I spent many years working in large corporations. These businesses maintained sizable human resource and information technology departments. It was only natural that they would make use of these vast resources when conducting an employee survey. Typically, employee surveys received less than a 25% response rate. These response rates were accepted as the norm, and aside from a few remarks lamenting the low response rate, little attention was paid.
I’m on the other side of the fence today. I’ve spent the last nine years providing various types of employee surveys to corporate clients. I recently had the opportunity to speak with human resources professionals from several organisations. These organisations scoured the internet for questions to use in their own employee survey. After compiling these questions, they intended to conduct their own survey in-house.
At first glance, this does not appear to be an unreasonable course of action. However, I requested that the human resources staff consider the following.
Confidentiality and Privacy
From an employee’s perspective, one of the most significant concerns about employee surveys is privacy and confidentiality. In our experience, most employees have a low level of comfort when they learn that their responses to a survey will be stored on their company’s computers. Regardless of how diligently a company works to prevent unauthorised access to survey data, the fact remains that company personnel are working with the data and conducting the analyses. Confidentiality has been violated already. There are numerous opportunities for abuse in these circumstances.
Several years ago, Entec encountered a situation in which the company president indicated that he was willing to proceed with an employee survey but desired both the database and the survey report we were going to prepare. We were forced to abandon this project. We were unable to provide unequivocal assurances to employees regarding their privacy and confidentiality.
These concerns can be significantly alleviated when employees are informed that the entire survey process will be conducted by a third party. For instance, in pre-survey communications, Entec informs employees that the organisation and Entec Corporation are separated by a firewall. Our computers are not accessible to any employee. Our database will not be visible to or accessible to any employee or company official. Any special requests by senior management or anyone else in the organisation to examine the data are categorically denied. In fact, this has occurred a couple of times in the last nine years. Privacy and confidentiality are vital concerns that must not be jeopardised. Employee surveys have a double edge. On the one hand, employees appreciate the chance to provide feedback. On the other hand, they will refuse to participate or provide candid responses if they believe their privacy will be jeopardised in the slightest.
Rates of Response
A large response increases the statistical validity of the findings. A high response rate results in the creation of a large database. A sizable database can be used to create data cuts that delve deeply into the organisation and produce meaningful results. A small database can only be used to conduct a cursory analysis that will not be able to pinpoint specific areas for improvement. As a result, obtaining a high response rate is critical for any employee survey. For instance, the Entec survey process has resulted in response rates ranging from 82% to 95%. This is significantly above average and enables detailed analysis. According to data, company-sponsored employee surveys typically receive a response rate of 30% or less.
Construct a Survey
Creating an employee survey requires three distinct steps: developing the questions, developing the scoring algorithm, and organising the questions. At Entec, we adhered to the principle that the organisation of questions is critical to the analysis phase’s effectiveness. Thus, if we wanted the analysis phase to result in clear recommendations for subsequent implementation, the questions needed to be organised in a way that reflected the desired outcome. This perspective informed the employee engagement modelling conducted by Entec as a preliminary step in the survey design process. After the modelling was complete, the survey questions were placed within the model’s appropriate sections. This process is described in greater detail in the section on Reliability and Validity below. The fact remains that the questions have a particular order. This organisation guides the analysis, which enables us to make specific recommendations for subsequent implementation.
Analyses of Surveys
The manner in which survey analyses are conducted is just as critical as the questions that are asked. The survey analysis process entails more than simply providing percentages. The analysis must provide context for the data. For instance, how do the responses to one set of questions relate to the responses to another set of questions? Certain questions are far more critical than others when it comes to employee motivation and performance.
For instance, in one company, the statement “There is little to no office politics and gossip” has a statistically significant correlation with the following leadership statements: “Takes appropriate disciplinary action against individuals who perform below expectations”, “Resolves conflicts fairly and appropriately”, and “Leads by example and action”. This type of analysis identified additional leadership behaviours that recurred as being critical to the culture of this particular organisation. The analysis resulted in the identification of the most influential leadership behaviours on best practises. As a result, the human resources department had a prioritised list of behaviours that needed to be coached. Additionally, the survey report included an assessment of how well all employees with supervisory responsibilities rated against these behaviours.
These identical statements do not always correspond to the same behaviours across organisations. They vary somewhat according to the organisational culture. Using office politics as an example, research has shown that a high level of office gossip is typically associated with a toxic work environment. In this case, this type of analysis provides the company with the confidence and knowledge that they are taking the appropriate steps to minimise gossip and thus improve performance.
Validity and Reliability
Anyone can create a survey by compiling questions. This, however, raises a question. How will you determine the validity and reliability of the questions? In other words, if an organisation designs its own survey, does it possess the internal capabilities and resources necessary to conduct the necessary reliability and validity testing to ensure the survey produces meaningful results?
Entec Corporation spent an entire year developing a series of surveys nine years ago. The procedure consisted of several steps. The first step was to bring together a diverse group of professionals with experience in strategic management, organisational development, leadership, psychiatry, and behavioural psychology. This group created models and questions to go along with them. These were pilot tested with numerous focus groups in a variety of business sectors. We conducted reliability analyses. We conducted principal component analyses. The various surveys were pilot tested and analysed again, with changes made to the surveys. This iterative process has continued to the present day in order to ensure that clients receive surveys that yield the best results possible.
If a person is ill, in pain, and has a fever, they have two options. They can either take their temperature or visit a physician. If they take their temperature, they have severely limited intervention options for regaining good health due to a lack of information. They will receive valuable information and an intervention plan from their doctor if they visit a doctor and undergo a battery of tests.
Conducting an employee survey is the same as conducting a customer survey. Organizations are highly sophisticated human systems. Using an untested employee survey, an untested survey process, and simple analysis will yield results similar to those obtained from taking a temperature. Taking the temperature multiple times impairs your ability to pinpoint specific performance-enhancing actions. This serves as a demotivator. Employee surveys inflate employee expectations. When the post-survey process demonstrates no discernible change, employee cynicism sets in and productivity declines.
If you decide to conduct a survey, a thorough diagnostic will produce results that will point you in the right direction for meaningful follow-up implementation. The organisation will receive the information necessary to proceed and accomplish its goals. The survey questions, survey structure, survey process, and survey analysis and interpretation are all critical components of generating meaningful results from an employee survey. The complexity of employee perceptions and expectations makes it difficult for an organisation to effectively implement all of the steps required to conduct a successful employee survey.