Lecture 10: Measurement

Constructs

• Variables that cannot be observed directly

  • Eg., traits, emotions, attitudes, abilities

• Defining your variables: Conceptually

• What do you think this construct is/means? How would you define it?

• Defining your variables: Operationally

• How will this construct be measured or manipulated?

• Types of measurement: Self-report measure

• Interviews, or questionnaire 

  • Downside: is that some people don't have insight into their thoughts or feelings. People want to present themselves differently so they're more likable with their answers

• Types of measurement: Behavioural measure (video recordings of people watching)

• Behavioural measures, observations of behaviour, could be naturally occurring (nail biting) or lab-induced

• Types of measurement: Physiological measures

• Assessments of bodily states

  • Eg., brain imaging (fMRI, PET), heart rate (have ppl wear heart monitors to record their heart rate)

How to choose the types of measurement you will use

• Research tradition: How was this variable measured in previous studies?

• Methodological advances: New measurement options can become available with new technology

• Feasibility: Resource limitations (eg., time, money) may constrain your choice

• People pick self-report because its the cheapest option

Reliability

Does your measurement consistently measure the same thing. Is it accurate

How to test the reliability of a self-report measure?

• Test-retest reliability

• Parallel-forms reliability (if applicable)

• Internal consistency

• Test-retest reliability

• Test it with test-retest correlation

  • The same test is given twice, with some time in between

  • Good for stable qualities (eg, personality), not temporary states (eg, mood)

• Parallel-forms reliability

• Different forms/versions of the same test are used

• Internal consistency

• Test it with split-half correlation

  • The top half of the questionnaire is compared to the bottom half

  • If theres a high correlation between the 2 it means the items are consistent with each other.

test internal consistency with cronbach’s alpha

• Measures how strongly all items on a test are intercorrelated.

• Considers all items at once, not just two halves

Folk psychology

everyday, intuitive way that people understand, explain, and predict human thoughts, feelings, and behavior

Heuristics

mental shortcuts or quick rules of thumb that people use to make decisions or solve problems efficiently.

Confirmation bias

Tendency to seek, interpret, and remember information in a way that supports your existing beliefs, while ignoring or discounting evidence that contradicts them

Skepticism

Pausing to consider alternatives and to search for evidence—especially systematically collected empirical evidence—when there is enough at stake to justify doing so

Tolerance for uncertainty

There is often not enough evidence to fully evaluate a belief or claim, that's why scientists accept that there are many things that they simply do not know