Background: It has been proposed that positive smoker identity may be an important factor undermining smoking cessation but very little research exists on this. This study tested the hypothesis that a simple measure of positive smoker identity would predict quit attempts over and above other known predictors in a population sample. More tentatively it explored whether this measure would also predict quit success. Methods: A representative sample of adult smokers in England (n=9456) was included at baseline and 2099 were followed-up at six months. Demographic and smoking characteristics, a single item measure of positive smoker identity (endorsing the statement: 'I like being a smoker'), measures of smoking-related attitudes, quit attempts and quit success were included. Results: A total of 18.3% (95% CI = 17.5-19.2) of smokers reported a positive smoker identity. Adjusting for all other predictors, those with a positive smoker identity were more likely to be older (p < 0.001), male (p = 0.013), more nicotine dependent (p < 0.001), have lower motivation to stop (p < 0.001), have not made a quit attempt in the past year (p = 0.025), enjoy smoking (p < 0.001), and consider themselves to be addicted (p <0.001). Having a positive smoker identity independently predicted failure to make a quit attempt at six months (p = 0.007). The independent association with quit success was similar in magnitude but did not reach statistical significance (p = 0.053). Conclusions: Only a minority of smokers in England have a positive smoker identity. However, where it is present it may be an important barrier to quitting smoking and merits further study. (C) 2013 Elsevier Ireland Ltd. All rights reserved.