Survey targets

It is important to ensure that response rates do not fall significantly over time, especially if they are likely to reduce the usability of survey results (e.g. through the generation of less precise estimates for smaller groups of population). Target response rates could therefore be one possible measure of survey performance while another would be the evaluation of data representativeness.

Graduate Outcomes response rate targets

ollowing the decision to discontinue the telephone data collection from non-EU overseas graduates1 the response rate in this group has decreased dramatically. Therefore, HESA in conversation with the Office for Students and other statutory customers decided to remove the non-EU domiciled response rate target from year 5 of the survey onwards. There is currently work being undertaken to develop a new target for the survey that considers representativeness of the data at a provider level as opposed to response rate. Current survey targets are as follows: 

UK domiciled full-time: 60%
UK domiciled part-time: 60%
Research funded: 65%
EU domiciled: 45%

These targets are mainly applicable for the entire set of survey results across a year but are also used to monitor performance of individual cohorts. This is to acknowledge the diverse composition of different cohorts in terms of graduate characteristics and how it impacts the response rates that can be realistically achieved.

The ‘headline’ response rates for the Graduate Outcomes survey are defined as:

• Numerator: Count of records with a valid response to a minimum[2] (pre-determined) set of core questions (classified with a status of ‘survey completed’).
• Denominator: Count of all records in the target population[3]

In addition to responses classified as ‘survey completed’, a status of ‘partially completed’ has been assigned where some of the core questions are missing but the first two questions have been answered. Although partially completed responses do not contribute to the ‘headline’ response rates, these are used alongside ‘survey completed’ responses in statistical outputs and form part of the base population for this purpose. Data from such responses will appear in published statistics with unknown values for questions that were not answered.

It should be noted that achieving a high response rate is not sufficient in itself to ensure good quality data has been collected. In later parts of this methodology statement, we will be discussing the other checks and processes we have undertaken in assessing and ensuring the quality of the data

Previous: Sampling frame     Next: Survey questions