Writing Good Survey Questions – Examples, Types, And Best Practices

Surveys are a fantastic way of conducting research on pretty much any topic. They can be done on a small or large scale and reach thousands or even millions of people depending the purpose of the survey and its target audience.

Experienced business owners can take advantage of market research as a means of getting feedback on products and services that are currently available or coming in the future.  It can save a ton of money before the expensive manufacturing, marketing and/or launch process, feeling confident a market or product is ready for consumption with the right messaging. To ensure a business gets the right answers to their business questions, it is essential to understand how survey questions work and which type to use and when. 

Multiple Choice Questions

Multiple choice questions are the most frequently used type of survey question.

This is for many reasons including the fact that they are arguably the easiest for the person answering the questions to answer and that they are uncomplicated to produce.

These questions provide the researcher with quantitative data, this is a countable result. In other words the researcher will be able to review their results and tally up the amount of times that each box has been ticked.

This allows them to figure out the popularity and frequency of certain things and is often used when trying to establish patterns and trends. 

Example: On average how often do you visit the grocery store?

1. Less than weekly
2. Once a week 
3. Two to four times a week
4. More than four times a week

The example above would allow the researcher to put all of the collected data together in order to figure out how often the majority of people visit the grocery store. 

Rating Scales

Rating scales are where a respondent must use the scale to rate something, for instance they may be asked the likelihood of something, the more likely that they thought something was the higher up the scale they would put it.  This once again creates quantitative data that can be compared and contrasted against itself. 

Example: Mark on the adjacent scale how likely you would be to recommend our services to a friend- 1 2 3 4 5 6 7 8 9 10

Researchers could use the responses from this question by adding all of the responses together and dividing that number by how many respondents answered to find out how likely the average customer would be to recommend their services to a friend.

This would allow them to figure out whether they need to improve customer relations and services in order to increase revenue or whether issues were elsewhere within the business. 

Open-Ended Questions

Open-ended questions are questions which allow the respondent more freedom when it comes to answering.

They have a box as an answer space which respondents are free to type whatever they want into. This leads to more personal and detailed responses, which is arguably more accurate as people aren’t having to settle for clicking the closest to accurate answer.

These sorts of questions leave the researcher with qualitative data. Qualitative data 

Is data that provides words as opposed to numbers, it is a lot more in depth than quantitative data and can provide a lot more insight and information however it is much harder to uncover patterns and trends from this so it really depends on the researchers main goal.

Sometimes having a mixture of quantitative and qualitative data providing questions can be a good idea as you have both the in-depth quality and the ability to establish patterns. 

Examples: What dinner, that your whole family will eat, do you find the easiest to make?

This example allows the respondent the freedom to answer exactly how they want to without being constrained to certain responses. This is a good idea as the researcher could have forgotten a simple meal which lots of respondents may find easiest to make.

The only issue with this question is that the phrase whole family could be argued to be subjective as is whole family the immediate family that you live with? Does it include older children that have left home?

What about aunties and uncles. When it comes to survey questions it is very important to be objective in order to gain reliable results. 

Image Choice Questions

Image choice questions are slightly different then the other types of question mentioned above. The respondent will simply be asked a question and have to select an image as a response.

These questions do provide quantitative data which is countable however for certain subjects they arguably lack depth and description.

Image choice questions are usually used for more visual subjects, having said that they can also be really useful when working with children as children are much more likely to understand and maintain interest when pictures are being used. 

Example: Which of these photographs is your favorite?

This question could be used in order to judge a photography competition. This would work well in this context as no further depth or detail is needed and it is a very visual topic. 

Final Thoughts

There are many different types of survey questions and the above are just some examples. The type of questions that you would use depends a lot on context, topic and what the purpose of the research is.

The average survey has a mixture of open-ended questions and more closed questions leading to a mixture of quantitative and qualitative data thus allowing the researcher to get a more in-depth understanding of the topic as well as establishing patterns and trends. It is important not to ask too many questions in an online survey or it may put off the survey taker.  For more in debt studies, often focus groups or online chat rooms are a better method for gathering and summarizing results.