OBJECTIVE: To identify factors associated with intention to attend a hypothetical eye health test and provide an evidence base for developing an intervention to maximise attendance, for use in studies evaluating glaucoma screening programmes. DESIGN: Theory-based cross-sectional survey, based on an extended Theory of Planned Behaviour (TPB) and the Common Sense Self-Regulation Model, conducted in June 2010. PARTICIPANTS: General population including oversampling from low socioeconomic areas. SETTING: Aberdeenshire and the London Boroughs of Lewisham and Southwark, UK. RESULTS: From 867 questionnaires posted, 327 completed questionnaires were returned (38%). In hierarchical regression analysis, the three theoretical predictors in the TPB (Attitude, Subjective norm and Perceived Behavioural Control) accounted for two-thirds of the variance in intention scores (adjusted R(2)=0.65). All three predictors contributed significantly to prediction. Adding 'Anticipated regret' as a factor in the TPB model resulted in a significant increase in prediction (adjusted R(2)=0.74). In the Common Sense Self-Regulation Model, only illness representations about the personal consequences of glaucoma (How much do you think glaucoma would affect your life?) and illness concern (How concerned are you about getting glaucoma?) significantly predicted. The final model explained 75% of the variance in intention scores, with ethnicity significantly contributing to prediction. CONCLUSIONS: In this population-based sample (including over-representation of lower socioeconomic groupings), the main predictors of intention to attend a hypothetical eye health test were Attitude, Perceived control over attendance, Anticipated regret if did not attend and black ethnicity. This evidence informs the design of a behavioural intervention with intervention components targeting low intentions and predicted to influence health-related behaviours.