Many companies, being seduced by the obvious cost savings, are trying to do their own Employee Surveys. There is an obvious dilemma with DIY Employee Research. It’s sort of like going up to someone and asking them what they like and dislike about you. Will they be honest? And even if you promise confidentiality, will they believe you? Research has shown that to get good information, you need to use an objective, third-party marketing research firm to survey employees.
None the less, many companies believe their companies are different and that they have “good enough” relationships with employees to get honest and candid answers, without employee fear of reprisals. However, whether that is true or not, whether you DIY or use a reputable and experienced marketing research firm to complete your employee research, there are several common pitfalls that you can avoid to improve your research.
- Use the KISS principal. Keep your employee survey short and simple. If your employee survey is too long, you risk employees becoming bored and frustrated. That results in poor quality data on the back-end questions, as well as lower completion rates. Simple questions are easier for employees to answer, as well as easier for analysts to interpret. Instead of “Are you provided with adequate opportunities for professional development?” how about asking “Do you get enough training to do your job?”
- Ask questions for action. To keep your employee survey short, you must limit yourself to those questions that can drive action and improvements. If your company is against giving cash bonuses, don’t ask whether employees are in favor of cash bonuses. One company, whose CEO was against remote working, repeatedly asked how employees felt about remote working. Not surprising, employees thought remote working was a great idea and grew increasingly frustrated when the company did nothing about it.
- Ask the right people. Even in employee surveys, you need to understand your respondents and ask them the right questions. Field personnel have different issues and challenges than home office personnel. New employees are probably different from longer-term employees. Being sensitive to these differences should be reflected in your question logic (“skip- patterns”), the types of responses you include for each question, as well as the language used in the survey.
- Develop good questions. If you ask bad questions, you will get bad data, but you will also annoy your employees, lower your response rate, and discredit the entire existence of your employee research program. So be sure to:
- Ask only one question at a time. Don’t ask people how much they agree with the statement “My manager delegates well and coaches me to better my performance.” If they agree, what do they agree with? Make sure you are only asking about one topic.
- Avoid vague questions that are open to interpretation. Make sure that respondents know what you are talking about so that everyone can answer the same question. Use non-technical language, include descriptions where needed, and avoid acronyms. Don’t use words that are difficult to define or that are subjective.
- Avoid leading or biased language. Political polls are the worst offenders, but businesses can also fall into this trap. Hyperbole has no place in survey research. If you ask, “What do you think of our fantastic new strategy that is going to lead XYC company into the future?” is probably going to get a more positive response than “Please rate our strategy.”
- MECE. ME what? MECE stands for “Mutually exclusive, collectively exhaustive.” That means, the responses to your close-end questions must include a response for every respondent – even if it is “don’t know,” “not applicable,” or “prefer not to answer.” If a respondent does not find a response that reflects how they want to answer the question, they will try to skip the question. If they can’t skip – or if this happens too many times in the same survey – they will simply break off and not complete the survey. In either case, a problem!
While DIY employee surveys may be a necessity, you can at least make sure that bad questions don’t result in questionable data and problem-ridden analysis. Take the time to examine every question, from as many different perspectives as you will have in your response base. By testing your survey extensively, you can identify some if not all the problem areas.
If you would like us to help you with your Employee survey, Contact Us!