Survey Research: The Process

Decision Sciences has over 20 years of experience in designing and deploying surveys for business, industry, government, academia, and non-profit organizations. The following is a brief guide to the process we follow when developing a research project.

Survey Project Planning

Decision Sciences analysts meet with a client "Survey Team" to develop a statement of goals for the survey project. This planning effort is designed to provide the groundwork for customizing the questionnaires and ensuring that the client firm's research goals are met. In this phase we will also look at the various media available for survey dissemination (mail, phone, web-based) and determine the best strategy for getting the most complete and cost-effective responses.

Survey Instrument Development

Decision Sciences survey designers construct a customized questionnaire based on our knowledge of the scholarly literature and our experience with customer and employee satisfaction issues. Questions for the instrument may be drawn from our library of over 500 sample questions or the client may choose to include their own questions.

Sampling Design

Designing the sampling framework is perhaps the most critical point in the survey process. Decision Sciences will work closely with the Survey Team to determine the goals of the research, the levels at which data needs to be disaggregated, and likely response rates. Based on that information, we will determine statistically the required sample size, and whether stratified sampling should be used. The sampling design will also take into account such factors as re-sampling, levels of aggregation, differential response rates, and time periods.

Survey Administration

Customer surveys are fielded by the agreed upon methods – including mailings, phone interviews, email, fax, and web-portals. For employee surveys, the questionnaire is distributed to employees at the work place or accessed through a link on the client's intranet.

Response Processing

Once questionnaires are received by Decision Sciences, they are coded and processed for analysis. Decision Sciences clerical staff reviews all questionnaires for response consistency and data are verified for accuracy. Open-ended responses are transcribed and sorted by comment topic. If this is part of an on-going Tracking Survey , response information is updated into the on-going series and made available to all users of the on-line reporting system (RMS) immediately after each update.

Analysis

Analyses are performed to determine which variables are most highly related to the desired outcomes of customer satisfaction, product quality, employee motivation and performance. Items are typically grouped according to important summary dimensions so that different regions, operational units, managers, departments or divisions of the company may be compared.

Survey Reports and Presentations

Detailed reports and analysis of findings are discussed at a results presentation. The reports themselves may be customized for each manager or department so that they can be compared against the company as a whole, or with other comparable units.

Intervention Plans

Using information from the surveys, companies may develop new policies and intervention plans to improve situations that have been shown to negatively impact customer satisfaction, product quality perceptions, or the workplace environment. Policy options are developed which will have the greatest impact on customer service or employee attitude.

Follow-up Surveys

Companies frequently plan periodic follow-up surveys so that they can assess the effectiveness of new policies and other interventions. Where there are continuous quality improvement programs, on-going reviews of policies or interventions are measured and reassessed. Policies that are ineffective or that have negative unintended consequences can be modified or eliminated. On the other hand, policies that have positive outcomes can be expanded into other areas of the company.