Over the years our researchers have built up a vast store of information and evidence on improving the quality of surveys and response rates.

Defining the sample

Perhaps the most important method for ensuring great response rates is having a well defined sample in the first place. If you are planning an anonymous census-type of survey, where you send out a request to take part in your survey to a large but ill-defined sample, the chances are that a good number of the people you send the questionnaire out to will have incorrect or unreachable addresses or no longer work where you think they do.

Having a measure or estimate on these 'unreachable respondents' can sometimes drastically improve response rates. For example: a email survey is sent to 1000 employee email addresses provided by a local firm, but over 70 were returned as either incorrect, undeliverable or 'not available'. This can make a sizable difference to your response rates.

Defining your sample has further advantages. If you know, for example, how many managers you have in an organisation, then based on your response rates you can calculate the percentage of managers that responded. If the survey is not anonymous you can then chase up the missing managers - or at least ask them why they didn't respond, and if the survey is anonymous you will at least be able to weight your manager category responses correctly compared to other categories in the sample when calculating means and response percentages.

You can find out more about specific response rate issues by following the links below.

How to improve response rates for postal surveys

How should questionnaires be sent back after surveys