• Psychometrics: area of psychological research that involves the development, validation, and refinement of surveys and tests for measuring psychological constructs
  • Construct validity: indicated that a survey measures the behavior it is intended to measure
  • Social desirability bias: bias created in survey responses from respondents’ desire to be viewed more favorably by others o Typically over­reports of “positive” behaviors and under­reporting of “negative” behaviors
  • Cronbach’s Alpha: method od testing scores’ internal consistency that indicates the average correlation between scores on all pairs of items on a survey o Calculates the overall correlation I. Chapter Summary:
  • Most surveys are used to answer descriptive or predictive research questions in correlational studies. o However, if a survey is used to measure behavior in an experiment, then it can be used to answer causal research questions, as well.
  • Surveys are constructed with questions or items that address behaviors and attitudes on specific topics o Open­ended or closed­ended items may be included
    • The goal is to construct the survey such that its validity and reliability are maximized
    • Validity and reliability are typically evaluated for new surveys that researchers conduct
  • Testing reliability:
    • Correlations between scores on different items of the survey can be measured o Correlations between scores on the same survey taken at different times can be measured
    • Strong, positive correlations in these cases would indicate reliable scores on the survey
  • Surveys can be administered in paper booklets to groups or individuals, as oral questions posed to groups or individuals, or on the web to individuals
  1. Questions:
    1. Survey items that can be completed with any response chosen by the participant are called __________ questions, whereas survey items that must be completed with one of the response options provided by the researcher are called __________ questions Open ended; close ended
    2. The scores are indicative of the psychological construct the survey was designed to measure
  • Construct validity
  1. The scores are similar each time a participant takes the survey
  • Test­retest reliability
  1. Participants are not restricted to the types of responses they can make to survey items
  • Open­ended scale
  1. Scores at later testing are affected by participant’s having taken the survey or test previously
  • Testing effects
  1. A measure of reliability based on the average correlations between pairs of items on a survey
  • Cronbach’s alpha