Imagine a watch that is said to be reliable. It records time at a consistent rate, it is calibrated against authoritative time keeping apparatus, it will work in similar ways wherever it is used and will be read in the same way by whoever looks at it - give or take a margin of error. This intuitive sense of reliability underpins the use of the term in your research. For example a reliable measure is a consistent one: a question might be reliable if it is clear enough to be interpreted the same way by different people; an application of a coding scheme is reliable if different coders use it in the same way; an observation might be reliable if noted by more than one person. Reliability is used in relation to different contexts but common is the sense of stable repeated measurement.

However, the question of reliability can become less clear cut in relation to qualitative data, particularly interview data For example imagine a teacher who describes him or herself as very favourably disposed to using ICT. The same teachers may be seen by some pupils and colleagues as quite resistant to using ICT. There are various ways in which you can test the trustworthiness of the teacher’s account for example by going back and seeking clarification of the meaning the manager put on 'disposed' and checking if can certain events did take place as reported. However unless there is a clear case of dissembling you are dealing with varied interpretations, not reliable versus unreliable reporting, almost certainly influenced by the role of the teachers and the context in which he or she works. It is the nature of the research that there is not a single consistent or reliable account of events.