What 2,000 Educators Should Tell Us (And How We Will Know If They Did)

Most edtech research asks vague questions: "What do you think about technology in education?" or "Would you use an AI-powered lesson plan tool?" These questions produce vague answers that do not help build better products.

Our research is designed differently. Every question maps to a specific product decision. Here is what we are trying to learn, why it matters, and how we will know when we have enough data.

Target: 2,000+ Responses

Why 2,000? Because Nigeria has approximately 1.7 million teachers in public secondary schools alone (UBEC Statistical Report, 2023). A sample of 2,000 gives us a 95% confidence level with a margin of error of approximately plus or minus 2.2% for the teacher population. Practically, it means we can draw conclusions about patterns across regions, school types, and subjects.

The minimum viable threshold is 500 responses. At that point, we have enough data for directional insights. The milestone target of 1,000 gives us regional representativeness. The 2,000+ target gives us statistical significance across subgroups.

Data Point 1: Resource Creation Workflow

We ask: "How do you currently create your scheme of work? What tools do you use? How long does it take?"

Why this matters: If 80% of teachers create schemes of work in spreadsheets, the tool should import from spreadsheets. If 60% copy from a colleague's existing scheme, the tool should support adaptation workflows. If the average time is 8 hours per subject, the tool should target reducing that significantly.

Product decision: This determines whether the scheme of work builder is the primary feature or a secondary one. If teachers spend 8+ hours on it, it is primary. If they spend 30 minutes, it is secondary.

Measurement: We define success as identifying the top 3 workflow patterns used by teachers, with at least 100 responses in each pattern to draw reliable conclusions.

Data Point 2: Technology Access and Constraints

We ask: "What device do you primarily use for work? What is your typical internet speed? How often does your connection fail?"

Why this matters: If 70% of teachers use Android phones with 3G connections, the application must be optimized for mobile-first, low-bandwidth use. If 40% experience daily connection failures, offline support is not optional. It is essential.

Product decision: This determines the technical architecture priorities. Mobile-first versus desktop-first. Offline-first versus online-first. Data transfer budgets. Caching strategies.

Measurement: We need at least 300 responses with device and connectivity data to make reliable architectural decisions. Below 300, we risk optimizing for the wrong constraints.

Data Point 3: Assessment Practices

We ask: "How do you create assessment questions? Do you write them from scratch, adapt from past exams, or use question banks? How much time does this take?"

Why this matters: If most teachers adapt from past WAEC/NECO questions, the assessment tool should start with a question bank and let teachers modify. If most write from scratch, the tool should generate questions aligned with curriculum objectives and examination patterns.

Product decision: This determines whether the assessment feature is a question bank with search and filter, a question generator with curriculum alignment, or a hybrid.

Measurement: We need at least 200 responses per assessment practice category (scratch, adaptation, question bank) to draw conclusions about each approach.

Data Point 4: Sharing and Collaboration

We ask: "Do you share teaching materials with colleagues? How? What prevents you from sharing more?"

Why this matters: If sharing is primarily via WhatsApp, the sharing feature should generate shareable links that work in chat applications. If sharing is primarily via physical copies, the tool should support print-friendly formats. If the barrier is attribution concerns, the system needs clear licensing and credit mechanisms.

Product decision: This determines whether the ecosystem feature prioritizes a marketplace (structured discovery), a sharing network (peer-to-peer exchange), or both.

Measurement: We need at least 150 responses per sharing method to understand the dominant patterns.

Data Point 5: Institutional Requirements

We ask: "Does your school require specific formats for lesson plans and schemes of work? Who reviews your materials? What happens when a teacher leaves?"

Why this matters: If schools require standardized formats, the tool must support export to those formats. If materials are reviewed by department heads, the tool needs review workflows. If materials leave with departing teachers, the tool needs institutional ownership and access controls.

Product decision: This determines whether the institutional workspace is a feature for all schools or only for larger institutions with formal resource management needs.

Measurement: We need at least 200 responses from educators who can describe their school's resource management practices.

Data Point 6: AI Readiness and Attitudes

We ask: "Have you used AI tools for teaching? What worked? What did not? What concerns do you have?"

Why this matters: If most teachers have not used AI, the tool needs to introduce AI gently with guided workflows. If many have tried and been disappointed, the tool needs to demonstrate clear value before asking for trust. If concerns are about accuracy, the tool needs transparency about sources and confidence levels.

Product decision: This determines whether AI features are front-and-center or tucked behind manual workflows that build confidence first.

Measurement: We need at least 200 responses about AI attitudes to identify the key concerns and opportunities.

How We Measure Research Success

The research is not successful when we hit 2,000 responses. It is successful when we can answer these six questions with confidence:

• What is the dominant scheme of work creation workflow, and how much time does it take?
• What are the technology constraints we must design for?
• How do teachers create assessments, and what would help most?
• How do teachers share materials, and what prevents wider sharing?
• What institutional requirements must the tool accommodate?
• What is the AI readiness level, and how should we introduce AI features?

If we hit 2,000 responses but cannot answer these questions (because the sample is skewed, or the questions did not elicit specific enough answers), the research has failed.

This is why we designed every question around a product decision. We are not collecting data for its own sake. We are collecting data that tells us what to build and how to build it.

Take the Survey

If you are a Nigerian educator, your experience directly shapes what Curriva builds next. The survey takes approximately 15 minutes. You can participate without creating an account. Take the educator survey.

Related Reading

For the methodology behind the survey design, see How We Designed Our Educator Survey. For the workflow we are researching, see How Teachers Actually Create Schemes of Work. For what we already know from curriculum analysis, see NERDC Curriculum by the Numbers.