Navigate

Explore Colleges

Career OS

School (6th–12th)

Mentorship

AI Tools

Career Counselling · Methodology

How we score — and what we don’t claim

We would rather be honest than impressive. This page is the whole method.

1. The questions

Interests: 36 statements for Classes 6–8 (Explorer) and 48 for Classes 9–12 (Decision), six or eight per Holland type — Doers (R), Thinkers (I), Creators (A), Helpers (S), Persuaders (E), Organisers (C). Each is answered on a five-point “would you enjoy this?” scale. We wrote these items ourselves for Indian school life, in English and Tamil. They are not the O*NET Interest Profiler and have not been normed.

Work values (Decision only): 12 statements, two per value — achievement, independence, recognition, relationships, support, stability & conditions — rated by importance.

Work-style preferences (Decision only): 10 short statements adapted from the public-domain IPIP item pool, two per factor, with reverse-keyed items. We show them as pairs of strengths (Planner ↔ Flexible), never as personality types or scores a parent could read as a grade.

Subjects: the subjects you enjoy, feel strong in and find hard (self-report), plus — when you have taken practice tests on FirstGrade — your measured subject accuracy. Measured evidence always outranks self-report.

2. Scoring

Each type score is the average of its items, rescaled to 0–100. Your three highest types form a code (for example ISA). If the spread between your highest and lowest type is under 15 points we do not force a code — we say the profile is still even and exploration is the right next step.

3. Matching careers

Every career on our Career Roadmap carries one to three interest tags. Each tag maps onto Holland types with these fixed weights:

A career’s fit is the average over its tags of (weight × your type score). Then, in order:

• Class 11–12 with a chosen stream: careers your stream cannot enter are removed, not down-ranked.
• Class 9–10: careers that start only after graduation lose 3 points (they are far away, not wrong).
• Values: if stability scores 70+, government and public-sector routes gain 5; if achievement or recognition scores 70+, careers with strong earning growth gain 3.
• Subjects: a measured strong subject that a career’s stream needs adds 4 (max 8); a self-reported one adds 2 (max 4); a measured weak subject removes 3.
• Demand: “very high” demand adds 2, “high” adds 1.

The top 12 careers and the top 3 fields (average of each field’s best three careers) are shown, each with the rules that fired written out. Nothing is hidden and the same answers always give the same result.

4. The written summary

An AI model writes the plain-language summary from your scores and the matched reasons only. It is instructed never to invent facts, never to diagnose, never to use words like intelligence or personality type, and its output is checked for those words before you see it. If the check fails you get a fixed, rule-based summary instead.

5. What we do not claim

• We have not published reliability or validity studies for these items. When we have enough completions we will publish internal-consistency figures per scale — and that still will not make it a validated test.
• There are no norms: your scores are not compared with other children, and you are never ranked.
• We do not measure intelligence, learning styles or “multiple intelligences”, and we do not believe those labels help a child.
• A profile can change within a year. That is normal. Refresh it yearly.

6. Your data

We store an age band, not a birth date; one parent email for consent and the parent report; your answers as numbers; your scores and matches. Students under 18 cannot start without a parent or guardian’s consent. You or your guardian can request deletion at any time. Privacy policy