Designing your first survey / questionnaire is challenging because it is not that easy to write it effectively for the first time around.
You either write too many questions trying to get lots of data and information from respondents. Or, you write too few and then you end up with very limited results. Your questions are too wordy, or too difficult, or may contain mistakes.
A lot can go wrong really with designing your first research tools.
3 important issues with designing a survey / questionnaire
The three important issues for market researchers to be aware of when conducting research surveys / questionnaires include:
- Who to ask in a survey? To determine the sample size.
- What to ask in a survey? To construct unbiased and unambiguous survey / questionnaire necessary to obtain valuable results.
- How to ask in a survey? To decide whether the survey / questionnaire will be self-completed and returned or filled in by an interviewer during a face-to-face session with the respondent?
1. Who to ask in a survey / questionnaire?
You need to identify the population and choose the sample. The population is simply all members of the group that you are interested in.
In most cases, it is impossible or too expensive for a business to survey all members of the target market. Therefore, the researchers need to select a ‘sample’ from the entire population which will closely reflect the characteristics of the survey population. The sample is a sub-set of the population that is chosen for the market research purposes.
The better the sample, the more accurate the survey / questionnaire is going to be. A key issue in choosing the sample relates to whether the members you have chosen are representative of the entire population.
To determine the sample size, it is usual to work back from how many responses are required for the completed analysis. Successfully completed surveys / questionnaires are called responses. One rule of thumb is to look for around 20 to 30 responses in each of the major categories of the sample. For instance, if a key aspect of the research is to compare male and female choices, then you should look for around 30 men and 30 women to be included in your sample.
This number then needs modifying by the anticipated response rate to determine the target sample size. It is quite common for survey response rates to be around 20%, which means sending out five times as many questionnaires as you want returning. For the above example, this would mean sending 300 surveys / questionnaires to receive 60 responses.
2. What to ask in the survey / questionnaire?
You need to make the research objective very clear and specific, so all of the questions in the survey / questionnaire can be focused on answering the research question. The research objective will be met when you gather enough data and information to solve the research problem.
1. Pay attention to what to ask.
Ask different types of questions. The mix of quantitative data and qualitative information will help to better address the research question.
Include both closed-ended questions and open-ended questions. Closed-ended questions help to gather quantitative data such as facts and figures which is easier to process and present on charts. They usually require the respondent to choose from a range of options. These kinds of questions make it much easier and quicker to complete a survey / questionnaire. Open-ended questions help to gather qualitative information such as opinions and attitudes which address the problem in more depth. They can also provide reasons why respondents behave in a certain way and ideas or suggestions that enable the business to make better decisions.
Include some questions that will allow a classification of results by gender, income, area lived in, occupation and so on.
Make sure that the survey / questionnaire has smooth flow and questions follow each other in a logical sequence. Nobody likes reading something that does not make any sense.
2. Pay attention to what not to ask.
Questions that do not help to answer the research question should not be included in the survey. Otherwise, irrelevant questions will make the survey too long.
Do not ask personal questions such as age or income levels unless absolutely necessary. Respondents will not be happy and will most likely not answer these type of questions. They may also refuse to finish the survey / questionnaire. One way around this issue is to ask people to indicate which of the following income levels they belong to: USD$10,000–USD$20,000 or USD$20,001–USD$30,000 and so on
Asking only open-ended questions to gather their opinions about the product is not a good idea because the answers will be so varied in length and content that the results will be impossible to put together and analyze. A better option is to give respondents a list of options to choose from or rate such as price, image, quality, reliability, safety, packaging, availability, scent/flavor, and so on.
Market researchers must avoid any sort of bias in order to collect meaningful and useful data. Absolutely avoid asking questions that will point people to give only one particular answer. The wording of questions should be clear and easy to understand not to instill any ambiguity among respondents.
3. Pay attention to writing clarity and visual design.
Due to the potential benefits of using surveys and the potentially high costs, effective survey design is of real importance. Design matters!
Writ clear and unambiguous questions. Use language that will be readily understood be everybody regardless of age, education levels and occupation.
To minimize unrepresentative answers, avoid jargon, slang or complicated technical language that may cause respondents not understand the questions.
It is always recommended to try and test you survey / questionnaire first through a trial with a small group of people in order to check the quality of questions as well as accuracy and usefulness of the results. This can help to identify any mistakes or omissions in the initial survey, so that the results from the final survey will be exactly what the researcher needs.
3. How to ask in the survey / questionnaire?
There are two main types of surveys / questionnaires including self-completed surveys and personal surveys.
A. Self-completed survey / questionnaire
Respondents usually complete the survey / questionnaire by themselves on the stop such as on the street, in the office, in the shopping mall, etc. Self-completed surveys / questionnaires can also be sent by post, conducted by phone or filled in online. Unfortunately, surveying people has quite low response rate, hence self-completed surveys / questionnaires may bring limited research outputs.
B. Direct interview
Personal interviews with the respondents are conducted by an interviewer usually on the street or in the respondents’ homes. A range of follow-up questions can be asked to clarify things and details regarding each question can be explained. However, direct interviews can be very costly especially when the sample size is significant.
In short, the biggest benefit of using surveys / questionnaires is the ability to generate both quantitative data and qualitative information depending on the research needs.
When properly designed, this method of market research is fairly straightforward, but can be costly and time-consuming as using a large sample is necessary to get statistically representative findings.
If not properly designed, the results will be biased and useless because of dishonest answers.
