Levels of reliability in fertility survey data

Area of Science:

• Survey methodology
• Social sciences research
• Data reliability analysis

Background:

• Assessing the accuracy of survey data is crucial for reliable social science research.
• The 1973 Taiwan KAP-4 survey provides a dataset for evaluating response consistency.
• Understanding reliability differences between factual and attitudinal questions is important.

Purpose of the Study:

• To evaluate the reliability of responses from the 1973 Taiwan KAP-4 survey.
• To compare the reliability of factual versus attitudinal questions.
• To determine factors influencing response consistency in surveys.

Main Methods:

• Reinterviewing 286 women from the 1973 Taiwan KAP-4 survey one month later.
• Analyzing consistency of responses for both factual and attitudinal questions.
• Examining how different measures and data aggregation levels affect reliability estimates.

Main Results:

• Overall response reliability at both aggregate and individual levels is reasonably good.
• Factual data demonstrated higher reliability than attitudinal data.
• Reliability of factual questions varied based on topic salience for respondents.
• Reliability estimates were influenced by the chosen measures and response distribution skewness.

Conclusions:

• Survey response reliability is generally adequate but varies significantly.
• Factual questions offer more dependable data than attitudinal ones in this survey.
• Topic importance and measurement methods are key factors in survey data accuracy.