Planning a Survey
Decide on the Survey Objectives
Without clear objectives the survey is likely to be ineffective. A clear set of objectives provides a firm foundation for the design phase and also sets clear expectations for the survey outcome. Objectives must also be communicated to the respondents so that they are aware of the purpose of the survey - failure to do this will result in poor response rates and inaccurate survey data.
Select the Survey Audience
Decide also on the survey audience - is the survey for all employees or just for a sample. Is it aimed at a specific group? Is the survey a one-off or will it be repeated on a regular basis? If the survey is going to be repeated then consider building up normative data - this could provide vital information on trends and can help identify early signs of problems within the organisation.
Confidentiality is also a key issue to address - will respondents be anonomous? If so how will confidentially be maintained?
Questionaire Design
Make the questions easy to understand. Respondents won't want to think too deeply when completing the survey so make sure your questions are straightforward and relate directly to your organisation. Phrase questions in a similar and consistent manner and use consistent answer schemes. Take care not to include too many 'free text' responses - these are difficult to analyse and can render results inconclusive. CorporatePulse questionnaires use a range of answer schemes (click to see example).
Don't ask too many questions - aim for a typical completion time (we recommend 20-30 minutes) and stick to it Research shows that short focused surveys have a much higher response rate.
Consider measuring importance as well as performance. Knowing how employees rate the company's performance in various areas is important, but in order to prioritise your action plan you need to know the importance of each of these areas.
Include demographic questions to provide additional analysis options. For example you want to ask how long the respondent has been with the company or what age or salary group they fall into.
We can use 'conditional branching' to control the path a respondent takes through the survey by making questions dependent upon one another. This allows us to design the questionnaire so that irrelevant questions are missed out.
Presentation of Results
Be clear from the start about how you want the results presented. Complex surveys can result in complex and often conflicting data and the results are often impossible to analyse or interpret accurately. Simpler surveys lead to simpler results and make analysis and interpretation much more successful - that's one of the reasons we recommend pulse type surveys. It's important to feedback results to respondent's too so at the very least a simplified report should be planned for respondents.
Considering comparing your results with Normative data either from external organisations or from data built up over a number of previous surveys within the business. Normative data, typically industry surveys conducted by magazines or research organizations, can be extremely useful in evaluating survey results but it does limit you to the 'standard' question sets used by these organisations.
|
|
| Try Our Demo Survey! |
Click here to see an example of a 'pulse' survey; fully customised for our fictitious company, "Global Telecom".
 |
|
|
