QUALITY FUNCTION DEPLOYMENT AND QUESTIONNAIRES

Questionnaires are commonly used to collect data for many different reasons, such as to learn people’s opinions, determine customers’ satisfaction, or identify employees’ needs.

Typically, they are used when there is not a more objective way to obtain data.
All questionnaires share two characteristics:

• The data source is a person
• statistical analysis is required to reach conclusions

A relatively large number of completed questionnaires have to be obtained to achieve results of statistical significance, which makes them one of the more costly methods of data collection.

It is, therefore, imperative that the questionnaire’s design be planned carefully.

Typically, three bodies of knowledge are needed when designing a questionnaire:

• the behavioral sciences
• the statistical sciences
• the subject matter being investigated (e.g., politics, food, or technical requirements).

To guarantee the quality of the results, the questionnaire’s implementation and analysis processes must also be considered during its design.

In many cases the use of a questionnaire to obtain technical data from customers is a flop.
There is an incompatibility between the engineering function’s needs and the customers’ capability to respond in technical terms.
As a result, we need to develop a more rigorous approach: a questionnaire methodology using Quality Function Deployment (QFD).

QFD provides a means for translating customers’ needs into the appropriate technical requirements.
The questionnaire methodology should consist of three phases:

• Phase 1: planning. The process starts by defining and analyzing the problem that is related to a product, process, or service.
  The problem has to be defined in an explicit and measurable way so that data can be collected and analyzed later in the process. The purpose of these steps is to break down the problem and identify the important factors, called objectives.
  Objectives are problem elements for which the questionnaire is supposed to find answers.
  Flowcharts, cause-and-effect diagrams, matrix diagrams, or other analysis techniques are typically used to determine objectives.
  The next steps involve questionnaire composition. At this point, the data obtained in the previous steps are translated into a set of unambiguous questions to be used in the survey and in the accompanying interview. The different aspects of the questionnaire, such as the clarity of questions and the response scale used to answer the questions, are also discussed.
  The goal is to rigorously identify relationships between the questionnaire’s stated objectives and the questions being asked.
  Once this is done, an analysis is performed to ensure that the questions are consistent with and cover all of the objectives. Often a revision of the questionnaire is required, in which redundant questions are eliminated and missing questions are added.
• Phase 2: pilot testing. After the planning phase is completed, a pilot test is performed, usually in-house. If necessary, the questionnaire is fine-tuned.
• Phase 3: implementation. In this final phase, the questionnaire and the accompanying interview are administered to the selected population.
  QFD and statistical analyses are then used to check the quality and validity of the results. Since the collected data also contain verbal comments, text analysis is performed.
  The text analysis provides a frequency distribution of terms used in the verbal comments. Those terms that have a high frequency represent important issues (problems) that were not covered by the questionnaire.

QFD makes surveys better

This questionnaire methodology ensures that all relevant parties make a timely contribution to data collection.

Furthermore, its use of QFD provides two major contributions.
First, during the planning phase, it ensures that the objectives have the proper coverage and that excess questions are eliminated.
Second, during the implementation phase, it ensures the quality and consistency of the questionnaire’s results.