Transcript Reliability and Validity of Researcher
Validity of Researcher-Made Surveys
Evidence of Validity
Evidence of Validity • Patterns of Association
Evidence of Validity • Patterns of Association • Comparing Results from Different Versions of the Same Question
Evidence of Validity • Patterns of Association • Comparing Results from Different Versions of the Same Question • Comparing Responses to Data from Other Sources
Evidence of Validity • Patterns of Association • Comparing Results from Different Versions of the Same Question • Comparing Responses to Data from Other Sources • Asking the Same Question Twice and Comparing Results
Evidence of Validity • Patterns of Association • Comparing Results from Different Versions Reliability • Comparing Responses to Data from Other Sources • Asking the Same Question Twice and Reliability
Evidence of Validity • Patterns of Association
Evidence of Validity • Patterns of Association • Scores from different measures believed to measure similar things should correlate. Scores from different measures believed not to measure similar things should not correlate.
• Responses to items believed to represent the same dimensions or factors should correlate.
Evidence of Validity • Comparing Responses to Data from Other Sources
Evidence of Validity • Comparing Responses to Data from Other Sources • Compare to records. Compare to physical testing. Compare to population estimates.
Face Validity of Survey Questions
Face Validity of Survey Questions • Have a reason for every question you ask.
• Keep questions simple.
• Keep questions precise.
• Avoid leading questions.
• Foresee social desirability.
• Response options should be mutually exclusive and exhaustive.
• Provide temporal frame of reference.
• Use Likert format correctly.