7.17 Existing routine monitoring data has the potential to fulfil some or, on occasion, all the data needs for planned evaluation. If this is the case then the policy makers and analysts have the advantage of reduced costs, reduced intrusion upon operations and potentially a longer historical time frame in which to place observed changes in context. There are, however, clear limitations to these data in terms of what questions can be answered, and the data require substantial processing in order to be applicable to impact evaluations. Indeed this type of data may be of lower or higher quality than those collected expressly for research purposes, and are not independent; this should be considered when deciding whether or not to use it.
7.18 For example, information about exact dates of joining or leaving a programme are likely to be recorded accurately on monitoring systems, whereas the individual participant, if asked in interview, is unlikely to have perfect memory. But data about disability, for example, is likely to be more reliable when collected as part of a research exercise than through monitoring systems. For this reason it can be useful to collect data using a number of methods, such as monitoring data and bespoke surveys, this is known as triangulation and is covered in more detail in Chapter 8.