Administering surveys is one of the oldest and most common ways internal communicators measure the engagement of employees. According to CEB, 92 percent of companies use them.
Do they really work?
Absolutely. However, they can just as easily paint an inaccurate picture of how employees are really feeling and if they are engaged in the way that matters to the performance of your organization.
Whether you choose to an employee engagement survey vendor such as the firm that started the trend (and currently claims we’re in an engagement crisis) or roll your own, a lot of effort goes into creating a survey that provides the answers communicators need to get an accurate look at the state of things and make decisions about how best to move forward.
Here are a few of those essential ingredients:
1. The right questions
You can’t find out if employees are engaged just by asking them, “Are you engaged?” You have to find the specifics that make it clear that they’re engaged or not. The International Association of Business Communicators offers this advice: “Survey questions need to be precise, unambiguous, efficient in the way they capture information and, in most cases, should employ answer categories that can be used to quantify responses.”
Given the results of 2016 election surveys, we all understand how the nature of the questions will determine what you are in fact measuring. Testing has shown the precise wording and order of questions will impact your results.
Some survey question sets, such as the Gallup G12, which asks, “Do you have a best friend at work?” tend to be focused on employee satisfaction. For more of an organization performance angle, a very good place to start would be the CEB’s list of the top nine survey questions.
Yes-or-no questions provide the clearest measurement results, yet not every question can be black-and-white. Engagement questions can be quite broad reaching, so it’s important to keep in mind some employees may be intimidated to honestly answer questions like “Do you understand the strategic goals of the company?” While open-ended, comment-style answers are quite difficult and time-consuming to measure, instead you might ask employees whether they agree or disagree with certain statements, have them rank options or make value judgments on a scale (“excellent” to “poor”). Other companies have borrowed from the Net promoter score concept, and ask, “How likely is it that you would recommend working at our company to a friend?”
2. A well-defined schedule and distribution strategy
Considering it can take two to three months to create, execute and report out your survey, timing is critical, as are your distribution methods.
Will it be available via email, on your intranet, at kiosks, on paper or all of the above? Will you attempt to survey every employee, or take a randomized sample approach? How will you ensure your sample is random and representative of the entire population?
While some companies conduct surveys every two years, Quantum Workplace offers research which suggests one survey per year is optimal. Others find a continuous feedback approach allows them to be more nimble and ahead of the curve given today’s rapidly changing economic and social environments. Companies such as CultureIQ and Agency BB&A offer reminders that focus groups, flash polls and asking for feedback on social media provide other useful ways of taking the pulse.
3. A statistically valid response rate and effective reporting of results
A survey isn’t worth much if not enough people answer it. That’s why it’s important to conduct outreach and conduct follow up that encourages employees to complete the survey for their own benefit as well as the company’s. This includes non-digital follow up with people in order to get a representative sample of responses.
To get accurate results, you don’t need everyone to respond to your survey, and the larger your organization the lower percentage of participants you actually need.
Generally, the larger the responding sample is, the higher your confidence will be, and the lower your margin of error. Simply speaking, the more response you get, the more accurate your results.
Many surveys aim for a 95 percent confidence level, which simply means that for every 100 people in the population, 95 will answer the same as your survey results. For an organization of 10,000 employees, for a 95 percent confidence level and 5 percent margin of error, you only need 370 responses. If you increase your accuracy to 99 percent confidence with a 2.5 percent margin of error, you will need 2,103 responses. For an organization ten times larger, those response numbers are just 383 and 2,594.
Sample size is important, but not as much as randomization. With surveys it is most important that you gather responses from a representative sample of your population. When you rely on only one method of distribution and allow a broad population to self-select their participation (such as a large email blast where you just take whatever responses you get), you can easily introduce selection bias errors. Perhaps only the more engaged employees respond or only the employees active on email, or only those with more free time. Obviously such bias may significant skew your results.
Communications teams play a critical role in the success of the employee engagement survey, as well as overall employee engagement.
To get higher levels of participation, it’s important to explain why you are doing the survey and provide examples of how you have acted on the results previously. Then follow up diligently and include data regarding your participation objectives and current status, but only send reminders to those who have not yet completed the survey, and send thank-you notices to those that do. When it comes to reporting out the data, CultureAmp offers good suggestions which include providing an overview of key findings, presenting results to executives first, and publishing an action plan.
Remember: Your survey is just one part of the larger whole of communications activity to help foster higher levels of employee engagement.