How to Conduct a Survey: For Beginners

Surveys are a great tool for companies looking to gauge customer feedback, analyze market trends and measure success in the current market. However, when planning a survey, it is important to consider the many factors that are necessary for the collected data to be meaningful. Before conducting a survey ask yourself these questions:

1)How many questions should be included in the survey?

2) How can you entice people to participate in the survey (what is the benefit to the survey participant for completing the survey)?

3) How many responses are necessary to get an accurate sampling of your target audiences’ opinions?

A recent post on the SurveyMonkey blog explains that the more questions there are on the survey, the lower the response rate will be. A short survey usually consists of no more than 20 questions. On the other hand, some  people are willing to answer longer surveys. Keep in mind, that for a longer survey, it is smarter to have multiple choice questions instead of open-ended – this way participants are more likely to complete it. However, it all depends on the value of the questions being asked.

Do not eliminate valuable questions in order to make the survey shorter. Instead, write out all the questions you think are necessary to achieve the data you want and then read over the questions, making sure each one is  mission critical. For instance, one section  of the survey that can often be condensed is the demographic section.

As for how to garner participation for your survey, try offering respondents a reward. The reward can be something as simple as a discount, a promotional item or a free consultation. If you do not offer a reward, explain to your audience the benefits they will receive from completing the survey. For example, perhaps responses from the survey will be used to determine employee benefits. Letting your audience know the outcome of the survey can help increase the participation rate, especially if the results will have a direct impact on them.

Lastly, be aware that depending on what you are surveying, the number of responses necessary for an accurate sampling may change. However, according to John Creswell, author of Research Design, 100 is the lowest number of responses you should probably aim to collect – validity (logical truth where the conclusion matches the premises) and significance (the results are unlikely to occur by chance) cannot be established with a response of fewer than 100 participants.

In order to estimate the response rate of your survey, always run a test survey. In other words, take your survey and distribute it to a randomly selected group of people (they will take the place/represent your target audience in the test survey). After a selected period of time, collect your responses and analyze the success of your response rate. If you are satisfied with the response rate, distribute your real survey to your target audience. Note – the response rate of your test survey may not be an accurate prediction of the actual response rate you will receive. However, a test survey is still beneficial to conduct.

For more resources on survey tips, visit the FAQ page on SurveyMonkey. Good luck!

Advertisements
%d bloggers like this: