Transparency & Rigour

Our Methodology

How we turn your responses into career insights — the frameworks, formulas, and scientific foundations behind every score.

Overview & Purpose

What This Assessment Measures

The SkillsBuff High School Assessment is designed to help students aged 14–18 discover career directions that align with their interests, values, and academic strengths. It is a career awareness tool, not a career prescription.

🎓For Students

Discover which career families, industries, and university majors match your unique profile.

👨‍👩‍👧For Parents

Understand your child's motivational drivers and how to support their exploration.

📋For Counselors

Use transparent, formula-backed insights to guide career conversations with evidence.

Assessment Structure
3RIASEC Scenarios
5Values & Motivation
5Work Style
14Career Anchors (8 scale + 6 tradeoffs)

Plus subject grades/interests and aspiration questions — ~27 total inputs

Scientific Foundations

Established Career Psychology Frameworks

Framework 1

Holland's RIASEC Theory

Developed by psychologist John L. Holland in 1959, RIASEC is one of the most widely used and empirically validated frameworks in career psychology. It classifies people and work environments into six personality types arranged in a hexagonal model:

RRealisticHands-on, practical, physical
IInvestigativeAnalytical, intellectual, research
AArtisticCreative, expressive, imaginative
SSocialHelping, teaching, counseling
EEnterprisingLeadership, persuasion, ambition
CConventionalDetail-oriented, organized, systematic
How We Apply RIASEC

1. Explicit measurement: 3 scenario-based questions where each choice maps to one RIASEC dimension. Each selected dimension receives 33.33 points (max 100 if all 3 match the same type).

2. Subject inference: Academic subjects map to RIASEC dimensions (e.g., Physics → Realistic, Computer Science → Investigative). Each subject contributes: (interest × 0.6 + grade × 0.4) × 20

3. Merging: Final score = explicit × 0.70 + inferred × 0.30. Explicit responses carry more weight, but subject data helps fill gaps and validate self-reports.

4. Adjacency smoothing: Holland's hexagonal model predicts that adjacent types (e.g., R–I, I–A) share traits. We apply a 10% neighbor boost: smoothed = score + (left_neighbor × 0.10) + (right_neighbor × 0.10). This prevents artificially narrow profiles.

Framework 2

Schein's Career Anchor Theory

Developed by MIT professor Edgar H. Schein in 1990, Career Anchors represent the core professional values that a person will not give up, even under pressure. They predict long-term career satisfaction more reliably than skills or interests alone.

TFTechnical Excellence
GMGeneral Management
AIAutonomy
SSSecurity
ECEntrepreneurship
SDService
PCChallenge
LSLifestyle
How We Apply Career Anchors

Step 1 — Base score: 8 Likert scale statements (1–5). Converted to: (rating − 1) × 12.5 producing a 0–50 base range.

Step 2 — Primary tradeoffs (4 pairs): Forced-choice “would you rather” questions pit paired dimensions against each other. Winner gets +30, loser gets −15.

Step 3 — Cross-cutting tradeoffs (2 pairs): These pit common co-winners against each other to break ties (e.g., Technical Excellence vs. Challenge). Winner gets +15, loser gets −8.

Step 4 — Clamp: All scores rounded and clamped to 0–100. Theoretical maximum is 95 (rate 5, win both tradeoffs).

Tradeoff Pairs
Primary: Expertise vs. Management, Security vs. Entrepreneurship, Autonomy vs. Service, Challenge vs. Lifestyle
Cross-cutting: Expertise vs. Challenge, Autonomy vs. Entrepreneurship

Why tradeoffs? Rating all anchors “5/5” doesn't differentiate. Forced choices reveal what you truly prioritize when you can't have everything.

Foundation Clusters

8 Cognitive & Skill Dimensions

We derive 8 foundation clusters from your academic performance and reported skills. These represent cognitive and skill dimensions that predict success in different career families.

Quant Reasoning
Logical Problem Solving
Communication & Writing
Visual/Design Thinking
Systems Thinking
Business/Market Thinking
Research & Curiosity
Collaboration/Leadership
How Clusters Are Computed

Each subject contributes a signal: (grade_signal × 0.60) + (interest_signal × 0.40) where signals are normalized to 0–100. Subject-to-cluster mappings are predefined (e.g., Mathematics → Quant Reasoning, English → Communication).

Skills and competencies contribute via keyword matching (e.g., “programming” → Logical Problem Solving). Final cluster score is the average of all contributing signals.

Subject ExamplePrimary ClusterSecondary Cluster
MathematicsQuant ReasoningLogical Problem Solving
PhysicsQuant ReasoningSystems Thinking
Computer ScienceLogical Problem SolvingSystems Thinking
English / LiteratureCommunication & WritingResearch & Curiosity
Art / DesignVisual/Design ThinkingCommunication & Writing
Business / EconomicsBusiness/Market ThinkingCollaboration/Leadership
History / GeographyResearch & CuriosityCommunication & Writing

How Career Pathways Are Scored

Role Family Fit Algorithm

Career pathways (role families like “Builder (Software)” or “Data Analytics”) are scored using a composite formula that balances four dimensions:

30%
Interest FitHow well your Holland RIASEC code matches the role family's ideal personality profile. Computed by comparing your top 3 RIASEC dimensions against the family's expected signals.
25%
Capability FitAverage of your foundation cluster scores for clusters relevant to this career family. A high Quant Reasoning score lifts Data-oriented families, for example.
20%
Values FitHow well your career anchors match the role family's typical values. E.g., 'Entrepreneurship' anchor aligns with startup-oriented families.
25%
Aspiration MatchBinary — did you explicitly express interest in this type of work? If you selected 'Building Apps' as an exciting outcome, the Builder family gets a full 25-point aspiration match.
Best (90+%)

Exceptional alignment across all four dimensions

Good (75-89%)

Strong alignment across core categories

Moderate (<75%)

Potential fit — requires more skill exposure

Gating Rules

Some families have hard prerequisites: Hardware/Embedded requires a Realistic (R) interest signal or Systems Thinking above 60, and AI/ML requires Quant Reasoning above 50 or an Investigative (I) signal. Without these, the capability score is halved to reflect the steeper learning curve.

How Roles Are Matched

Individual Role Fit Scoring

Within each career family, individual roles (e.g., “Frontend Developer”, “Data Analyst”) are scored with a weighted formula optimized for high school students:

30%
RIASEC FitCosine similarity between your RIASEC profile and the role's ideal profile. Measures personality alignment.
25%
Career Anchor FitAverage of your anchor scores on dimensions relevant to this role. E.g., Software roles map to Technical Excellence + Challenge + Entrepreneurship.
20%
Skill MatchComparison of your reported skill levels against the role's requirements. Partial credit for related skills.
15%
Workstyle FitHow your preferred work pace, structure, and environment align with the role's typical demands.
10%
Constraint FitPractical considerations — timeline to graduation, geographic flexibility. High school students receive generous defaults.
Why Personality Outweighs Skills for High Schoolers

At 14–18, students are still building skills. Research shows that interest alignment (RIASEC) and value fit (Anchors) are stronger predictors of long-term career satisfaction than current technical ability. That's why personality dimensions carry 55% of the weight, while existing skills carry only 20%.

How Industries Are Scored

Industry Fit Algorithm

Industries are scored as aggregates of the roles they contain. The intuition: if you're a strong fit for many roles within an industry, you'll likely thrive in that sector.

70%
Weighted Role FitWeighted average of your fit scores across all roles in the industry, where popular/core roles carry more weight.
20%
Industry InterestDid you express interest in this industry? Preferred industries score 90, others receive a 40 baseline.
10%
Role CoverageWhat percentage of the industry's roles are you a good fit (>50) for? Higher coverage means more career flexibility within the sector.

How Majors Are Suggested

University Major Fit Algorithm

Major suggestions help students explore which university degree programs align with their academic strengths and career direction. The formula prioritizes academic foundation because degree choice depends heavily on readiness.

50%
Foundation AlignmentCosine similarity between your 8 foundation cluster scores and the major's ideal cluster profile. A Computer Science major weighs Quant Reasoning and Logical Problem Solving heavily.
40%
Workstyle AlignmentCosine similarity between your workstyle preferences and the major's typical learning environment (structured vs. creative, collaborative vs. independent).
10%
Target Role AlignmentHow well the major's career outcomes overlap with your target role families. If you're aiming for Software Engineering, a CS major gets a boost.
Cosine Similarity

We use cosine similarity — a standard vector comparison technique — to measure how “directionally aligned” two profiles are. Two profiles pointing in the same direction score near 1.0, even if their absolute magnitudes differ. This means a student with moderate but well-distributed scores can still match strongly with the right major.

Limitations & Disclaimers

Important Context for Interpreting Results

Starting point, not a final answer

This assessment is designed to spark exploration and conversation. It does not determine your career. Use it as one input among many.

Self-reported data

All scores are based on your own responses. Accuracy depends on honest, thoughtful answers. Social desirability bias (answering how you think you 'should') can distort results.

Limited question set

With ~27 questions, we cannot capture every nuance. RIASEC is measured with only 3 scenario questions plus subject inference — some personality dimensions may be underrepresented. Professional RIASEC assessments typically use 30-60 items.

Interests evolve

Career interests are not fixed, especially during adolescence. We recommend retaking the assessment every 6-12 months to track how your profile develops over time.

Not an aptitude or IQ test

This assessment measures interest alignment and value fit. It does not measure cognitive ability, aptitude, or intelligence. A low score in a career family does not mean you lack the ability to succeed there.

Not a substitute for professional counseling

While built on established frameworks (Holland's RIASEC, Schein's Career Anchors), this tool should complement — not replace — guidance from qualified career counselors, teachers, and mentors.

Cultural and socioeconomic context

Career assessments developed in Western academic traditions may not fully capture the values, constraints, and opportunities of every cultural or socioeconomic context. Interpret results with your own lived experience in mind.

Transparent design choices

Our formulas use specific weights (e.g., 30% Interest, 25% Capability) that reflect professional judgment informed by career psychology literature. These are not absolute truths — they are reasoned design decisions that we make transparent here.

No employment guarantees

SkillsBuff provides career guidance and learning resources. Outcomes depend on individual effort, market conditions, educational choices, and many other factors beyond the scope of any assessment.

References

Scientific & Industry Sources

Holland, J.L. (1997). Making Vocational Choices: A Theory of Vocational Personalities and Work Environments (3rd ed.). Psychological Assessment Resources.

The foundational text on RIASEC theory, used worldwide in career assessment.

Schein, E.H. (1990). Career Anchors: Discovering Your Real Values (Revised ed.). Pfeiffer & Company.

Defines the 8 career anchor dimensions used in our values assessment.

O*NET OnLine (Occupational Information Network). U.S. Department of Labor. https://www.onetonline.org/

Industry-standard database for occupational skill requirements and work contexts. Informs our role-to-skill mappings.

Nauta, M.M. (2010). "The Development, Evolution, and Status of Holland's Theory of Vocational Personalities." Journal of Counseling Psychology, 57(1), 11-22.

Meta-analysis confirming the empirical validity of Holland's RIASEC model across cultures.

Feldman, D.C. & Bolino, M.C. (1996). "Careers within Careers: Reconceptualizing the Nature of Career Anchors and Their Consequences." Human Resource Management Review, 6(2), 89-112.

Examines how career anchors evolve and their predictive validity for career satisfaction.