismiss workers’ labour by asking: Tum log kitna kaam karti ho? ( How long do you even work?) Without official time-use data, without any logging system, their effort is invisible by design; a micro dataset of time use could directly contest that invisibility. As Lakshmi put it: “If government officers have log-in and log-out systems, why not us? This is one way to establish our status.” Micro data, in other words, offers a counter-narrative, making legible and visible what official systems are structured to ignore.

The intersection of caste and care work in India’s public health system is under-examined, not because researchers have looked and found nothing, but because the questions have rarely been asked with sufficient specificity and proximity to the people most affected. The ASHA programme has generated surprisingly little research that takes caste seriously as a structural variable rather than a demographic footnote.

We felt an urgent need to record this, and we chose to begin now, imperfectly, at a small scale. A micro survey is a small instrument but questions it raises need not be so.

That’s all for this month’s Postscript, Behans. Our eyes, ears, hearts, and inboxes are always open for your thoughts. Write to us or comment below.