Multiple Intelligence Mapping
A web-based psychometric assessment and academic recommendation system designed to evaluate students’ multiple intelligences and vocational interests to guide them toward suitable academic strands and career pathways.
Tech Stack
Why I Built This
Most academic strand selection processes rely heavily on grades, generalized academic performance, or subjective guidance counseling, which often fail to reflect a student’s actual cognitive strengths and vocational interests. Existing multiple intelligence assessment systems also tend to use extremely limited question sets, producing weak and unreliable recommendations.
I wanted to build a more structured and evidence-driven assessment platform that combines Multiple Intelligence profiling and RIASEC vocational interest analysis into a single system capable of generating personalized academic strand and career recommendations.
The system was designed to help Grade 10 students better understand their dominant intelligences, evaluate vocational tendencies, and receive more aligned recommendations for Senior High School strands such as STEM, ABM, HUMSS, and TVL.
How It Works
Students begin by completing a structured psychometric assessment composed of Multiple Intelligence and RIASEC-based questionnaires using Likert-scale responses. Each question is mapped to a specific intelligence or vocational interest domain.
The backend processes all submitted responses and computes normalized scores across nine Multiple Intelligence categories and six RIASEC vocational dimensions. These scores are then passed through a weighted recommendation engine that evaluates strand compatibility based on predefined cognitive and vocational mappings.
Assessment data, computed scores, and recommendation results are stored inside PostgreSQL, allowing the system to generate historical reports, intelligence profiles, and analytics dashboards for both students and administrators.
The React frontend visualizes intelligence distributions, dominant strengths, strand rankings, and career recommendations using interactive charts and dashboards. Administrators can also manage assessment questions, monitor participation statistics, and review aggregated intelligence analytics across all students.
The system additionally supports downloadable PDF reports containing complete intelligence profiles, strand recommendations, and career pathway suggestions that students can use as guidance for future academic planning.
Key Decisions
Weighted recommendation engine
Instead of directly assigning strands based only on the highest intelligence score, the system uses a weighted scoring model that combines Multiple Intelligence and RIASEC dimensions to generate more balanced and explainable recommendations.
Normalized psychometric scoring
Different intelligence categories may contain varying numbers of assessment items. To avoid biased results, all domain scores are normalized before recommendation calculations are performed, ensuring fair comparisons across all intelligence and interest categories.
Database-driven scoring architecture
Recommendation mappings between intelligences, vocational interests, strands, and careers are stored in relational database tables instead of being hardcoded in the application logic. This makes the system easier to maintain, recalibrate, and expand without modifying the scoring engine itself.
Assessment versioning support
Psychometric systems evolve over time as question banks improve. To preserve historical consistency, the system separates assessment versions and locks the active questionnaire version during each assessment session, ensuring older results remain reproducible even after future updates.
What I Learned
This project significantly improved my understanding of psychometric system design, recommendation engines, and large-scale assessment workflows.
I learned how to design weighted scoring systems, normalize multidimensional assessment data, and structure relational databases for dynamic recommendation architectures.
The project also strengthened my backend engineering skills, particularly around transactional assessment processing, role-based access control, analytics aggregation, and PDF report generation.
Beyond the technical implementation, I gained deeper experience in translating theoretical research frameworks such as Multiple Intelligence Theory and RIASEC into a scalable and maintainable real-world software system capable of supporting educational decision-making.
Let's Connect