Random Sampling and Employee Engagement Surveys:

What are the advantages and disadvantages?

Do you have to survey everyone?

You don't, but before you decide to use a random sample of employees for your survey process, it's important to understand how simple and stratified random sampling work and under what conditions they work best for organizations.

What is simple random sampling?

This is a common method in which researchers select a subset of people from a larger group. It’s important that every single person in the larger group has the same probability of being selected. This ensures that the subset of people is representative of the whole group.

Simple random sampling is used daily: for jury duty, pharmaceutical testing with a placebo group and drug-test group, and pollsters.

In regard to an employee engagement survey, some organizations might choose to use simple random sampling instead of inviting every employee to fill out the survey.

What is stratified random sampling?

When you're surveying a big group of people, it usually isn't enough to just look at everyone together. Sometimes you need to understand how different groups within that big group feel or think. For example, you might want to compare leaders with employees, different age demographics, different branches, etc.

To do this, use something called stratified random sampling. Instead of just picking random people from the whole group, you pick random people from each sub-group within the big group. It's a bit more work because you have to survey more people, but it can provide the organization with valuable information.

When you're planning to do this kind of sampling, consider the single most relevant demographic division in your organization. It's usually best to pick just one way to divide them because it gets too complicated otherwise. For example, if you want to look at how employees feel based on their job level and job type, you'd have to look at each combination of job level and type. But sometimes, you might end up with very small groups, and it doing a stratified sampling doesn't make sense anymore.

What are the disadvantages of random sampling for employee engagement surveys?

  • Only a select group of employees are given the opportunity to voice their concerns and opinions. This can cause staff to feel left out or excluded.
  • Staff might not “believe” it to be random and might question the validity of the data. (The data doesn’t reflect their views).
  • It can increase the level of variation in your survey results.

What are the advantages of random sampling for employee surveys?

  • It can save time and resources.
  • It can help mitigate bias. When ‘everyone’ is filling out the survey, you might only get employees who are enthusiastic or disengaged (wanting to praise or stump).
  • It can reduce survey fatigue. (It might be more appropriate for follow-up pulse surveys for specific actions taken post-survey.)

  • Employee Surveys: Should you use a random sample?

    We’ve been working with organizations for over 25 years, and we believe that random sampling isn't typically effective for smaller organizations. However, for larger companies with tens of thousands of employees, it becomes a viable option.

    When surveying a segment of a population, there's inevitably some margin of error in the findings. However, once this margin narrows to just a few percentage points, it often becomes negligible.

    In smaller populations comprising only a few hundred individuals, all of them probably need to be surveyed to attain the desired level of precision. Conversely, as the population size grows, the proportion of individuals necessary to achieve high accuracy decreases.

    Larger population = Smaller proportion surveyed
    Smaller population = Larger proportion surveyed

    Your random sample will yield statistically significant results at the organizational level but may not suffice for statistical significance at the departmental level.

    Significant employee engagement or satisfaction issues frequently lurk within smaller subgroups. Using random sampling may inhibit the detection of these concealed pockets of discontent since there won't be enough employees selected within those small groups to measure local employee attitudes.

    How to Use Random Sampling Effectively for Your Employee Survey:

    • This can be effective in organizations with over 1,000 employees.
    • Use our our simple random sample calculator to calculate how many people you need for a random sample (if you haven’t conducted the survey already) for your employee survey process to be effective. (Likewise, some organizations want to know how accurate the data are in a survey they’ve already conducted, taking into account the number of responses received.)
    • Communicate with employees, sharing with them the random selection process.
    • Collect feedback on broad, organization-wide priorities like leadership, company confidence, and learning and development.

    How engaged are your employees? How engaged are you?

    Try this sample survey to get a personal performance and engagement profile.

    Contact us today, and we can help you determine what might be best for your organization.

Random Sampling and Employee Engagement Surveys:

To receive periodic articles & research updates, sign up for our newsletter mailing list.
Email address