It may not be the most logical thing to comprehend but surveys are a vital cog in the machine of industry. They give companies accurate feedback as to where they and their products stand in the market from the people who they most need that information. This information has taken on even greater importance over the last few years with the advent of mobile survey apps, also known as CAPI software (Computer Assisted Personal Interviewing). These apps have overcome many of the problems associated with traditional forms of survey resulting in the collection of accurate and useful data.
However, even the best survey technique and the best survey app will come asunder if the survey is not properly structured. As a survey is comprised of questions, it is vital that the survey questions will result in useful responses.
Following are five types of questions to ask followed by five types of questions not to ask when conducting a survey to ensure optimal survey results.
Questions to Ask
1. Questions that require short answers – let’s face it, even the most accommodating interviewee has better things to do with their time than answer survey questions, so don’t abuse their generosity and make the questions lengthy and complex.
2. Questions that provide relevant information – every question must count, so every question must result in the surveyor receiving actionable data that can be easily collated and analyzed by the company to best suit their needs.
3. Questions that provide information you can’t receive elsewhere – in these days where so much information is so freely available, surveys should be used to access information that cannot be readily accessed. You only get one shot at surveying someone, so make it count as the freely available data will always be easy to access.
4. Questions that are properly worded – the language used in structuring a question and the actual structure of the question are of vital importance in attaining the required information. If words used are too complex or a question is not clearly worded, the responses will be inaccurate.
5. Questions that are balanced – this differs from point 4 above in that it refers to the balance of responses offered. For example you want an even balance of positive or negative response options to ensure the accuracy of responses.
Questions not to Ask
It would be easiest to say that the questions not to ask are the opposite of those that you should ask – and you would be correct; however that would be cheating. So here are questions not to ask in a survey:
1. Questions that are too personal – just because people are good enough to answer a few questions, it doesn’t mean that you own them. If you start to ask questions about religion and detailed financial information, many respondents will refuse to answer not only those questions, but all questions – and rightly so.
2. Questions that don’t cover all possibilities – if, after showing the interviewee all possible responses, the interviewee asks, “Well how about this?” then you know there is something wrong with the question and the answers that must be promptly addressed.
3. Questions that don’t offer an option not to answer – a survey should be fun, not an ordeal. If the interviewee doesn’t want to answer a question for any particular reason, it is their prerogative and they should not have to give the interviewer any justification for their decision.
4. Questions that require a direct response – such questions mitigate the chances of the interviewee providing answers that don’t answer the question being asked. It is important to structure questions to avoid this occurring.
5. Questions that use leading language – incorrect use of a word or using wording that is too strong will result in a self-fulfilling response to the question, so keep the wording of the questions objective.
As well as generating invaluable data for the business that commissioned a survey, a well constructed and executed survey has the ability to strengthen the bond between client and company.
And in today’s world where information is disseminated at lightning speed, this is possibly worth more than the cost of the survey itself and the data received.