Six Sigma Tool Navigator: The Master Guide for Teams

2017-07-07 02:10:07

AKA

Survey Analysis

Classification

Data Collecting (DC)

Tool description

The surveying technique is used to collect large amounts of written or verbal data from customers, employees, or other sources in a very objective, efficient, and organized manner. Applied survey methods such as questionnaires, interviews, observations, and focus groups can provide great flexibility in format and content and usually use data collection instruments that have a high degree of validity and reliability. The type of survey to be used is determined by a team or researcher.

Typical application

To collect data for the purpose of solving problems, verifying customer satisfaction, or comparing performance against organizational goals.

To collect data from a large group of employees to verify job satisfaction, employee morale, or communication problems among organizational units.

To obtain data, general to specific, on customer perception of quality on products or services.

To acquire benchmarking information and "best practices" data.

Problem-solving phase

→	Select and define problem or opportunity
→	Identify and analyze causes or potential change
	Develop and plan possible solutions or change
	Implement and evaluate solution or change
→	Measure and report solution or change results
	Recognize and reward team efforts

Typically used by

1	Research/statistics
	Creativity/innovation
	Engineering
	Project management
	Manufacturing
3	Marketing/sales
	Administration/documentation
4	Servicing/support
2	Customer/quality metrics
5	Change management

links to other tools

before

Data Collection Strategy

Random Numbers Generator

Interview Technique

Questionnaires

Observation

after

Frequency Distribution (FD)

Responses Data Encoding Form

Thematic Content Analysis

Descriptive Statistics

Hypothesis Testing (Chi-Square)

Notes and key points

Comparison of Survey Methods

Advantages	Mailed Questionnaires	Mass Administered Questionnaires	Face-to-Face Interviews	Self-Administered Questionnaires (Face-to-Face)	Telephone Interviews
Low cost	High	Med.	Low	Low	Med.
Speed	Low	High	Low	Low	High
Anonymity	High	Med.	Low	Med.	Low
Immediate response—no time pressure	High	Med.	Low	Med.	Low
Ability to obtain a valid sample from the population^[*]	Low	Low	High	High	Med.
No interviewer bias	High	Med.	Low	Med.	Low
Establishing rapport with respondent	Low	Med.	High	Med.	Med.
Ability to ask complex questions	Low	Med.	High	Med.	Med.
Ability to get full, detailed responses	Low	Low	High	Low	Med.
^[*]Note: Always refer to a sample size determination table.

Step-by-step procedure

STEP 1 A team determines the specific data needed and identifies the data source, and defines the population to be sampled and surveyed.

STEP 2 Team consensus is reached on the type of survey to be used (questionnaires, interviews, observations, focus groups) and the formatting of the data collection instrument.

STEP 3 A survey instrument is developed and field-tested on the basis of valid instrument construction guidelines. See example Development of Customer Satisfaction Surveys.

STEP 4 Refer to a more detailed description of the development and administration of the following types of surveys and related tools in this handbook:

— Questionnaires	— Data collection strategy
— Interview technique	— Sampling methods
— Observation	— Random numbers generator
— Focus group	— Response data encoding form

Example of tool application

Development of Customer Satisfaction Surveys

Survey Instrument Goal: To obtain written responses on a particular product or service Survey Instrument Design Considerations:

Keep the instrument short. Only ask questions that are absolutely necessary.

Restrict the instrument to a single issue, if possible.

Provide a cover letter that explains the purpose of the survey, respondent selection process, and the issue of confidentiality. State the return date.

Write explicit instructions on how to answer the questions for the respondent.

Utilize terms that are clearly understood. Avoid loaded words and phrases, avoid jargon.

Determine how response data will be analyzed, tested, and displayed.

Avoid rank order questions; difficult to analyze.

Encode each response when using rating scales.

Development of questions:
- Start with nonthreatening, easy to answer.
- Use general before specific.
- Use open-ended only if all possible responses are not known.
- Use multiple-choice for increased date volume.
- Phrase in positive, rather than in the negative.
- Avoid researcher bias or prejudice when phrasing.

Pretest the instrument for reliability and content validity.