When compared to telephone surveys and face-to-face interviews, online surveys (questionnaires) are a more efficient research method. However, there are a number of factors that can influence the success of your project and impact results.
By following these best practices, you’ll avoid some of the pitfalls others make when conducting survey research.
Subject and Aims
Before you begin writing survey questions, you should be aware of your project purpose (or survey methodology). Whether it’s for academic research, social research or gauging customer satisfaction, your whole survey will be centred around that purpose (or subject). You’ll also have specific aims for your research or, in the very least, a set of expectations for your survey results.
Each of your questions should clearly relate to that purpose and contribute in some way to meeting those expectations. You can manage those expectations by writing a short introduction to your project at the start. You can read more about this in the providing extra information section of this article.
This will make respondents feel as if they’re contributing to something, rather than the survey being just another task. However, we don’t recommend including your aims or goals in this section as they may influence respondent’s answer choices.
You must remember, respondents may not always want to share the information you’re asking for. If they feel you’re asking more of them than they expected, they may well leave.
Tldr: Be as transparent as you can be with respondents. You should be creating a survey to meet your research aims, not gathering as much data as possible.
Voice and Tone
Let’s face it, taking surveys isn’t the most exciting activity. And when respondents have to plough through volumes of overly formal text, they’re going to switch off. Make your survey more engaging to respondents by writing in a conversational tone.
Using a voice that matches participants’ goes a long way in building a rapport with them. It may also discourage forms of response bias caused by disengagement.
Tldr: Use a conversational tone to build a better relationship with your respondents.
Spelling and Grammar
This should go without saying, but you need to pay close attention to spelling and grammar when creating a survey. Clarity is key to understanding and if your respondents don’t understand something, their answers are all but useless.
Tldr: Proof read your surveys before sending them.
Use simple language
As with any piece of content, you should cater to the widest audience possible. If you assume that all your respondents are educated to your level, you’ll alienate a portion of them.
For example, if you’re trying to gauge public opinion on a topic, you’ll want your questions to be understandable by as large an audience as possible.
By using simple and clear language, your project will be more inclusive. This will go some way in improving your response and completion rates.
Tldr: Don’t use complex language or industry specific terms unless it’s essential to your project.
Don’t use double negatives
Let’s get the definition out of the way. ‘Double Negatives’ are sentences where two negative words are used in the same clause, e.g. “I’m not unhappy”. These switch the meaning of a sentence from a negative to a positive.
They may not seem confusing when written but can cause ambiguity. This can lead to respondents choosing a different answer choice to what they would have or leaving altogether. Keep in mind that they’re giving up their time, make your question text as clear as possible.
This is especially true for respondents who speak English as a second language and aren’t familiar with informal English.
Tldr: … don’t use double negatives.
We’ve already covered this above, but we want to emphasise the point. The use of subject specific terminology, undefined abbreviations and other complex terms will alienate a portion of your target population
It’s important to keep your respondents in mind when conducting survey research. After writing a list of questions and answer options, go through them with a critical eye. Ask yourself if you’d interpret them in the same way without your knowledge or ask someone who isn’t involved in the project to do the same thing.
Alternatively, if you have more time for the project, run a pilot survey and analyze its effectiveness.
Keep in mind that this point won’t apply to you all. Some types of surveys will require the use of specific terminology or acronyms, but it’s up to you to know when (and when not) to use them.
Tldr: Write for your audience’s understanding, not your own.
Don’t Assume Understanding
Make no assumptions about what your survey respondents know. If your questions require some prior information or understanding, you should provide it.
For academic surveys, this information may be defining key terms or a brief explanation of your research topic. For market research, it may be information on a service or product.
However, it’s important that the information you provide does not influence your respondent’s answer choices.
Tldr: Provide the information respondents will need to understand and answer your questions.
When providing extra information
If you are including extra information in your survey, we don’t recommend putting it in the question fields. This is something we see this all too often and it’s off-putting for respondents. Include all the general information your participants will need at the start of your survey.
There may also be cases where a question needs further information. An example of this may be in customer service skills assessments for job candidates, where you detail a hypothetical customer scenario.
You should use a text field for providing this information, as opposed to writing the scenario out in the question space. Or use images and videos to provide that information, these are usually more engaging to respondents. When creating a survey with KwikSurveys, you can find a text, video and image fields among our Presentational Items.
Tldr: Don’t stuff question fields with text. Keep them short and easy to read.
Measuring Change over Time
If you’re using a survey to measure change over time, always use the same language as any surveys conducted previously. This way you ensure that any changes in your results aren’t caused by language or structure.
Tldr: Don’t change question phrasing in bench-marking research.
To run a successful survey research project, your content must be clear and engaging for respondents. Keep in mind that it’s likely to be seen by a range of people with different levels of education and understanding. The more accessible it is, the better your survey data will be.