Validity in research

Validity in research

1. CONSTRUCT validity

   • define the construct
     • define how you observe it
       • define how you measure it
   • abstract variable
     • psychological state
       • e.g. motivation

2. INTERNAL validity

   • does my research investigate what it claims?
   • interventionist and causal-comparative studies
     • Does the IV really influence DV or some other factors do?
       • ! Establish the causal relationship between variables by...
         • ...excluding any other intervening factors
           • ...by keeping other factors stable
       • ! Isolate variables
         • Establish 2 groups: control & experimental
           • pre-test & post-test and measure progress
           • both groups should have similar characteristics
   • qualitative studies
     • Survey: Do the answers really reflect respondents thoughts?
       • No leading questions!
         • Anonymous!
     • Observation: careful with Observer's paradox!

3. EXTERNAL validity

   • can the results of my research be generalized?
     • participants
       • sample should be representative of the population
         • sampling strategies
           • probability sampling strategies
             • Simple random sampling
               • when you do not have much information about the population
                 • when the population is uniform
           • non-probability sampling strategies
             • when pressed for time LOW VALIDITY
               • Convenience sampling:
                 • Purposive sampling
                   • Snowball smapling
     • statistical inference