Students are often very disappointed when the questionnaires or surveys they send out as part of their course have a very low response rate, and this may even affect the validity of their research. It is very important, therefore, that questionnaires be well prepared and well constructed in order to minimize non-response.
General considerations
Keep the questionnaire as short as possible
Most people are busy. Certain companies and organisations may also receive many other similar questionnaires and other requests from students around the country, or even internationally. If your questionnaire is long and complicated, it will greatly lessen the chances of receiving a response.
Just stick to the essential or very useful information needed; discard any questions which relate to supplementary, secondary or unnecessary data. This means preparing well, and making sure that you know exactly what your aims are. Are you trying to identify attitudes, needs, behaviour or what?
Target the questionnaire carefully
You must make sure that your questionnaire is completely relevant to the person you’re asking to complete it. If they’re unable to answer the first few questions, then they’re very likely to give up. So do some research on who would be the most appropriate people to target within the organisation, and always address your request to them directly. It may be a good idea to send a preliminary email to ask if it would be OK to send the questionnaire, and if in fact it would be relevant to the person.
In addition, try to target people and organisations who you think are less likely to be bombarded by similar requests. Some sort of personal contact or relationship would of course be ideal; failing that, a phone call or even an email in advance may increase the chances of a response. Don’t just send your questionnaire to a general address.
Consider the question of anonymity
If your questionnaire contains sensitive or personal questions, you need to convince potential respondents that their answers will be confidential. If not, they will not respond.
Maybe offer something in return
Commercial questionnaires try to tempt people into replying by offering the chance to win holidays or money. You obviously can’t do this, but some sort of incentive is important to motivate people into responding. You could maybe offer to send the respondent a copy of the survey results, or make it clear in the introduction just how worthwhile your project is and that the respondent may benefit from it at some stage. It might just persuade them to answer.
Don’t make people pay for stamps or phone calls!
Use follow-up reminders
Your target respondent may have every intention of replying, but has got lots of other important things to do to, so a polite reminder (or two) may be necessary. This is probably best done by email or phone, and you might have more chance of a response on a Friday than a Monday. Remember, however, the question of anonymity: it may be necessary to use codes to be able to track who has replied or not, but you must make it clear that the link between codes and personal will not be used and will be destroyed.
One other possibility may be to have a shortened version which someone could quickly reply to by email or on the phone. It may be better than nothing.
Types of questions
Keep your language simple
Phrase your questions in simple, unambiguous language, which will be instantly understood. Don’t use any jargon or abbreviations which might be unknown to some of your target sample.
Check over your grammar and spelling very carefully, and if necessary get someone else to do this aswell. Mistakes look very unprofessional and give the impression that you are not very serious about what you’re doing. If you couldn’t be bothered, why should someone else be bothered to reply?
Don’t make it too formal or informal
It’s not an official document; just use “standard”, neutral vocabulary and grammar, being careful, however, not to use colloquial language.
Start with interesting questions
Just as a good introduction to an essay will grab the attention of the reader, so interesting, relevant initial questions will immediately involve the respondent in the questionnaire, and make it more likely that they will continue on through the whole thing. Save the more complicated ones for later. You could begin with one or two non-specialised personal questions, or maybe phrase questions in the third person rather than the first: How does your company feel about equal pay rather than How do you feel about equal pay?
Avoid leading questions
These are questions which presume a specific response. For example, Which train company do you use most often? presumes that the respondent travels by train. If they don’t, they can’t answer.
Limit, or avoid, the use of open-ended questions
Open-ended questions require an original, personal response to a question. For example, What do you consider to be the most important qualities of a Personnel Officer? Answers to such questions obviously take a lot more time than yes/no answers or rating scales. Definitely don’t start with questions like this; if you want to include you or two, put them at the end. And don’t leave too much space for the answer; if the respondent sees half a page of blank paper, they may feel that they have to fill it all, and consequently be discouraged from answering at all. Open-ended questions are also more difficult to analyse and categorize.
Use simple rating scales or lists of choices
If the respondent is faced with a long list of scales or choices, they may be put off. Maybe consider five as a maximum.
Put your questions in logical order
There should be a flow of questions; one question should logically follow the previous one. You could perhaps start with one or two general questions, and then become more specific.
No comments:
Post a Comment