There’s a fundamental truth which underpins nearly everything we do in quantitative market research and that is: numbers count. Sounds glib, but think about it. Society is so diverse and mobile that even very well-defined user groups or geographical areas can present sampling issues if the initial survey planning is not undertaken rigorously. An often-overlooked element of this planning is that of non-response and the more obvious sample bias issue.

There are formulae relating to rate of contact and participation we apply during the very early stages of planning and these give us a working base from which to anticipate any shortfalls in coverage. These impinge primarily on the interviewer numbers and duration of the sampling exercise. But there are other measures which can impact quite dramatically on response rates and the level of engagement of the respondent.

The time of day makes a huge difference in the availability of all types of people and we can programme in a variety of at-home patterns to cover off the group or groups who make up the sample. For instance, looking for a sample representative of the population we’ll plan a schedule of interviewing to cover as wide a day part as possible. Continuous sample monitoring further allows these timings to be geared to avoid sample bias.

It’s also important to remember that an interview is a conversation. So, whilst it’s the answer to the question we’re focused on, the introduction to it forms the beginning of the conversation and it’s here we either engage with, or lose the respondent. By revealing the sponsorship of the survey during the introduction participation rates can be increased. If that sponsor is a government agency, respondents feel almost as if participation is an extension of the democratic process and are more likely to engage with the interviewer. That is equally true when the purpose of the survey is succinctly put to the respondent.

Effective solutions such as these are part and parcel of the everyday process at NEMS. On the few occasions when a bias has been unavoidable, our expertise in post-survey adjustments through statistical weightings of the data ensures robust findings.

For more information on this or anything feel free to get in touch, contact details are on our website