The OECD's PIAAC Cycle 2 (2023) represents a conceptual evolution in the measurement of adult competencies (OECD, 2021).¹ The first cycle of PIAAC (2012) assessed "Problem Solving in Technology-Rich Environments" (PS-TRE), which focused on assessing individuals’ ability to use “digital technology, communication tools and networks to evaluate information, communicate with others and perform practical tasks”.²

‍

This paper replicates and expands upon the seminal work of Hanushek et al. (2015), which used PIAAC Cycle 1 data to demonstrate robust wage returns to cognitive skills across the globe.³

‍

APS has three features: 

‍

1. capacity to flexibly and dynamically adapt their problem-solving strategies to a dynamically changing environment 
2. ability to identify and select among a range of available physical, social and digital resources.
3. monitoring and reflecting on progress in solving problems, through metacognitive processes.⁴

‍

^{1. ibid}

^{2. PIAAC Expert Group on Problem Solving in Technology-Rich Environments (2009). PIAAC Problem Solving in Technology-Rich Environments: A Conceptual Framework. OECD Education Working Papers, No. 36. Paris: OECD Publishing.}

^{3. Hanushek, E. A., Schwerdt, G., Wiederhold, S., & Woessmann, L. (2015). Returns to skills around the world: Evidence from PIAAC. European Economic Review 73, pp. 103-130}

^{4. OECD (2024), Do Adults Have the Skills They Need to Thrive in a Changing World?: Survey of Adult Skills 2023, OECD Skills Studies, OECD Publishing, Paris}

A central challenge in human capital policy is bridging the gap between high-level measurement and practical intervention. While PIAAC provides a robust, cross-national measure of Adaptive Problem Solving,⁵ it does not, by design, offer a pedagogical framework for developing it.

‍

This is the role of the Skills Builder Universal Framework. It provides the necessary translation mechanism, demonstrating how the construct measured by PIAAC can be systematically taught and acquired. The APS construct, defined by its focus on "dynamic situations" and "metacognitive processes"^{6 7}, aligns directly with steps of the Framework’s Problem Solving skill.

^{5. OECD (2021). The Assessment Frameworks for Cycle 2 of the Programme for the International Assessment of Adult Competencies. OECD Skills Studies. Paris: OECD Publishing}

^{6. OECD (2021). The Assessment Frameworks for Cycle 2 of the Programme for the International Assessment of Adult Competencies. OECD Skills Studies. Paris: OECD Publishing}

^{7. OECD (2024). Survey of Adult Skills – Reader’s Companion. OECD Skills Studies. Paris: OECD Publishing}

• Previous installments in the Essential Skills Tracker series found a wage premium of between 8.9% and 14.6% associated with moving from the lower to upper quartile of essential skill level, including problem solving. 
• Following the methodology of Hanushek (2015)⁸, we used a Mincer-style earnings regression to model the natural log of wages (W) as a function of years of schooling (S), potential experience (X), and controls (Z)⁹:

‍

• Our models showed that higher levels of APS resulted in increased earnings across all countries analysed. A one standard deviation increase in APS proficiency is associated with earning increases ranging from 3.7% (Poland) to 17.3% (Chile). In the UK, this proficiency is associated with a 13.9% increase in wages.

• APS had the greatest impact on prime-age workers in most of the countries explored in this analysis

‍

^{8. Hanushek, E. A., Schwerdt, G., Wiederhold, S., & Woessmann, L. (2015). Returns to skills around the world: Evidence from PIAAC. European Economic Review 73}

^{9. Jacob A. Mincer (1974), "The Human Capital Earnings Function," in Schooling, Experience, and Earnings (NBER), pp 83–96}

Previous instalments in the Essential Skills Tracker series found the existence of a persistent “Skills Trap”, where individuals from less advantaged backgrounds face a cycle of restricted opportunities leading to worse outcomes.

• The PIAAC data supports previous findings that essential skills can help individuals escape the Skills Trap and boost social mobility. 
• Both an individual’s formal education and their parental background remain strong predictors of their overall APS proficiency. However, the statistical interaction between these two variables is not significant. 
• This lack of interaction means that privilege does not unfairly multiply the value of learning: the economic benefits of developing problem solving skills are consistent regardless of a person's starting point
• Indicating that while social advantage may currently influence access to skill development, the skills themselves, once mastered, act as a powerful mechanism for escaping the skills trap and fostering genuine social mobility.

The introduction of Adaptive Problem Solving as a replacement for Problem Solving In Technology Rich Environments, and the links between higher levels of this skill and income are reflective of a global shift towards more human-centered skills.

‍

Previous Essential Skills Tracker installments found links between Essential Skills, such as Problem Solving, and an increase in wages, wellbeing and reduction in unemployability within the UK¹⁰. PIAAC Cycle 2 data provides a platform to explore whether these correlations persist globally through the lens of Adaptive Problem Solving (APS). Although this construct is different from problem solving as defined by the widely adopted Universal Framework, mapping using Language models highlights the overlap between the Universal Framework and APS. 

‍

While the PIAAC data shows a clear relationship between Adaptive Problem Solving and economic and social outcomes, the data reveals the limitations of this single assessment construct. In particular, there is high multi-colinearity between literacy, numeracy and Adaptive Problem Solving in Cycle 2 of PIAAC. While each independently predicts other outcomes, when adding the other terms to regression models, there is little or even negative improvement to their predictive value. 

‍

The correlation scores between the three skills sits at approximately 0.88. During Cycle 1, the correlation between PR-TRE and literacy and numeracy was lower at 0.79 and 0.73, respectively¹¹. While this shows a strong relationship between these competencies, it is important to note that this level of correlation is consistent with the relationship between literacy and numeracy themselves (0.89). Literacy and numeracy measure different aspects, and given that the correlations are consistent, it therefore does not suggest that APS is assessing the same constructs as literacy and numeracy, nor does it indicate that APS is merely a proxy for general intelligence.

‍

The high correlation between the constructs does, however, suggest that there is a need for differentiated measures of essential skills. The Universal Framework provides this granularity by breaking this into eight essential skills, each with 16 steps. The APS was designed as an objective assessment, in contrast, the purpose of the Universal Framework is as a teaching, learning or workforce planning tool. We therefore do not seek to draw any direct comparisons between the two constructs. Instead, this paper analyses some of the differences and their implications.

‍

‍

^{10. Seymour, W. and Craig, R. (2023). Essential Skills Tracker 2023. London: Skills Builder Partnership}

^{11. Hanushek, E. A., Schwerdt, G., Wiederhold, S., & Woessmann, L. (2015). Returns to skills around the world: Evidence from PIAAC. European Economic Review 73}