You worked through all of your employee communication research planning decisions, but you’re stuck on one detail: “Should I send the survey to all employees (census) or a representative portion of the organization (sample)?”

Here are three questions to help you determine the target population (census vs. sample):

1.    How many employees work here?
When the company population is fewer than 1,000 employees, it’s better (and faster) to survey everyone. But with larger organizations, a sample is often easier to manage. For example, with a population of 10,000, you only need 964 responses for a statistically valid sample.

Calculating your required sample size is straightforward. Search “sample size calculator” on the web and you’ll be presented with a number of calculators. Target a confidence level of 95% and a margin of error of +3.

2.    How long is my survey?
When fielding pulse/spot surveys—up to 5 questions—a census strategy is fine, since these quick surveys don’t place a big demand on participants or the organization.

3.    What else is going on in the company?
It’s critical to understand how busy or distracted employees are. Ask other groups about their survey plans or if other important events (such as a new product launch or a new initiative) might get in the way. If multiple surveys are scheduled throughout the year or if everyone is working at a feverish pace, consider using a sample of the population.

Quick tip: If you decide to run a sample survey, don’t forget to invite more than your required responses. If you need 964 responses and assume a 50% response rate, you’ll need to invite 1,928.