Organizational Surveys

Organizational surveys are also known as employee opinion surveys or employee attitude surveys. Most experts prefer to call them organizational surveys to clarify that the sponsor and user of such surveys is almost always the organization. Further, the people asked to complete such surveys may be employees at any or all levels, including top executives. Recent estimates conclude that about 75% of all medium- to large-sized firms conduct organizational surveys, typically every year or two. Surveys are also used extensively within the United States federal government, including the armed forces.

Organizational Surveys Purpose and History

The size and content of such surveys may vary widely, reflecting the different purposes to which organizational surveys are put. These purposes range along a continuum, so surveys can be seen as tools for assessment and change. Historically, surveys have been used for assessment, much like taking a broad-scale annual medical examination to see how we are doing. In recent decades the emphasis has been more on stimulating and measuring change in specific areas of strategic value, such as product quality, work-life human resource initiatives, or customer satisfaction.

The survey is a popular methodology for conducting research in many areas of industrial and organizational psychology. Catherine Higgs and Steve Ashworth noted that surveys can be used for several types of research. In early stages they can be merely exploratory and later move on to be fully descriptive of the phenomena measured. At the highest level, they can be used to test causal relationships. It is quite common to see journal articles that have used surveys to collect data. Practitioners often learn or borrow survey methodology from allied fields such as sociology.

The typical content of the organizational surveys has changed since the early popularity of employee surveys in the 1930s and 1940s. That was an era of concern for employee morale and emotional adjustment, often as desirable ends in themselves and sometimes to prevent unhappiness that might cause unionization. Typically, survey questions asked about the individual employee’s contentment with different aspects of work, management, and pay. They even asked about environmental issues such as parking lots, cafeterias, and lighting. Both interviews and questionnaires were used.

In the 1950s and 1960s the emphasis shifted to the individual’s job satisfaction because of presumed links of satisfaction to better organizational productivity. Typical surveys were based on paper-and-pencil questionnaires. In the 1970s and 1980s attempts were made to link employee satisfaction to organizational outcomes like turnover, absenteeism, and stress. The worker’s job level and part of the organization were seen as important moderators. The technology for doing surveys also evolved to the use of self-administered, standardized, scannable forms.

Since the early 1990s an extraordinarily different view of surveys has taken hold in most organizations. Behind the change is a set of assumptions linking employee opinions and perceptions to the achievement of strategic organizational goals. In service-dominated industries, employee behaviors are now seen as directly influencing customers’ reactions and loyalties and thus spilling over to bottom-line measures like financial growth, profit, and product success. Set against an increasingly competitive global marketplace, the emphasis in surveys is to capture employee views, perceptions, and reports of how their organizations work, with the aim of achieving more productive teams, better quality products and services, and more satisfied customers.

This view of surveys has led to several conceptual models, some with names like the service-profit chain. Still others are known as linkage research models. They help describe to organization management just why the surveys measure the concepts they do, and show how they are linked to important outcomes. Several recent studies seem to support this way of viewing and using survey results. The relationship between employee views and organizational outcomes at the unit level (rather than the individual level) have proved to be more powerful than previously believed by a few meta-analytic studies, which distill the relationships found in many studies.

Organizational Surveys Methodology

Survey questions may cover a wide variety of issues. Closed-ended questions are typically written as Likert-type items, to be answered on a five-point scale of satisfaction or agreement.

An example of the first type would be “How satisfied are you with the recognition you receive for doing a good job?” Possible answers:

  1. Very satisfied;
  2. Satisfied;
  3. Neither satisfied nor dissatisfied;
  4. Dissatisfied;
  5. Very dissatisfied.

An example of the agreement type would be “I like the kind of work I do.” Possible answers:

  1. Strongly agree;
  2. Agree;
  3. Neither agree nor disagree;
  4. Disagree;
  5. Strongly disagree.

Most surveys also include one or more open-ended questions that ask respondents to write in their response. Questions may be very general, such as, “Any other comments?” Or they may be very specific, such as, “What kind of training would help you to be more effective?” To encourage frank and honest responses to all questions, strict confidentiality is almost always promised by the survey sponsor.

In the last decade, organizational surveys have been administered largely by computers, using e-mail and the Internet. This has made astonishing changes in the administration, collection, and use of survey data. Surveys can now be sent electronically to eligible samples, avoiding the physical effort and expense of postal or other distribution systems. Surveys are cheaper and much more flexible once the infrastructure is available. Although printing a large survey might take weeks, an electronic version can be changed at the last minute (and even during a survey if a major crisis has occurred). These surveys can also use branching techniques, so respondents answering certain questions unfavorably can be offered more detailed follow-up items to help in diagnosis. Write-in comments are keyed in by respondents, avoiding the laborious transcription needed in paper-and-pencil surveys.

In most ways, the electronic version of organizational surveys is quite superior to earlier versions. Careful studies show no distortion of replies versus paper-and-pencil versions, except that write-in comments are typically twice as long. Still, computer programs make write-in comments much easier to analyze and report than in the past. In general, reporting survey data has been shortened from many weeks to a few days. Reports to managers can also be made via computer, giving managers a chance to receive reports sooner and in more flexible formats.

Organizational Surveys Response Rates

However, electronic surveys may create problems because surveys now seem quite easy to do. Many groups in a firm launch their own internal surveys, often of dubious quality. Along with this are employee cries of oversurveying, and a decline in response rates. In the past it was common to survey all of a firm’s population to do a census survey. Recently, many firms have shifted to doing more frequent surveys, or pulse surveys, often of small samples of respondents.

Typical response rates are hard to pin down with certainty. It is believed that census surveys of employees typically get response rates of anywhere from 30% to 95%, averaging about 65%. Sample surveys seem to get 10% to 15% lower response rates. Poor response rates undermine the credibility of survey results. Some recent research, however, suggests that nonrespondents are mostly people who have other priorities and are not actively opposed to the survey or its topics. Still, it would be wise to weight any subgroups that respond at higher or lower rates than typical, to properly represent each subgroup in the firm. Of course, survey researchers should always compare a sample’s demographics with the known population demographics to be sure the sample is representative.

Most experts believe that the size and content of a survey questionnaire will influence the response rates. Surveys with clear, well-written questions on topics that are obviously important to the individual and organization will gain more participation. Respondents will be turned off by surveys they think are overly long. In prior decades, it was not unusual for paper-and-pencil surveys to have more than 200 questions. Recent electronic surveys typically have from 40 to 75 items and can be completed in 10 to 20 minutes. This, too, reflects the competitive, fast-moving business climate in many firms.

Normative Data

Management getting survey reports often ask how they compare with other firms. Different consortiums of companies have been born from this desire. The granddaddy of survey consortiums is the Mayflower Group (www.mayflowergroup.org), a group of roughly 40 large firms that ask the same two dozen core items in their respective surveys and share the normative data with other member firms under strict confidentiality. There is also the Information Technology Survey Group (www.itsg.org), made up of 18 high-technology firms, that operates in a similar way. The representatives of the consortium’s member firms meet twice a year and share what they believe are best practices in doing organizational surveys. In addition, several survey vendors offer data norms based on their client data sets, or based on specially collected national data sets.

Other types of norm data are those that represent cultural or national differences. With many large firms now truly global organizations, some international data is available for norm comparisons. Basic advances in the social sciences are also helped through such data. Three decades ago the Dutch organizational scholar Geert Hofstede used the international survey data from the IBM Corporation to lay out several cross-cultural dimensions, such as the tendencies of people in different countries to be oriented toward the group, or collective (as in Asia), or toward the individual (as in the United States).

Organizational Development and Action Taking

The most important outcome of surveys is meaningful and responsive action. This is also the most elusive aspect of the organizational survey process. But experts in organizational development, working with survey researchers, have developed some excellent techniques. They recognized that survey data can be energizing and motivational. Naturally, meaningful action requires a supportive top management that is knowledgeable about the overall process. They must be advocates and champions of the organizational survey. Survey practitioners responsible for doing the organization’s survey must educate all levels of management; they must provide the training and infrastructure for data to be collected, analyzed, and reported and then track the actions taken.

Some experts favor reporting survey data first to top management, then letting the data cascade down to lower levels, with top management acting as role models for how to discuss and act on the data. Other experts prefer having the data bubble up, with lower levels seeing it first and then reporting their findings and action plans to higher levels. Over the years detailed protocols have been developed to feed back the survey results to respondents; and this is seen as a critical step to good survey practice. Recent research, however, has made it clear that action taking, not data feedback, is the critical ingredient for success. In fact, providing employees with survey data feedback and no action causes more unfavorable consequences than giving no feedback at all.

Companies that have described their experiences make it clear that the critical factor is to ensure that managers must act. It may even be best to focus on only one or two high priority areas to work on. Many firms use survey results, and improvements or declines, as the basis for performance appraisals, incentive bonuses, and promotions. If the topics measured in the survey are truly important to the organization, rewards and punishment for their achievement seem quite appropriate.

Organizational surveys have been used for many purposes in recent years. These include topics as different as managing the progress of mergers and acquisitions, improving a climate of diversity, reducing employee turnover, reinvigorating an organization after a business turndown, and coordinating practices in a large global organization. Surveys have also been modified to provide multisource feedback (360-degree feedback) for management development purposes. The organizational survey is a powerful tool for many purposes and seems destined for continuing use and influence.

References:

  1. Church, A. H., & Waclawski, J. (2001). Designing and using organizational surveys: A seven step process. San Francisco: Jossey-Bass.
  2. Kraut, A. I. (Ed.). (1996). Organizational surveys: Tools for assessment and change. San Francisco: Jossey-Bass.
  3. Kraut, A. I. (Ed.). (2006). Getting action from organizational surveys: New concepts, methods and applications. San Francisco: Jossey-Bass.