Data Challenge 5.0 — Sustainable Well-Being
Explore how education and employment connect to well-being across Malaysian states. The Malaysian Well-Being Index (MyWI) reveals the full picture: from schools to jobs to quality of life. Click any state to see its complete well-being profile.
Currently showing: Graduate Employment Rate by State (2024)
Select any state from the list or click on the map to highlight it
Click on any state name to view its metrics
💡 Tip: Use this list for smaller states like Perlis or W.P. Labuan
👆 Hover over or click a state on the map to see its details
3.3%
Latest data (2023)
68.6%
Latest data (2023)
92.5%
National average (2024)
4.1M
Graduates produced
RM 6,338
Latest data (2022)
98.5%
Average (2022)
Engineering
94.9% employed (2024)
+24%
2023 → 2024 improvement
Select 2-5 states to compare their well-being indicators side-by-side
👆 Select states to start comparing
Choose 2-5 states to see their trends and metrics
Understanding unemployment patterns and employment composition across Malaysia (2016-2025)
Comparison of unemployment rates (2016-2025). Youth unemployment consistently runs 7-9% higher than overall rate.
Latest Youth Rate (2025)
10.2%
Latest Overall Rate (2025)
3%
Current Gap
7.2%
Distribution of employed persons by employment status (Latest data: 2025)
Employee
12,797.4k
74.8% of total
Own Account Worker
3,272k
19.1% of total
Employer
596k
3.5% of total
Unpaid Family Worker
464.1k
2.7% of total
Total Employed: 17,111.4 thousand
Source: Department of Statistics Malaysia - Monthly Labour Force Survey
States ranked by LFPR (highest to lowest). Colors indicate performance tiers: green (75%+), blue (70-75%), orange (65-70%), red (below 65%).
Highest LFPR
78.6%
Putrajaya
National Average
68.2%
across all states
Lowest LFPR
57.3%
Kelantan
State Range
21.3pp
percentage points spread
Exploring the economic prosperity-unemployment relationship across Malaysian states
Examining the relationship between economic wealth and unemployment across Malaysian states. The trendline and R² value (R² = 0.212) reveal a very weak correlation, indicating that GDP per capita is not a reliable predictor of unemployment; other factors are far more important.
R² Value (Model Fit)
0.212
Very weak relationship
Avg GDP per Capita
RM 37.8k
across all states
Avg Unemployment
3.5%
national average
Data Points
15
states analyzed
Regression Model
Unemployment = 3.25 + (0.0077) × GDP_per_capita
Model Fit
R² = 0.212 (21.2% of variance explained)
Interpretation
The data shows a weak positive correlation (R²=0.212), meaning states with higher GDP per capita have slightly higher unemployment rates. However, this relationship is very weak. GDP per capita explains only 21% of unemployment variation. The remaining 79% is determined by other factors. This counter-intuitive finding likely reflects that high-GDP urban centers (like Kuala Lumpur and Penang) attract job-seeking migrants, temporarily inflating unemployment rates. Additionally, wealthier states may have more educated populations who are selective about employment opportunities. The weak R² indicates that GDP per capita alone is not a meaningful predictor of unemployment. Other factors such as education-employment alignment, industry structure, migration patterns, and labor market policies play far more significant roles.
This finding appears counter-intuitive because economic theory typically suggests wealthier regions should have lower unemployment. However, in Malaysia's context, several factors explain this pattern:
Most importantly: The very weak correlation (R²=0.212) indicates that GDP per capita is not a reliable predictor of unemployment. State-level unemployment is much more strongly influenced by education quality, industry composition, labor market policies, and migration patterns: factors that vary independently of economic wealth.
Above Trendline (Higher unemployment than expected)
These states have higher unemployment rates than their GDP per capita would predict, suggesting structural labour market issues or economic transitions.
Sabah
Unemployment: 7.5% (Expected: 3.3%, Difference: +4.2pp)
GDP per capita: RM0.9k
Perlis
Unemployment: 4.4% (Expected: 3.5%, Difference: +0.9pp)
GDP per capita: RM26.7k
Labuan
Unemployment: 6.8% (Expected: 6.0%, Difference: +0.8pp)
GDP per capita: RM358.2k
Negeri Sembilan
Unemployment: 3.2% (Expected: 2.6%, Difference: +0.6pp)
GDP per capita: RM-79.5k
Perak
Unemployment: 3.9% (Expected: 3.4%, Difference: +0.5pp)
GDP per capita: RM19.2k
Below Trendline (Lower unemployment than expected)
These states have lower unemployment than their GDP per capita would predict, indicating effective labour market policies or strong job creation.
Melaka
Unemployment: 1.6% (Expected: 3.5%, Difference: -1.9pp)
GDP per capita: RM31.7k
Pahang
Unemployment: 2.0% (Expected: 3.5%, Difference: -1.5pp)
GDP per capita: RM32.1k
Pulau Pinang
Unemployment: 2.2% (Expected: 3.5%, Difference: -1.3pp)
GDP per capita: RM36.9k
Terengganu
Unemployment: 3.4% (Expected: 4.2%, Difference: -0.8pp)
GDP per capita: RM121.7k
Johor
Unemployment: 2.6% (Expected: 3.4%, Difference: -0.8pp)
GDP per capita: RM14.4k
Selangor
Unemployment: 2.7% (Expected: 3.3%, Difference: -0.6pp)
GDP per capita: RM8.8k
The Formula:
What this means:
How it's calculated: We use Linear Regression (Ordinary Least Squares) to find the line that best fits all 15 state data points. The algorithm minimizes the total distance between the line and all actual data points.
R² = 0.212 means the model explains 21.2% of unemployment variation.
The remaining 78.8% is explained by other factors not in this model.
How it's calculated:
Interpretation:
R² = 0.212 (21.2%) is very weak. GDP per capita alone is NOT a reliable predictor of unemployment.
The very weak R² (0.212) tells us that GDP per capita is not a good predictor of unemployment. Other factors like education-employment alignment, industry structure, labor market policies, and migration patterns are far more important in determining state-level unemployment rates. These factors vary independently of economic wealth, which is why two states with similar GDP can have very different unemployment rates.
Sectoral productivity analysis: output per hour and per employee
Comparing productivity across Malaysia's main economic sectors. Higher values indicate greater economic output per unit of labour.
Highest Productivity
RM 536.1
Mining and quarrying
Average Productivity
RM 137.0
across all sectors
Lowest Productivity
RM 22.0
Construction
Compare how graduate unemployment rates have changed from 2020 to 2024. Select states to compare.
1. Selangor
6.0%
2020: 13.0%-7.0pp
Highest unemployment
⚠️ Methodology Change: 2024 data uses a revised binary methodology (Employed vs Not Working Yet) and is not directly comparable to 2020-2023 data which used a 5-category classification.
Examining education supply, completion rates, and graduate production trends
Average completion rate across primary, lower secondary, and upper secondary levels (2022)
Highest Rate
Putrajaya: 102.6%
National Average
99.4%
States Above 100%
4 states
⚠️ What caveats I should bear in mind when using this data?
Because Malaysia's dropout rates are extremely low, completion rates may exceed 100% due to a small number of students repeating grades or transferring between states. Furthermore, it should be noted that the data refers to government schools only.
Bubble size represents student-teacher ratio. Hover over bubbles for detailed information.
Avg S-T Ratio
22.6
students per teacher
Lowest Ratio
18.5
better teacher access
Highest Ratio
29.4
more students per teacher
Total States
16
included in analysis
National aggregate of graduate output by gender across all states. Shows steady growth in tertiary education completion.
Total Growth (2020-2024)
+19.9%
994,200 more graduates
Latest Year Output
5.98M
graduates in 2024
Female Representation
53.6%
of total graduates (2024)
Male Representation
46.4%
of total graduates (2024)
Strategic insights from composite indices and K-means clustering analysis
This quadrant analysis reveals the strategic positioning of each state in terms of education supply and employment outcomes. States are classified into four categories based on median ESI and EOI values.
Optimal Alignment
High ESI + High EOI
Oversupply Risk
High ESI + Low EOI
Education Shortage
Low ESI + High EOI
Development Priority
Low ESI + Low EOI
Optimal Alignment
2
states performing well
Oversupply Risk
6
need job creation focus
Education Shortage
6
need education expansion
Development Priority
2
need comprehensive support
Policy Insight: States in different quadrants require tailored interventions. Optimal states should maintain balance, oversupply states need economic diversification, shortage states require education infrastructure, and development priority states need holistic support.
States grouped by similar education and employment characteristics using Ward's linkage method. Cluster centroids (marked with black border) represent the average position of each group. 8 features were used for clustering (ESI, EOI, unemployment, income, GDP, participation rate, completion rate, MyWI).
Advanced Economies
Developing Markets
Emerging Regions
Clustering Insight: Hierarchical clustering with Ward's linkage identified 3 distinct groups based on 8 standardized indicators including education supply, employment outcomes, unemployment rates, income levels, GDP, participation rate, completion rate, and well-being metrics. States within each cluster share similar characteristics and may benefit from similar policy approaches. Silhouette analysis confirmed k=3 as optimal.
⚠️ 2024 uses revised methodology - not directly comparable to prior years
Percentage of graduates not working, sorted from highest to lowest
National Rate
7.5%
Best Performer
Engineering Manufacturing & Construction
5.1%
Worst Performer
Education
10.2%
National Unemployment
📊7.5%
Public University
🏛️5.1%
Private HEI
🏢12.8%
Public vs Private Gap
⚖️2.51x
multiplier
Best State
🏆WP Putrajaya
4.7%
Worst State
⚠️Sabah
11.0%
Best Field
✅Engineering Manufacturing & Construction
5.1%
Worst Field
❌Education
10.2%
Comprehensive analysis of graduate employment outcomes across fields, states, and institution types
⚠ Note: 2024 uses different methodology (binary GE rate) vs 2020-2023 (5-category classification)
Engineering consistently outperforms, while Education and Services show higher unemployment rates. Note: 2024 uses different methodology (binary classification).
Best Performing
5.1%
Engineering Manufacturing & Construction
Highest Unemployment
10.2%
Education
Sabah, Kelantan, Terengganu show highest graduate unemployment in 2024. Note: 2024 uses different methodology.
Worst
11.0%
Sabah
National Avg
7.8%
All states
Best
4.7%
WP Putrajaya
Private HEI graduates consistently face higher unemployment rates than public university graduates. The gap has widened from 1.55x in 2020 to 2.51x in 2024.
📊 Private HEI graduates are 2.5x more likely to be unemployed than public university graduates (2024)
Gap: 7.7 percentage points | Public: 5.1% | Private: 12.8%
| Year | Public | Private | Gap | Multiplier |
|---|---|---|---|---|
| 2020 | 13.3% | 20.6% | 7.3pp | 1.55x |
| 2021 | 11.8% | 19.7% | 7.9pp | 1.67x |
| 2022 | 7.2% | 14.5% | 7.3pp | 2.01x |
| 2023 | 7% | 17% | 10pp | 2.43x |
| 2024⚠ | 5.1% | 12.8% | 7.7pp | 2.51x |
Methodology Note: 2024 uses binary GE rate (Employed vs Not Working Yet) instead of 5-category classification used in 2020-2023. This may affect direct comparability but the trend remains consistent.
Public universities and polytechnics show significantly better employment outcomes than private institutions. Note: 2024 uses different methodology.
Best Performing
1.2%
Polytechnics
Highest Unemployment
13.7%
Vocational
| Institution | 2020 | 2021 | 2022 | 2023 | 2024⚠ |
|---|---|---|---|---|---|
| Public Unis | 13.3% | 11.8% | 7.2% | 7.0% | 5.1% |
| Private HEIs | 20.6% | 19.7% | 14.5% | 17.0% | 12.8% |
| Polytechnics | 8.6% | 6.3% | 3.8% | 2.9% | 1.2% |
| Comm. Colleges | 5.8% | 5.2% | 3.2% | 2.0% | 1.2% |
| Vocational | 18.4% | 18.5% | 15.1% | 18.6% | 13.7% |
Policy Insight: Public universities consistently produce graduates with better employment outcomes than private HEIs. Polytechnics and community colleges show excellent performance with under 2% unemployment in 2024, suggesting strong industry alignment.
Geographic Disparities
Sabah (11.0%), Kelantan (10.9%), and Sarawak (9.8%) show persistently high graduate unemployment, while Putrajaya (4.7%) and Labuan (6.0%) outperform the national average of 7.5% (2024).
Field Performance
Engineering graduates excel with only 5.1% unemployment (2024), while Education (10.2%), Services (10.1%), and Agriculture (10.0%) face structural challenges.
Public vs Private Gap
Private HEI graduates are 2.5x more likely to be unemployed than public university graduates, with a persistent gap of 7-10 percentage points across all years (2020-2024).
Institution Type Impact
Polytechnics (1.2%) and Community Colleges (1.2%) show exceptional employment outcomes, suggesting strong industry-aligned training programs compared to degree-granting institutions.
This interactive dashboard visualizes Malaysia's education and employment outcomes using official data from the Department of Statistics Malaysia (DOSM). The dashboard is designed to support data-driven decision-making around sustainable well-being initiatives.
Universiti Teknologi MARA
Data Integration: This dashboard integrates the Malaysian Well-Being Index (MyWI) 2010-2024 with structured CSV data from OpenDOSM and extracted PDF data from Ministry of Higher Education (Graduate Tracer Studies 2023-2024, Graduates Statistics 2020-2024). The integration provides a complete view of the education → employment → well-being connection across all Malaysian states.
Note on Wilayah Persekutuan: MyWI data combines Kuala Lumpur, Putrajaya, and Labuan into one "Wilayah Persekutuan" entity. The same MyWI values are applied to all three federal territories on this dashboard.