Teacher Survey in India: Question Bank, Sampling and How to Read the Results

Each purpose has a different owner and a different cut of the data. Pick one per survey.

The policy gives you yardsticks. NEP 2020 says each teacher is expected to take part in at least 50 hours of continuous professional development a year (paragraph 5.15). The same document says teachers should no longer be engaged in work not directly related to teaching (paragraph 5.12). It also says teachers will have more autonomy in choosing pedagogy (paragraph 5.14) (NEP 2020, NCERT copy).

Use a five-point agree-to-disagree scale unless the table says otherwise. Ask about the current term so every teacher answers about the same period.

Question 4 is the one principals act on fastest. Put the hours in bands, because teachers will not add up a week in their heads. Question 6 matters even where the official ratio looks healthy. UDISE+ reports a pupil-teacher ratio of 10 at the foundational stage and 21 at secondary (PIB, 7 July 2026), but those are national averages, not one teacher's largest section.

Professional development items should produce a list you can fund. Curriculum items should name the book and the grade.

Question 10 maps to the 50-hour expectation in NEP 2020 paragraph 5.15, and question 17 to paragraph 5.14 on autonomy (NEP 2020). Report the share of teachers in each band, not an average. In a board or publisher study, pipe the textbook name into questions 14 to 16 so each teacher sees only the book they teach.

Edtech items should separate access from use, and use from value. In 2025-26, 69.9% of schools had computers and 67.4% had internet (PIB factsheet, 17 August 2026), so ask about the device before the product.

Question 24 is the item an edtech company should watch. High use with a low score on 24 means the tool is used because it is compulsory. Pair it with 23 to see whether teachers believe it helps learning.

Anonymity comes first. Do not collect a teacher's name, employee ID or phone number with the answers. The only fields you need are the cuts you will report: school type, stage taught, subject group, years of experience in bands. Set a minimum group size before fieldwork and drop any cut below it. With an average of seven teachers per school (PIB factsheet, 17 August 2026), a single school rarely has enough people for a cut by subject. Report it as one group or not at all.

Women are 54.9% of teachers nationally (PIB factsheet, 17 August 2026), but a primary school may be almost entirely women, and a gender cut there identifies the men. Check each school's cell, not only the total.

Sampling across school types matters for boards, publishers and edtech companies. Teachers in government, government-aided and private unaided schools face different workloads, pay cycles and devices. Set quotas by management type, by stage (foundational, preparatory, middle, secondary) and by medium of instruction, and report each quota's base. For a chain survey, invite every teacher and report the response rate by school.

Decide who is in scope. Contract and guest teachers, special educators and pre-primary staff often carry the heaviest loads and are the easiest to leave off the list. Include them.

Avoid the first two weeks of the year, board exam months, result weeks and the week before a vacation. Good windows are the middle of each term, after unit tests and before exam preparation. For a textbook study, ask after the book has been taught for a full term. For an edtech study, ask about eight weeks after roll-out and again at the end of the year with the same questions.

Run the survey in the language of instruction at each school, with English as an option rather than the default. Back-translate every version, then have a teacher at that school read it on a phone before launch. Keep scale labels identical across languages and across waves. The employee survey question bank has the wording rules that survive translation.

Keep it short enough to finish in a free period on a phone: 12 to 20 items from the bank, plus two open questions. Send it through the principal with a note that answers are anonymous and that results will be shared back, and say when.

What we would do. We would pick the middle of the second term, send the link through the principal on a Monday and close it on Friday. That gives you one clean window to repeat next year and a trend you can trust. Keep the questionnaire to one free period; a teacher who runs out of time between classes answers the last items without reading them.

Report every agreement item as the share who agree or strongly agree, with the base. Do not average the scale. A mean score tells a principal nothing. "This share of secondary teachers agree their load is manageable, out of this many who answered" tells them where to look.

Read the bands in questions 4, 10 and 26 as distributions. Suppose a large share of teachers report more than seven hours a week of non-teaching work. You now have a number to take to the management committee, and NEP 2020 paragraph 5.12 to put next to it (NEP 2020).

Cross use against value. For edtech, put question 20 (how often) against question 23 (helps learning). High use with low value is a compliance problem. Low use with high value is an access or time problem, and questions 19, 21 and 22 tell you which.

Read the open answers yourself before anyone codes them. Group them into themes, then quote two or three word for word. Within a month, share the results with teachers and name the two changes you will make.

Internal teacher surveys usually go to the school's own staff list. Our panel suits the market version of this job, such as an edtech product study among teachers found with a screener, and this is what a pilot and main study cost at list price with respondents included. Screened-out respondents do not count towards a bundle.

Pro is sold in bundles of 400 to 1,000 responses and costs 20% less per response than Plus. Enterprise plans come with a dedicated research team and are priced on request.

The SuperJ network has 2 crore Indian respondents. The survey-eligible panel is a subset, and availability depends on your targeting. Members include professionals, so name the teacher profile you need and ask for a feasibility count before you commit. Members complete Aadhaar-based KYC once, and each study receives a zero-knowledge proof of verification. Surveys run in as many Indian languages as the study needs. Quality checks flag speeders, straight-lining and inconsistent answers. Read /panel-methodology and pricing.

The limit: the app reaches smartphone users only, so teachers without a smartphone need a paper or face-to-face version.