Click on the image to download a PDF version.

It is important to avoid possible invalid data inputs, such as a participant answering both ‘yes’ and ‘no’ to a question. In this case, you will have, say, 100 participants in the survey, but 121 responses. Where did the extra responses come from? You will then have to discard all your data and start over again.

The picture above was taken from a survey created by some students. I’ll use this as an example of what you should avoid when creating an online survey or questionnaire for your project or report.

Between squares, circles and the Matrix
Know when to use checkboxes and radio buttons in your survey. They’re not for stylistic purposes. Checkboxes allow participants to select multiple values while the radio buttons allow them to choose only one answer.

Most of the questions in this survey use checkboxes. Which should be changed to radio buttons and why? Are there other input methods that can be used?

Another input method is the matrix question, commonly used when several questions share the same answer choices, such as rating questions like strongly agree to strongly disagree or a scale from 1 to 10. In this survey, question 2 could use a matrix, but matrix questions use either checkboxes or radio buttons. Which type of matrix should question 2 use?

To text box or not to text box?
Use the text box when you’d like your participants to share their opinion or feedback on something. If your question has limited answers like age, race, income then it would be better to create a drop-down list instead. Generally, you should only use text boxes if the answers are open-ended.

In this survey, questions 1, 7 and 10 have text box inputs. Which ones should be changed to drop-down lists and why?

Direct the flow of traffic
If there are questions that must be followed-up by other questions then it would be better to split the questionnaire into several sections or pages and direct participants to the respective follow-up questions. So, in this case questions 7 and 8 should be on one different page and question 9 on yet another page. If the participant answers ‘no’ to question 6 then she is moved to question 9. She does not get to answer questions 7 and 8.

All surveys are equal, but some surveys are more equal than others
Surveys can be very laborious. You have to type out a survey in a Word document, print a copy then make photocopies. Go out to malls or areas where you think your target audience will likely congregate and cajole them to complete your survey. After collecting the forms, you have to type everything out in a spreadsheet and generate charts. Then comes your analysis of your findings.

Of course, these days you have plenty of online survey tools that do most of the heavy-lifting for you. Create your survey in minutes and spread it through your preferred social network. Sit back and wait for the results to come in. Some of these tools also instantly create beautiful charts of of your data. All you have to do is print them for your report and you’re done. Simple.

However, there are drawbacks. If you share the link through email and your social network, chances are you’re reaching out to a limited audience – your friends and your friends’ friends. You also don’t get to select or verify your target audience since you never know who is actually answering your survey. For this reason, it may be a good idea to conduct a portion of your surveys on paper as a control.

List of some online survey tools
This list is in no way intended to be comprehensive. I usually create surveys using forms in Google Docs. Before trying any of these web tools be sure to read the limitations of their basic (free) accounts, including number of responses and surveys you are allowed.

1. Google Docs
2. Survey Monkey
3. Wufoo
4. Zoomerang