Village, block and district-level polls that work where the network does not
Bharat Survey is a panchayat election survey agency for candidates for pradhan or sarpanch, block and zila parishad seats, and for party district units and researchers. Rural local body contests turn on hamlets, water, roads, scheme access and the candidate's personal standing. Our field teams interview on an offline Android app, every record carries GPS and time, and results are reported for the seat you are actually contesting.
Panchayat polls are three contests at once in most states: gram panchayat, the intermediate block tier and the zila parishad. A voter can pick different people for each. Our teams go hamlet by hamlet, including the outlying tolas that a roadside survey skips.
Panchayats are set up under Part IX of the Constitution, and their elections are run by the State Election Commission under Article 243K. Rules on reservation, party symbols and campaign timing are set by each state.
The most common mistake in a panchayat election survey is treating it like an assembly poll. At gram panchayat level a candidate's family, caste networks, past work on the village road or handpump and the way scheme lists were handled often matter more than party. In several states candidates at this level do not contest on party symbols at all. We ask about the candidate as a person and about local delivery, and we keep party questions for the tiers where they make sense.
Sampling starts from the panchayat's own geography: revenue villages, hamlets and wards. Households are selected at random within each hamlet in proportion to its size, so the main village does not speak for the whole panchayat. Where a zila parishad seat spans many panchayats, we cluster by panchayat first.
Rotation of reserved seats changes who can contest and often who voters expect to win. Questions about reserved categories are asked only where legally relevant, in neutral wording, and reported in aggregate. No individual's caste or vote is ever shared.
Use this to match the tier you are contesting with a sensible survey size and scope.
| Tier | Typical geography | Main survey question | How we sample |
|---|---|---|---|
| Gram panchayat ward member | One ward of a village | Recognition and local standing | Often a near-census of households in small wards |
| Pradhan or sarpanch | Whole gram panchayat | Preference, past work, scheme access | Random households per hamlet, weighted by size |
| Block tier member (panchayat samiti or BDC) | Several panchayats | Preference and issue priorities | Panchayats as clusters, households within |
| Zila parishad member | Many panchayats in a district ward | Preference, party role, turnout intent | Two-stage cluster sample |
Names of the tiers and whether the intermediate tier exists vary by state. Article 243B allows states with population under 20 lakh to skip the intermediate level.
All political research services.
Learn moreSampled pre-election polls.
Learn morePolling-day voter interviews.
Learn moreField partner for regional news desks.
Learn morePolitical surveys in Hindi.
Learn moreSend the district, block and post.
Learn moreStraight answers to the questions people ask most about this topic.
Panchayat elections are conducted by each state's State Election Commission under Article 243K of the Constitution. The SEC notifies the schedule, reservation of seats and rules for the poll, so our survey timing and any exit poll release are checked against that state's orders.
Yes. Interviews are recorded fully offline on Android phones with GPS and time stamps, and the data syncs once the phone reaches network. Supervisors can still see where and when each interview happened.
Yes. For block and zila parishad seats we sample panchayats as clusters and households inside them. The report gives the seat estimate and, where the sample allows, results by panchayat.
Only where it is relevant to reserved-seat awareness or analysis the client needs, in neutral wording, with consent, and reported in aggregate. We never share an individual household's answers.
Send the district, block, post and expected polling month. We will reply with a hamlet-wise sample plan.
