Chapter 4 Survey

Chapter 4 Survey

Survey

Surveys ask respondents for information using oral or written questioning. The benefit includes quick, inexpensive, efficient, accurate, and flexible. However, the disadvantage of it will be poor design and improper execution.'

Total error

Random sampling error: A statistical fluctuation that occurs because of chance variation in the basics selected for the sample.

Sample bias - when the results of a sample show a continual trend to move away in one direction from the true value of the population factor.

System error (bias): Systematic error results from some imperfect aspect of the research design or from a mistake in the execution of the research.

Respondent bias

Some respondent action or inaction create the sample bias. It includes non-respondents error and response bias.

Non response error

People who refuse to cooperate

Not-at-homes/refusals. Self-selection bias. Over represents extreme positions. Under represents those who are unresponsive.

Responses bias

A bias that occurs when respondents tend to answer questions with a certain angle that consciously or unconsciously misrepresents the truth.

Purposely misrepresent the true

Purposely give false answers to avoid embarrassment.

Unaware misrepresent

May occur from question format, question content, wording.

Admistrative error

Improper admin of the research task

Types of Response Bias

1. Acquiescence Bias

A category of response bias that results because some individuals tend to agree with all questions or to concur with a particular position.

2. Extremity Bias

A category of response bias that results because some individuals tend to use extremes when responding to questions.

3 Interviewer Bias

A response bias that occurs because the existence of the interviewer influences answers.

4. Auspices Bias

Bias in the responses of subjects caused by the respondents being influenced by the organization conducting the study.

5. Social Desirability Bias

As in responses caused by respondents’ desire, either conscious or unconscious, to gain status or appear in a different social role.

Type of admistrative error

1.      Interviewer cheating - filling in fake answers or falsifying interviews.

2.      Data processing error - incorrect data entry, computer programming, or other procedural errors during the analysis stage.

3.      Sample selection error - improper sample design or sampling procedure execution.

4.      Interviewer error - field mistakes: mis-recording of data.

Reduction of sampling and systematic error

Sampling Error

      Can only be reduced by taking larger samples.

      Can be estimated by confidence levels.

Systematic Error

1.      Difficult to estimate.

2.      Can be reduced by:

a.      Good survey design

2. Good sampling design
3. Training of fieldworkers
4. Care in data processing

Classifying Survey Research Methods

1. Time Period for Surveys

Cross-Sectional (representative) Study

      A study in which various segments of a population are sampled

      Data are collected at a single moment in time

Longitudinal Study

      A survey of respondents at different times, thus allowing analysis of changes over time.

      Tracking study - compare trends and identify changes

–        Consumer satisfaction

–        Purchase behaviour: consumer diary sheets, single source data research

Structured and Disguised Questions

How much structure or standard in the questionnaire?

1.      Structured questions (descriptive survey research)

Closed ended questions with choices

2.      Unstructured questions (exploratory research)

        Open ended questions

3.      Disguised questions (exploratory research)

Indirect questions for sensitive issues.  Eg. Using projective techniques.

4.      Undisguised questions (descriptive survey research)

Assumes respondent is willing to reveal information.  Eg. Do you have dandruff problems?

                              

Basic Methods of Communication with Respondents

1.      Personal Interviews.(e.g. Door to door, shopping mall intercepts.)

2.      Telephone Interviews

3.      Mail Survey

4.      Internet Research

How to Increase Response Rates for Mail Surveys

1.      Effective cover letter

2.      Money helps : As a token of appreciation and for a charity

3.      Stimulate respondents’ interest with interesting questions such as reminder letters

4.      Advanced notification and sponsorship by a well-known and prestigious institution

The advantage and disadvantage of using survey

1.      Advantage: High confident question by using mail survey.

2.      Disadvantage : To prevent interview bias, we need to scarify speed of data.

Selected Questions to Determine the Appropriate Technique:

1.      Is the assistance of an interviewer necessary?

1. Yes – Use personal interviews

2.      Are respondents interested in the issues being investigated?

1. Yes – Use faster methods

3.      Will cooperation be easily attained?

1. Yes – Use faster methods while maintaining high response rates
3. How quickly is the information needed?
1. Fast – Use faster methods
4. Will the study require a long and complex questionnaire?
1. Yes – Use personal interviews
5. How large is the budget?
1. Small – Use mail or Internet

Pretesting

A trial run with a group of respondents to solve fundamental problems in the design of a questionnaire.

1.      Unclear instructions

2.      Spelling mistakes

3.      Ambiguous terms

4.      Question wording and language

5.      Confirmation of asking the right questions

6.      Estimate time needed to complete