Internal reliability can be checked using repeat questions in the survey, worded slightly differently. This checks for the consistency of the questionnaire within itself as respondents who give two very different impressions on repeat questions can be classed as unreliable.