We use linked data for 1,460 workplaces and 19,853 employees from the Workplace Employee Relations Survey 1998 to analyse the incidence and duration of employee training in Britain. We find training to be positively associated with having a recognised vocational qualification and current union membership. Whilst being non-white, shorter current job tenure, and part-time or fixed-term employment statuses are all associated with less training. Furthermore, in line with recent non-competitive training models, higher levels of wage compression (measured in absolute or relative terms) are positively related to training.