‘Prefer-Not-To-Say’ Is Not An Option In Employee Surveys

Summary: Here I discuss the potentially negative impact of the ‘Prefer-Not-To-Say’ (PNS) response option in employee opinion survey feedback.

“An approximate answer to the right question is worth far more than a precise answer to the wrong one” (John Tukey)

Many times, I am asked by HR professionals and clients if they should include the ‘Prefer-Not-To-Say’ (PNS) response option in their employee opinion surveys, especially in the demographic section of their questionnaires. They believe such PNS response will help employees during the completion of their survey and will limit any potential inaccuracy in their responses. Their rational is that it is better that employees choose a PNS response option rather than giving a random and / or potentially inaccurate response to the survey question.

Basically, they think that if they do not include the PNS response option, they will be driving employees to make choices about something they have no willingness to comment about. They sense that ‘forcing’ employees to answer specific survey statements will lead to lower reliability of their employee feedback. Also, they argue that PNS response options provide employees without knowledge on the subject or positive attitude towards certain survey questions an opportunity to express that specific opinion directly.

Should I Include ‘Prefer-Not-To-Say’ In My Employee Surveys?

So, now we understand why some HR colleagues are keen to include PNS response options in their listening interventions. Unfortunately, when it comes to employee surveys (rather than market research), I truly believe this sort of reasoning is WRONG! I personally believe that including PNS response options in employee surveys actually create a problem by allowing employees to actually disengage from the survey. Given the opportunity, employees, especially the apathetic or indifferent ones, will just select PNS options to avoid having to think too much about their answers and the survey as a whole.

My 15+ years-experience in employee research indicates me that the PNS response option mostly attracts all the disengaged employees, those that don’t want to be identified, are not bothered reading the questions, or simply don’t feel like reasoning their answers properly.

Yes, employee opinion surveys involve thinking. Sometimes it is not easy to answer opinion surveys as employees are actively requested to create a precise judgment on each survey statement topic and translate their verdict into the selection of one response choice. This is hard work for many employees, especially those with lower education levels or limited literate ability.

When the amount of mental work exceeds the employee’s motivation or ability, they will start looking for ways to avoid the assignment while still appearing as if they are carrying on with the survey. The bad news for social scientists is that employees have a simple way of avoiding this intellectual workload by simply selecting for the PNS response option in the survey... Very easy way-out for them!

Relevance Of The ‘Prefer-Not-To-Say’ Response Option

In general, in employee surveys, I would try to avoid the PNS response option all together. The exclusion of a PNS option will encourage employees to consider an answer and think harder about their potential response choice. If employees don’t want to answer, then no problem, they can simply skip the question by leaving it blank/unanswered. This alternative should be clarified in the ‘how to respond to the survey’ instructions. All survey questions should be voluntary. We should not force employees to answer all survey questions by making each of them compulsory. So, make sure you offer employees this escape route for not answering unwanted questions instead of offering the PNS option.

Some HR colleagues argue it is probably a good idea to use the PNS response option only for very sensitive survey questions, where employers need to be ‘politically correct’ on certain delicate subjects such as sexual preference, religious belief, gender, ethnic background, etc. However, as already discussed, these questions should always be voluntary and never compulsory. The employee should be able to leave the question blank if he/she wishes to do so. Hence, the PNS response option becomes irrelevant in these cases.

When we consider attitude / opinion survey statements, if using a five-point Likert scale type of response, the middle point already caters for the undecided respondents, so again, the PNS response option becomes irrelevant.

It is simple, questions that exclude the "PNS" option produce a greater volume of accurate data. Not using the PNS response option offers a greater chance of getting a valid response. Getting an answer or even a guess is better than getting no answer at all.

Impact Of ‘Prefer-Not-To-Say’ On Sample Size & Response Rates

There is generally no impact on response rates depending on the inclusion or exclusion of PNS response options in the survey. However, those organisations that decide to include a PNS response option in every single survey statement will unnecessarily increase the length and complexity of the survey which will have a negative impact on response rates.

On the other hand, offering PNS response options will most likely reduce your effective sample size. Consider that a significant proportion of respondents will opt for the PNS option and, as a consequence, their feedback will most likely not be included in the core analysis of your survey results. The negative impact of PNS responses in the data, due to the sample size, will be even stronger for small socio-demographic groups.

As a rule of thumb, approximately 10 to 15 percent of employees tend to opt for the PNS option. There are several potential reasons for this tendency, being lack of employee engagement the most common one. Consider the following:

• Lack of motivation in answering the survey. Employees are tired, bored, simply don’t care or they feel that management will do nothing with their feedback.
• Employees just want to take an easy route to avoid thinking too much.
• If it is a demographic related question, employees may just want to remain anonymous / avoid any potential way of being recognised.
• Employees simply don’t understand the question, or the survey exceeds their ability to answer, or the survey is too long (Bad survey design).
• Employees do not want to commit themselves with an answer, do not really have a view on the subject, do not want to offer an opinion on the surveyed topic, or are unable to respond, they truly don’t know the answer.

‘Prefer-Not-To-Say’ Can Render Your Employee Feedback Unusable

Providing PNS response options is likely to produce less valid data. From a data analysis perspective, there is nothing we can do with the PNS response choice. It is an invitation for disengaged employees to hide behind that option. My own employee research experience shows that there are greater chances disengaged employees will choose the PNS option. This response gives these employees the perfect opportunity to cut across the questionnaire without focusing on the questions nor the answers.

“The ‘Prefer-Not-To-Say’ response option provides an easy way for dejected employees to avoid engaging with an employee survey.”

The inclusion of PNS response options in the demographic section of employee surveys makes the employee feedback pretty unusable, this is more predominant in small organisations with reduced employee headcounts. The whole point of the survey is to identify where potentially disengaged employees sit in the organisation so that leadership can act on the results. If employees do not specify their demographic profile and where they sit in the organisation, how can leadership take focused corrective actions on the organisation’s survey results?

By not being able to identify the demographic and organisational background of employees, the PNS option reduces the effective sample size which is available for analysis. Employees who choose PNS options cannot be included in the data analysis. This is even worst for lower educated respondents who tend to opt for the PNS option far more often than higher educated respondents. You are simply inviting employees not to think - This PNS preference simply biases the whole analysis.

‘Prefer-Not-To-Say’ Case Study

Consider the following real-life case study in the global retail sector (see table below). In this case study, out of 5,060 effective respondents, 612 employees opted for the PNS option (12% of the overall employee population). We can also note that the PNS column shows the most negative results in all themes being surveyed (red column) with opinions well below the organisation’s global average (see 'Overall' yellow column). Unfortunately, because these people were given the chance to chose the PNS response option, we do not know what is their job level or where they sit in the organisation.

This same situation also happened with all other demographic questions: gender, age, ethnicity, sexual orientation, physical ability, etc. where the less engaged respondents also opted for the PNS option. Basically, we lost the chance to analyse 10 to 15 percent of our workforce’s ‘most negative’ feedback. In the example below we have lost the feedback of our 612 most disengaged employees!

The PNS response option has a negative impact on all post-survey action planning activities, because we cannot analyse the opinions of our most disengaged employees, and moreover, we don’t know where they are located in the organisation and who they are (demographic profile).

The inclusion of the PNS option misrepresents the percentage of employees in each category being investigated, in this case job level. In the example below, those who prefer not to say were treated as a separate category in the analysis. In similar situations, it is best to treat the PNS option as a non-response (missing value).

Imagine organising a global survey of 150,000 employees and wasting 22,500 responses (15% of all effective survey responses) – This would be unforgivable! Using the PNS response option in employee surveys will make you look rather amateur in the eyes of key stakeholders and executives. It will be embarrassing to have to defend this position in your survey results presentations.

Some Final Thoughts…

My personal view on the inclusion of the “Prefer-not-to-Say” response option is that it does not reliably improve the quality of the employee survey data. Instead, it is possible that those employees attracted by the “Prefer-not-to-Say” response choice would have provided more substantive answers if no such PNS response option had been offered at all.

Yes, the PNS response option is potentially wasting useful responses from employees that otherwise would have opted for a more substantial answer. As a general rule, it is best to avoid using PNS response options in employee surveys. This will minimize the risk of getting sloppy results in your next employee survey.

Let’s face it, organisations cannot afford to waste useful employee feedback by including unnecessary PNS response options in their surveys. Implementing complex global employee surveys to gather incomplete data is rather an inexcusable waste of money (survey vendor fees), internal resources, time, team effort and, most importantly, loss of enthusiasm for future employee listening initiatives in the organisation.

• Read more: Ensure ‘authentic and meaningful’ survey participation
• Read more: Don’t survey employees just because you can!

Thanks for reading my blog & your kind support. Were any of the insights provided of value to you? I would welcome your feedback - please do ‘Like’ or ‘Comment’ your experience on this subject in the space provided! -?Follow me on LinkedIn?for more articles and insights!

Disclaimer: The author is making this ‘Opinion Blog / Guide’ available in his personal effort to advance the understanding of best practices in workplace related matters. The author assumes no responsibility or liability for any errors or omissions in the content of this ‘Opinion Blog / Guide’ or for the results obtained from the use of the information provided. The information is provided on an ‘as is’ basis with no guarantees of completeness, accuracy, usefulness or timeliness and without any warranties of any kind whatsoever, express or implied. The views expressed are solely those of the author in his private capacity and do not in any way represent the views of any entity whatsoever with which the author has been, is now, or is to be affiliated in the future.