As part of a survey service developed to assess bullying in schools, anonymous questionnaires were given to over 6,000 pupils in 17 junior/middle and seven secondary schools in the Sheffield LEA. The results are analysed in terms of frequencies of being bullied, and bullying others; year differences; gender differences; types of bullying; where bullying occurs; whether teachers and parents are informed; and attitudes to bullying. Rates of reported bullying are disturbingly high; they vary with year, gender and school location, partly as a result of opportunities for bullying. With the addition of data from six other schools, it was found that school size, class size and ethnic mix were not linked with bullying. Social disadvantage is linked with bullying to a small extent, and schools with high bullying rates also tend to have pupils who dislike, or are alone at, playtime. Implications for intervention against bullying are briefly discussed.