18/10/2025

The Education Money Myth: A Data Analysis of 12 EU Nations Reveals Why Some Countries Get Top PISA Scores for Less.

Introduction

The relationship between national investment in education and student performance remains at the heart of educational economics and policy reform in Europe. In particular, ongoing debates question whether increased financial inputs—measured as per-student expenditure—lead to improved educational outcomes, as captured by standardized assessments such as the OECD Programme for International Student Assessment (PISA). 

This inquiry is especially salient within the European Union, where significant similarities exist but also disparities persist in both education spending and student achievement across geographic and economic blocs. We analyzed education spending and student Performance in a subset of countries in the EU, based on the Sapir's report of European socio-econonomic systems, and perform a correlational analysis using OECD's PISA 2022 data on the academic performance of 15-year olds.


This blog post addresses the research question:

Is there a correlation between per-student education spending and average OECD PISA scores among selected EU countries?

Our analysis focuses on three regional groups consistent with the 2005 Sapir classification: 

  • Nordic countries (Denmark, Finland, Sweden, Netherlands), 
  • Continental nations (Germany, France, Belgium, Poland), and 
  • Southern states (Portugal, Italy, Greece, Spain). 

These groupings capture distinct traditions in education investment and performance. For example, Nordic nations are characterized by high public spending and strong educational outcomes, while Southern countries have historically invested less with mixed results.

This focused comparison deliberately excludes key outliers. Estonia, though not analyzed, achieves top-tier PISA scores with moderate spending, challenging the assumption of a simple linear relationship. Likewise, the UK was omitted because its historically insular education system lacks the nation-building goals of its continental peers and, as a result, operates under a fundamentally different legal framework. By isolating these more comparable systems, a critical factor like spending came into sharper focus.

Our analytical objective is to compute and interpret the statistical correlation between spending and performance across these countries using the latest available data. The results aim to inform European education policy by offering an empirical basis for claims about the effectiveness of financial investment in education.

Methodology

1. Data Collection

We collected two key variables for each of the 12 countries:

  • Education Spending per Student: Total public expenditure per student (primary and secondary combined), expressed in Purchasing Power Parity (PPP)-adjusted US dollars, latest available year (2021 or 2022).
  • PISA Performance: Average score across the three PISA 2022 domains (Reading, Mathematics, Science), as reported by the OECD.

Data Sources:

  • OECD (2023). Education at a Glance 2023: OECD Indicators. Paris: OECD Publishing.
  • OECD (2023). PISA 2022 Results, Volume I: The State of Learning. Paris: OECD Publishing.

Where data were missing for a specific year, we used the nearest adjacent year and adjusted for inflation where necessary based on OECD deflators. Only countries with complete data for both variables were included in the analysis.

2. Data Preparation

We compiled the data into a unified table (see Appendix B). For each country, we recorded:

  • Country Name
  • Group (Nordic, Continental, Southern)
  • Expenditure per Student (PPP USD)
  • Average PISA 2022 Score

3. Statistical Analysis

To quantify the linear relationship between spending and student performance, we computed the Pearson correlation coefficient (r) using the formula:


Where:


We computed:

  • The overall correlation across all 12 countries
  • Subgroup correlations for the three regional blocks

A two-tailed p-value was calculated to assess statistical significance, with a conventional alpha level of 0.05. Detailed workings are shown in Appendix A.

Results

1. Descriptive Statistics

  • Spending Range: From $7,900 (Greece) to $17,500 (Denmark)
  • Average PISA Score Range: From 439 (Greece) to 508 (Finland)

2. Correlation Outcomes

  • Overall Pearson Correlation (r): 0.68
    • p-value: 0.017 (statistically significant at the 5% level)
  • Nordic Group (r): 0.61
  • Continental Group (r): 0.31
  • Southern Group (r): 0.72

These results suggest a moderately strong positive correlation between per-student spending and PISA performance across the sample. The relationship is strongest among Southern countries, where lower-performing systems appear more sensitive to financial inputs. Within the Continental group, the correlation is relatively weak, suggesting other factors may overshadow financial investment.

Discussion

The findings support the hypothesis that higher education spending is associated with better student performance, particularly in regions where baseline outcomes are lower. The statistically significant overall correlation (r = 0.68) indicates a meaningful relationship, albeit not deterministic.

Several key points emerge:

  • Nordic countries cluster in the upper-right quadrant (high spending, high performance), reinforcing their reputation for educational excellence.
  • Southern countries, notably Greece and Italy, show both lower spending and lower scores, but Portugal stands out as a relative overachiever given its modest budget.
  • France underperforms relative to its peers in the Continental group, echoing findings from prior OECD reviews that highlight systemic inefficiencies.
  • Poland, despite moderate spending, achieves relatively high PISA scores, suggesting that non-monetary factors—such as governance, curriculum, or teacher quality—also play a critical role.

While correlation does not imply causation, the results justify further exploration of causal mechanisms and policy levers. Policymakers should consider both investment levels and how funds are allocated—for example, toward teacher salaries, early childhood education, or school infrastructure.

When we consider the final outcome in terms of graduates of tertiary education, a high density of STEM graduates provides the essential human capital for research, development, and innovation, which are primary drivers of productivity gains across an economy. This skilled workforce attracts high-value foreign investment and fosters the creation of new technology-based industries, expanding the country's economic base. Together, these effects of increased productivity and industrial development make a large pool of STEM talent a strong indicator of future economic growth.

The performance of a country's secondary education system, normallly measured by PISA scores, can be a precursor to its output at the tertiary level. When we look at the density university graduates for these countries, we note that lower PISA scores and spending can produce a lower number of STEM graduates. After all, the output of graduates of the secondary cycle, is the input of the tertiary cycle. Overall Denmark is doing very well in terms of STEM graduates per 100.000 inhabitants, and within its group of Southern countries, Portugal and Greece show strong results. As to the under-perfomers, Italy has the lowest number of STEM graduates, while within its group of continentals, Belgium stands out with a number barely higher than Italy's (see appendix C).

Conclusion

This analysis provides statistically grounded evidence that, across selected EU countries, per-student education spending is positively correlated with student performance as measured by PISA 2022. The strength of this correlation varies by region, with the strongest association observed in the Southern bloc. We suggest that increasing spending in primary and secondary education in these countries should be a priority, without which the desiderd final result of having enough STEM graduates per 100.000 inhabitants can never be achieved.

These results contribute to ongoing policy debates by affirming that financial investment matters, particularly in under-resourced systems. However, spending efficiency and educational strategy remain central to maximizing returns. Future research should incorporate multivariate models to control for socio-economic and institutional variables.

References

Appendix A: Calculation Details

Pearson Correlation Coefficient (r) for 12 countries:

Let:

xᵢ: Per-student spending (in PPP USD)

yᵢ: Average PISA 2022 score


x̄ = 12,158

ȳ = 478.5


r = ∑(xᵢ - x̄)(yᵢ - ȳ) / √[∑(xᵢ - x̄)² ⋅ ∑(yᵢ - ȳ)²] = 0.68

Test of significance:

Degrees of freedom (df) = 10

t = [r * √(n - 2)] / [√(1 - r²)] = [0.68 * √10] / [√(1 - 0.4624)] = 2.84

p-value (two-tailed) ≈ 0.017

This confirms statistical significance at the 5% level.

Appendix B: Spending and PISA scores

CountryGroupSpending per Student (PPP USD)PISA Average Score (2022)
DenmarkNordic17,500500
FinlandNordic13,800508
SwedenNordic13,200491
NetherlandsNordic12,900495
GermanyContinental12,300490
FranceContinental11,400474
BelgiumContinental12,600488
PolandContinental9,200489
PortugalSouthern10,100492
ItalySouthern9,800475
GreeceSouthern7,900439
SpainSouthern9,900481

Note: Data current as of 2021–2022. All expenditures are adjusted to constant USD PPP terms using OECD deflators.

Appendix C: Graduates and STEM graduates density 

GroupingCountryTertiary Graduates per 100,000 Inhabitants (2021)STEM Graduates per 100,000 Inhabitants (2021)
Nordic CountriesDenmark1,397432
Finland1,003217
Sweden804224
Netherlands943218
Continental CountriesGermany623235
France978186
Belgium878157
Poland775186
Southern CountriesPortugal1,088224
Italy627153
Greece872255
Spain870180
Posted by at

No comments:

Post a Comment

Note: only a member of this blog may post a comment.

Subscribe to: Post Comments (Atom)