A study compared the GCSE benchmarks of all local education authorities (LEAs) in England and Wales. The impetus was that a comparison on almost any performance indicator of initial education flattered England, and comparisons led to the conclusion that the education system in Wales was failing. The performance measure was the GCSE benchmark of the percentage of the relevant age cohort gaining five or more grades A-C; the 1994 figure was lower for Wales. Independent variables used to characterize LEAs were social class, free school meals, population, area, and population density. Correlations between data were calculated using Pearson's r; a multiple linear regression analysis was conducted. All independent variables correlated significantly with the GCSE benchmark, with the percentage taking free school meals accounting for two-thirds of the variance in results between LEAs. No evidence was found of poor school performance in Wales. The initial schooling and background characteristics of each participant were used to create a series of logistic regression models to analyze post-compulsory learning trajectories. Nine predictor variables at age 15 or less were significant in predicting examination results at 16; interactions between variables were more significant in many cases. The type of secondary school attended was the major determinant of qualifications at age 16 until fairly recently. (Contains a 53-item bibliography.) (YLB)