Employee Surveys: 7 Common Mistakes that Will Ruin Your Data
By Carolyn Nevitte
June 17 2019 - Employee surveys remain one of the best methods to measure and track employee engagement. In an age where managers and business leaders need to be increasingly concerned and invested in the employee experience, it's important for organisations to keep up to date with employee feedback, views and insights so appropriate changes can be made for better success.
There are a number of reasons to conduct an employee survey. On top of accurately measuring employee engagement, employee surveys give your workforce a voice. They show how much your company genuinely values their opinions and they direct organisational growth. But although writing questions for an employee survey might seem like an easy task at first, without appropriate planning and forethought, your poorly written survey questions stand to ruin the potential for perfectly good data.
Word choice is everything when it comes to employee surveys. After all, words have a profound impact on our minds - they are powerful and influential, so we need to be careful how we phrase questions if we are looking for meaningful, accurate results. This is why the best employee surveys are designed by business psychologists, who are able to craft the right questions to get to the bottom of an organisation's problems.
Below are 7 common problems that can ruin your survey data.
1. Not Knowing What You Want to Achieve or Measure
When it comes to employee surveys, there is nothing more useless and frustrating than hindsight. It's a horrible feeling when you get to the stage of sitting down to analyse your employee survey results and you realise you don't have the right data at all. You haven't asked the right questions and the results you do have aren't reflective of anything. The result is a loss of time, money and a huge sense of disappointment.
To avoid this common pitfall, you need to be thorough and understand what it is you are attempting to measure. What problems have you noticed in your organisation? Have you received employee feedback about aspects of the employee experience? Your survey needs a goal - what are you trying to understand? Thinking this through ahead of time will increase the likelihood of you obtaining useful data you can turn into meaningful change.
2. Question Overload
More isn't always better. Sometimes, it's better to keep things short and simple. Design questions that allow you to get to the point without taxing (and potentially boring) your employees with too many questions.
When employee surveys get too long, employees might lose patience or focus. They might get annoyed and they might be deterred from completing the survey at all. Your response rates will suffer, but the quality of your data may also decline. Make sure none of your questions duplicate each other and be strict about whether each and every question is entirely necessary to serve your purposes.
3. Unclear Language and Word Choice
When creating questions for your employee survey, try to be as clear and concise as possible. In fact, it's best to adhere to George Orwell's rules for writing. Specifically,
- Steer clear of metaphors, similes and figures of speech
- Always use a short word over a long word
- If you can cut out a word, cut it
- Never use the passive over the active
- Never use a scientific or a jargon word (or an acronym) if you have an everyday English equivalent
If you have to use industry words, even if you believe your employees will all understand their meaning, be sure to include definitions or examples for clarity.
4. Leading Questions
Your employee survey questions should never be worded in a way that might sway your employee one way or another. The best survey questions make use of neutral wording. A bad choice of wording, for example, would be:
"How stressful do you find performance appraisals?"
The word "stressful" immediately puts the respondent in a particular, negative, frame of mind, which will skew your data.
5. Loaded questions
Loaded questions are written in a way that assume the respondent agrees with the question being asked. Similar to leading questions, it puts employees in a particular frame of mind, but it doesn't allow them the opportunity to qualify or disagree.
Surveys need to be written in such a way that solicits unbiased, honest answers. A bad choice of wording, for example, can be seen below:
"What aspect of working at (company name) do you like the most?"
The question above makes the assumption that the employee enjoys their time at the company at all. It might be that you are forcing your employees to answer dishonestly. To get around this, you can first ask employees whether they enjoy working at the company (and how much, on a scale of 1 to 10). You can then ask them about aspects of the employee experience. If an employee is entirely dissatisfied with the company, you can use skip logic so employees don't have to answer questions they feel don't apply to them.
6. Double-Barreled Questions
This is an extremely common survey mistake. It occurs when you force your respondents to answer two questions at once. Each question of your employee survey should be written in such a way that allows you to measure one thing at a time. For example, it would be a bad idea to ask the following question:
"How satisfied or dissatisfied are you with the levels of support and communication provided by your manager?"
An employee might, for example, feel that they are afforded a lot of communication with their manager, but they might be entirely unhappy with the level of support provided. Such a question should be separated to create two unique questions.
7. Boring or Ugly Survey Design
You want to keep your employee's attention long enough for them to complete a survey. You are more likely to get a good completion rate from a well-designed survey than from a poorly designed one and, of course, the more complete and considered your responses are, the more accurate your data is.
So What SHOULD You Do?
To avoid falling into all the traps above, it's wise to adhere to an employee engagement survey strategy, such as the PEARL™ model. An employee survey requires a methodical, calculated approach and carefully considered questions that will give you solid, actionable results. The science behind models such as this makes them robust, as they have been based on the latest research on employee engagement, stress and wellbeing.
When in doubt, keep your surveys simple and lean. Keep your employees involved. When done right, you will be left with happy employees and accurate, meaningful data.
About the Author:
Carolyn Nevitte is HR Director at People Insight, a company that helps organisations measure and improve the employee experience through employee surveys, 360-degree feedback and expert consulting.