Surveys may be one of the most commonly used, but also most misunderstood, tools in the world of business. How often have you found yourself answering questions and soon wondering why a company would need to know such ‘irrelevant’ details? How often have you regretted saying “yes” to a supposedly quick survey after over 10 minutes of answering such questions? And how often have you been stuck for how to answer a question due to unclear wording, leading you to simply throw out an answer, just to get on with it? Unfortunately, it is the repetition of such experiences that makes us all more likely to decline a survey, even when it may aid in the development of something valuable to us, let alone when the subject is further from home.
In previous editions of Empowerment Through Knowledge we have mentioned the use of surveys to gain insight into the views and habits of employees within your organisation, such as when using the Cultural Web as a tool for the analysis and development of organisational culture. Surveys are also a widespread tool in the field of market research, gaining information on the opinions and behaviours of those outside your organisation, to aid in business development. They are also commonplace in the gathering of feedback from current or past customers. Therefore, they are a tool that nearly everyone will need to make use of at some point in their career.
Why do you want to conduct a survey?
First, let’s explore the reason for conducting your survey. What do you want to find out? This is the most important question that you will need to keep in mind throughout the development of your survey. With each question you put down, ask yourself; “how does this question help me to reach the goal of my survey?”. If you cannot answer, or if your reasoning is stretched, it’s a question that will simply prolong the survey, wasting the participant’s time, and yours. If you had a real person asking the questions, wouldn’t you rather they spent their time gathering more opinions from different people, rather than asking irrelevant questions? If your survey was conducted by chatbot or via email, then this is still a data point that you will need to process in some way. Just because you notice that most surveys ask questions like “What city do you live in?”, this does not mean that the question will be relevant to your exploration of customer satisfaction with the new design of your website, for example. The idea that this question may come in handy at some point in the future should also be avoided. Try not to take on the attitude of a hoarder when it comes to your questions. Very often, when we keep something because we might need it one day in the future, we spend years storing it just to throw it away when it becomes too old to be used.
The same thing happens with data, just that data tends to become outdated and unusable a lot quicker than physical items. It will be much more effective to simply conduct a new survey in the future, when the question is relevant. The previous, positive survey experience that your participants had, will likely mean that they will be happy to participate again, knowing the different goal of the new survey.
Furthermore, knowing the goal of a survey puts each question into context for the participant and tends to make them feel more comfortable while taking part. Of course, it is not always possible to provide the full truth of why the survey is being conducted, but knowing the general aim is often enough for someone to consider saying “yes” to answering your questions.
Types of Data
There are two main, distinct types of data: Quantitative and Qualitative data that require very different methods of analysis.
Quantitative data is what most of us are most familiar with. Numbers. That being said, the answer to a question need not be a number, such as on a scale of 1 to 5, to be quantifiable. ‘Yes’, ‘No’, or ‘Not Sure’ answers can also be counted and compared based on characteristics of the respondents. Similarly, asking someone for their nationality also produces quantitative data as the responses can be grouped by country and counted. These answers come from closed-ended questions.
Qualitative data is abstract. The question prompts an answer through descriptive words that are not chosen from a list of options, but that come from the participant’s own thinking. The answer must be more than a simple ‘yes’ or ‘no’ as these results cannot be directly counted and generally result from open-ended questions. The analysis of such data tends to be a little trickier and more delicate than that of quantitative data. More about this in a soon-to-come edition of Empowerment Through Knowledge.
How will your survey be conducted?
Consider the types of questions being asked. Will the participant need to be prompted to give a more detailed answer? In the case of primarily qualitative data, it is often better to have a real person asking the questions. The fact that the participant can simply speak about their experience and does not need to find the right way to express it in writing generally allows for a much more detailed response. When people need to do the writing themselves, they tend to give you a lot less content to analyse. In certain situations this may be beneficial as it keeps the answer on point. However, a well-trained surveyor who knows how to direct the participant’s answer towards the key components of their experience, will still allow for a more in depth answer about what was relevant than if the participant had to write it down themselves.
When you deal with quantitative data, on the other hand, having a person repeatedly asking someone to rate something on a scale of 1 to 5 may be a waste of your resources if you are dealing with high-level participants. You need to keep your audience in mind. The participant will likely also complete the survey in less time if they are able to simply read the questions themselves and answer, rather than having to occasionally ask the surveyor to repeat the question, depending on its complexity. If you, however, are conducting a survey with a group of participants who would likely have an apathy towards reading the questions, who are not tech-savvy enough to answer the survey online, or who simply cannot read, you should consider having your survey conducted by a real person.
Coming back to the topic of your survey, it is also highly important to consider the sensitivity of the questions being asked. To take an extreme example, let’s say you were involved in the development of a new contraceptive. How comfortable do you think people would feel answering questions about their sexual habits to someone on the phone, or even in person?! Not much at all.
In Part 2 of this article, we will dive into depth about how to avoid skewed or low-quality results due to your survey’s delivery method, audience, question wording and other factors.