Employee Surveys

What is the primary objective of your survey?

The bottom line for any survey is to improve organisational performance.

There are many paths to achieving this goal.  Enhancing employee experience to attract and retain talent, increasing employee engagement to raise productivity, improving wellbeing to reduce absence, enhancing diversity and inclusion to foster innovation, changing organisational culture for better customer service.

Essentially, surveys help to identify what needs to change, suggest where to focus efforts, evaluate the effectiveness of actions and celebrate successes.

So an important first step is to be clear about your objective - the thing that you would like to improve. It might be employee engagement, wellbeing, organisational culture or something else or a combination of things.

Once your objective has been decided, the next step is to identify the things that could lead to improvements - the drivers of your objective.

What are the potential drivers of your objective?

You might have identified increasing employee engagement as your objective. There are many factors that drive employee engagement but which ones should you consider measuring in your survey?

To get some ideas you could review what's been written about your objective to see what, according to research, drives it.

Looking at employee engagement, there's lots of research on the job demands - resources model and engagement. Guided by this model your survey might include things like workload, emotional demands, support, autonomy, organisational climate.

A good starting point but does this fit your organisational context?

Interviews with senior leaders, asking for their views on what they would like to get from the survey and what it should focus on will help to refine your thinking. Furthermore, consulting and involving leaders from the start will help to secure their buy-in to the survey process.

Focus groups with employees is another great option for identifying and refining the drivers measured by your survey.  It’s helpful to consult employees to get a sense of their concerns and get close to the issues.  They're best placed to provide insight into what helps and hinders engagement in your organisation.

Plus focus groups are an opportunity to communicate clearly that employee concerns are taken seriously, and the survey is more than a top-down exercise.

The next step is to identify suitable questions to measure your objective and drivers.

What are the best questions to ask in your survey?

You will need questions to measure your objective and drivers you have identified as potentially being important in your organisation.

Measuring your objective

Questions about your objective will usually focus on individuals and their thoughts and feelings. They are fairly universal and apply across organisational contexts. You will find lots of examples of questions from practitioners and researchers.

For engagement, a good source is the work of a group of psychologists from Utrecht University who define work engagement as a state of mind characterised by high levels of vigour, dedication and absorption.

Other measures of engagement focus on commitment - employees' attachment to the organisation.

Measuring your drivers

Driver questions differ in their focus to questions about your objective. They are focussed on tangible aspects of working life in your organisation. A critical feature of driver questions is that they should be actionable. In other words it should be clear from the question what action would be needed to improve the thing it is measuring.

Your driver questions will be more tailored than your objective questions to fit the context of your organisation. Data collected from senior leader interviews and employee focus groups will help you write suitable questions.

Care should be taking when creating questions to ensure they will provide useful, reliable and valid data.  There is a fair amount of skill in asking the right questions in such a way that you end up with useful data you can use to inform decision making.

How to create good survey questions?

Response scales

What response scale should I use?

Likert scales are the most common.  You will have seen these many times before.  The most common scale for employee surveys is a scale ranging from strongly agree to strongly disagree.  

The key features of Likert scale are:

• Equal number of positive and negative options
• A description of each option - strongly agree / agree / disagree / strongly disagree
• A numeric value for each option - 4 / 3 / 2 / 1

Other scales can be created to suit your questions.  For example,

• Very likely / likely / unlikely / very unlikely
• Very confident / confident / unconfident / very unconfident

Four to seven options are best – longer scales use up too much of respondents’ cognitive energy deciding on their response.

Likert scales are popular because they allow respondents to indicate whether they feel positive or negative as well as the strength of their feeling.  Dichotomous (yes / no)  scales can be used but Likert scales give more options for follow-up statistical analysis.

Should the response scale include a mid-point?

There are differing views on whether to use a neutral option such as neither agree or disagree or to force respondents to choose a positive or negative response by not including a midpoint.

Research shows that results are not distorted either way.  In other words, there’s no right or wrong answer, but importantly, once a decision has been made the same scale must be used across surveys to ensure valid comparisons.

How long should the questionnaire be?

The length of your questionnaire will depend on the type of survey. There's a balance between sufficiently covering the things you would like to measure and the length of the questionnaire. If it's too long there's a risk of low response rates or poor quality responses with little thought given by respondents to answering the questions.

The length of a questionnaire can be measured by the number of questions or how long it takes to complete.  Of course, the two are related but time to complete is more helpful.

A well-designed questionnaire with 30 simple and unambiguous questions might take less time than a poorly designed questionnaire with 15 questions.

Five to ten minute completion time is a reasonable target.

What type of survey should we run?

There are different types of survey related to their purpose, their frequency and who takes part. Many organisations use a combination of survey types.

Assessment / baseline surveys

Assessment / baseline surveys seek feedback on a wide range of topics to set a baseline and identify areas that need improvement.

Assessment / baseline surveys are common for organisations running their first employee survey and are usually repeated annually. They typically involve all employees.

Trend / monitoring surveys

Trend / monitoring surveys can be used in combination with assessment / baseline surveys and are often referred to as pulse surveys.  

These surveys track changes in views on issues that have been identified as requiring attention to gauge the effectiveness of interventions and indicate where adjustments might help or monitor engagement levels to pick up early signs of disengagement before the next assessment / baseline survey.  

Much shorter in length than an assessment / baseline survey. Usually all employees are involved.

Diagnostic / deep dive surveys

Diagnostic / deep dive surveys take an in-depth look at a specific topic that might have been identified in an assessment / baseline survey. Involving specific demographic groups might be appropriate.

Employee experience / employee life cycle surveys

These surveys seek feedback on employees’ experiences at different stages of their time with an organisation - recruitment, onboarding, work anniversary, role change, post parental leave, return to work after long term absence, employee exit.

Preference surveys

Preference surveys gather information on employee preferences on things like benefits, flexible work arrangements, training to help make better informed decisions. Sometimes a stand-alone survey but often incorporated in an assessment / baseline survey.