Strategies for Surveying Employees

	Bookmark This Page Email This Page Format for Printing Cite This Article Submit an Article Read More Articles
	Related Tools & Articles Additional VOC Resources Reflecting the Customers' View Three Simple Questions: How to Focus Your VOC Research
	Six Sigma Quick Poll How often does your company conduct employee surveys? More than once per year Once per year Once every two years Greater than once every two years Has never conducted an employee survey

By P

New from iSixSigma Tying in to Buy-in Research Report Reducing Help Desk Turnaround Time Project Example Six Sigma Certification Research Report

amela Hunter, Ph.D.

Most mid to large companies or organizations can at one point or another, benefit from surveying their employees for the primary purpose of gauging morale. A survey may be needed for a number of reasons, including changes in employee behavior such as high absenteeism or a reduction in output; before or after major adjustments in corporate structure; or to simply obtain baseline data on morale from which to compare changes in future years. Improvements in morale generally indicate a more satisfied work force, which often translates into a more effective and efficient team that plays a critical role in the company's quality and bottom line. Many Six Sigma companies integrate regular employee surveys into their measurement system to capture the "voice of the employees."

Some employers consider such a survey superfluous because they believe that they know their employees and their gripes and do not need a survey to tell them what they already know. Many times employers do know when morale is low and why it is low, but often do not know what to do about it. They may have a superficial view of the reasons for low morale. In other cases, employers may be completely wrong in assessing morale so a survey can serve as a reality check. Quantifiable surveys are often the only way to know for certain what employees as a whole think.

Surveying Employees Is Tricky
Surveying employees, however, is tricky. To obtain the most objective data that best reflects the actual situation, the most critical elements are anonymity and confidentiality.

Anonymity suggests that the answers to a questionnaire cannot be traced back to the actual respondent. Confidentiality, however, means that the answers can be connected to a specific person, but it is difficult to do so and/or is in the hands of an objective third party who will not allow anyone access to the data. For example, names might be eliminated from a list of employees from which a sample is selected, leaving only telephone numbers for a telephone survey. The researcher needs, at least temporarily, to keep track of all responses and pertinent information and maintain a link to the employee if follow-up is required. Instances in this category include whether the interview was completed or whether the respondent needs to call back. If push came to shove, the phone number could probably be matched with the employee and so the survey data is confidential, not anonymous.

Employees need reassurance that their responses to the survey are either anonymous or highly confidential. Employees will not be honest if they think the information can be used against them. No matter how benevolent an employer may be, employee concerns about identity must be addressed or the data will, at best, have limited validity.

Planning Your Survey
An important first step in a successful survey is to hire an outside firm to do the research. If the human resources department, or any other internal department, conducts the research, the data may reflect what the employees think you want to hear rather than the truth. The time, effort, and money spent on research will have been wasted.

Surveying employees is relatively inexpensive. Although exactly how the cost of research is determined varies among survey companies, the most important determinant is usually time; i.e., how long it takes to gather the information. For example, studying men ages 18-29 with incomes greater than $100,000 is much more time consuming than randomly surveying adults in general because the 18-29-year olds in that income bracket are more difficult to locate. Employees are usually easier to survey because, presumably, the company has a current and complete list of employees and all pertinent contact information. The researchers do not have to search for the respondents; therefore, the cost should reflect that time savings.

Employees need reassurance that the researcher, even if from an outside company, is impartial. The outside researcher can show independence from the employer by directing employees to the researcher's web site, not the company web site, for information about the survey. The web site should, for example, provide details about the steps taken to preserve confidentiality, and perhaps offer testimonials from employees of other companies to vouch for the researcher's independence.

It is important to inform employees specifically how the data will be kept anonymous or confidential. If the survey is web-based, provide each employee with a random number in lieu of a name. Employees must understand that tracing their answers back to them will be difficult if not impossible.

Upper management, preferably the head of the company, should enthusiastically endorse the survey. This executive should tell employees that the survey is important to him or her and that management is only interested in the data in its aggregate form.

Allow the outside firm latitude in constructing the survey instrument. Ask the difficult questions. Employees will know whether you really want to evaluate morale or whitewash it simply by the questions that are asked. Morale will be harmed more than helped by research that employees believe is irrelevant.

A variety of questions should be used to collect data on morale. A list of sample questions is included at the end of this article. A survey might, for example, ask respondents to rate their own overall morale. The questionnaire might also include a series of words such as "fair," "disorganized," and "leadership" and ask respondents how well those words describe their unit on a 1-10 scale. A rating of "one" would mean that the word does not at all describe the unit or department and a rating of "ten" would mean that the word describes the unit very well. A researcher should use numerous types of questions to evaluate morale.

Once the data are collected, the researcher should assess morale and, if necessary, provide direction on how to improve it. The third party should also provide the results to the employees perhaps on a web site that, again, is maintained by the outside consultant, not the company.

Surveys of employees can provide essential information to help management improve morale and, as a result, save the company money by lowering absenteeism, minimizing the number of mistakes, and so on. The worst thing management can do is shroud the research in any mystery whatsoever. Employees should feel that they know how the survey process works, the results of the survey, and how management views the results.

Sample Survey
Exact questions would depend on the specific situation facing the company and/or the goal of the survey. Below are examples of questions an employer might consider along with brief comments in italics:

1. Overall, is Company A headed in the right direction or the wrong direction?
   (Start with a general question and move to the specific questions.)
   
   
2. On a scale of one to ten, with one meaning completely dissatisfied and ten meaning completely satisfied, how satisfied are you with your job?
   (The 1 to 10 scale is good to use primarily because people are comfortable with it; it is part of our vernacular in the U.S.)
   
   
3. Which two words best describe Company A?
   (READ LIST. RANDOMIZE ORDER)
   Proud
   Egotistical
   Fair
   Stingy
   Truthful
   Out of touch
   Open
   Closed
   Don't know
   Refused to answer
   (This question begins to get at morale by providing insight into employee attitudes.)
   
   
4. If you were to ask your supervisor a question about the company's reorganization, how likely is it that he/she could really answer your question: Very likely, somewhat likely, somewhat unlikely, or very unlikely?
   
   
5. If you were to ask your supervisor a question about the company's reorganization, how likely is it that he/she would really answer your question: Very likely, somewhat likely, somewhat unlikely, or very unlikely?
   [SKIP IF ANSWER TO QUESTION 4 WAS "SOMEWHAT UNLIKELY" OR "VERY UNLIKELY".]
   (These two questions indicate whether employees consider themselves adequately informed. Another approach is to ask specific questions about the company that the employee should know such as, "Did you receive Memo X?")
   
   
6. How long do you think you will probably work for Company A?
   [DO NOT READ LIST. IF RESPONDENT SAYS, "REST OF MY CAREER," "A LONG TIME," OR A SIMILAR PHRASE, ASK: ABOUT HOW MANY YEARS WOULD THAT BE?]
   Less than a year
   One or two years
   Three or four years
   Five to ten
   Ten to 20
   More than 25
   Other
   Don't know
   Refused to Answer
   (Another indicator of morale; this question suggests whether you have short-terms or long-term employees.)
   
   
7. On a scale of one to ten, with one meaning completely dissatisfied and ten meaning completely satisfied, how satisfied are you with the following? [READ LIST. RANDOMIZE ORDER.]
   Your job
   Training
   Promotions
   Ability to advance
   My work environment
   My supervisor
   My salary
   (This battery could include any number of specific issues. This type of battery of questions is useful because it takes minimal time and provides a lot of useful information.)
   
   
8. How often do you visit Company A's internet or intranet site: At least once a day, every couple of days, every few days, hardly ever, or never?
   (This question starts to get at the best way to communicate with the employee.)