The global labour market is confronted with a significant challenge of labour market mismatch, notably evident in the prevalence of overeducation. Historical and contemporary research into this phenomenon has been heavily biased towards developed economies, suggesting either that (a) overeducation is not an issue for developing countries, or that (b) it actually is a significant concern, but has been under-researched for a plethora of reasons including a lack of suitable data and perhaps a lack of focus given the emphasis on economic growth and development instead. This thesis resolves the latter, at least by looking at available Sri Lankan Labour Force Survey (LFS) data, a developing country with access to better data sources than most of its peers. In Sri Lanka specifically, and in the developing world generally, the augmentation of educational investment and attainment has exacerbated the imbalance between labour supply and demand, culminating in adverse labour market outcomes characterised by diminished wage returns, wage penalties, and wage disparities. This thesis explores the overeducation phenomenon in a developing country (Sri Lanka) by studying (1) alternative measurements of overeducation, (2) incidences and returns to overeducation, including a specific look at graduate overeducation, and (3) gender wage inequalities, and its relationship to overeducation. Four research objectives were developed to address the overeducation phenomena, which in turn are organised as four studies in the thesis. Despite almost 50 years of research into overeducation, there is a significant absence of a comprehensive synthesis regarding the alternative measures of overeducation. In Chapter 3; a systematic quantitative literature review addresses this gap by providing a wide-ranging overview of alternative measures of overeducation debated in the contemporary literature, thereby contributing novel insights into the methodological diversity and implications of different measurement approaches. Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines are used to synthesise a database using Covidence and Notion software. Relational analysis by Leximancer software identifies six major clusters within the segmented review that highlight the divergences in results due to the use of alternative measurements and clarify the sensitivity of results to specification to both the academic community and policymakers. Chapters 4 and 5 focus on the incidences and returns to overeducation. Specifically, Chapter 4 provides a comprehensive analysis of overeducation in Sri Lanka, comparing the incidences and returns to overeducation utilising alternative measurements. A pseudo-panel is created out of five continuous (2015 - 2019) cross-sectional data to control individual unobserved heterogeneity using fixed effect estimations in assessing returns to overeducation. Consistent with the stylised facts of overeducation, the study found significant pay penalties for overeducation and pay premia for undereducation vis-a-vis adequately matched peers with similar levels of education, irrespective of measurements used, though magnitude differences were evident. Moreover, the study found that overeducated males no longer suffer from a wage penalty whereas the pay penalties associated with overeducation only apply to females, again irrespective of measures utilised. Chapter 5 explores the impact of graduates' overeducation on wages across the wage distribution accounting for the individual unobserved heterogeneity attributed to ability, given the governmental focus on developing skills via the higher education sector in Sri Lanka and other developing countries. Overeducated graduates were distinguished from their non-overeducated counterparts using the Job Analysis (JA) measure. Unconditional Quantile Regression (UQR) approach along with standard Ordinary Least Square (OLS) regression employed to estimate the wage effect of overeducation across the entire wage distribution while stratifying by gender and field of study sub-samples. Results suggest a wage penalty associated with overeducation across the wage distribution, irrespective of sex and field of study. Both male and female graduates with lower abilities experience higher wage penalties, supporting the notion of a close relationship between low ability and overeducation. Conversely, higher-ability male graduates suffer lower wage penalties compared to higher-ability female counterparts prompting discussion on the existence of the glass ceiling effect. Chapter 6 contributes to the gender wage inequalities of overeducation. The study estimates the gender wage gap (GWG) in urban and non-urban areas in Sri Lanka in the presence of overeducation, using Blinder Oaxaca decomposition (BOD), Counterfactual Decomposition using Quantile Regression (CDQR), and Decomposition using Unconditional Quantile Regression (UDQR). Unsurprisingly, the chapter finds GWG in favour of males, especially in lower and higher quantiles of the wage distribution. The non-urban localities evince a higher GWG compared to urban areas consistently in all three methods. Decomposition results indicate that the GWG is mainly driven by discrimination i.e. by the unexplained factor in comparison to the explained factors. Indeed, the explained factors suggest that any existing wage premia should actually accrue to females rather than males, which is a unique finding. A priori, one would expect that explained factors like skills and work experience would already favour males in terms of wages, and that this would only widen with the presence of the discriminatory (i.e. unexplained portion) of the models. Instead, findings suggest a significant discrimination effect on wages against females, as the explained portion results in favour of females is completely flipped on its head. Overeducation contributed positively to the GWG in which overeducated males are penalised less in terms of wages compared to their female counterparts. Especially so in urban areas compared to the non-urban counterparts. Post-empirical studies, Chapter 7 provides a comprehensive overview and detailed empirical findings of the research where the thesis provides policy recommendations on each specific research chapter (i.e. chapters 4-6) to reduce the labour market mismatch from the supply side. Finally, this chapter also addresses the limitations encountered during the study and proposes directions for future research.