PSY 301 Lecture Notes - Lecture 3: Zumba, Sleep Deprivation, Internal Validity
17 January 2018
Psy 301
Quiz:
1. Three necessary criteria for causal claims are:
○ Covariance, temporal precedence, and internal validity
2. An association claim has:
○ Two or more measured variables
3. Which of the following headlines is a frequency claim?
○ Exercise: 45% of you shake your booty in zumba
Class Notes
Variables, Claims, and Validities
● “Most students don’t know when news is fake, Stanford study finds”
○ Wall Street Journal, Nov 21, 2016
○ This is an association claim, frequency claim
What counts as good evidence?
● As a research consumer, you need to know what type of evidence is convincing
Variables
● A variable is something that varies or can take on different values
○ Levels are the values that a variable takes on
● Variables can be measured
○ Levels are observed and recorded
● Variables can be manipulated
○ Researcher controls assignment of levels
● A constant is a potential variable that is kept as a single level
Variables in Developmental Psychology
● Baby makes way down runway towards parent with a toy
● Measuring the frames of what babies are looking at
● Variables: toy height (low, middle, high), whether babies were walking or crawling
(posture variable), proportion of frames (0-1)
● Age of babies was kept constant, to keep similar developmental model in mind when
making conclusions
From conceptual variable to operational definition
● When we talk about research, we speak in terms of conceptual variables
○ Abstract concepts (e.g. “happiness,” “intelligence”), also called constructs
● To test hypotheses, researchers need to create operational definitions of variables
○ Operationalize the variables
● Ex: How you could operationalize happiness
○ Self-report scale, number of positive tweets
● Ex: How could you operationalize “school achievement”
○ Look at GPA
Three types of claims
find more resources at oneclass.com
find more resources at oneclass.com
● Frequency claims
○ Involve one variable
○ Express rate pr degree
○ Ex: 80% of college students have been depressed during the last year
■ Can be more vague like “most” or “the majority of” students
● Association claims
○ Involve two (or more) measured variables
○ Assert that the value of one variable varies systematically with the value of
another variable
○ Ex: heavy cell phone use tied to poor attention
○ Types of associations: positive, negative, curvilinear, zero (slope of graph)
● Causal Claims
○ Involve two (or more) variables
○ Argue that one variable is responsible for changing another variable
○ Ex: sleep deprivation makes you crabby
Criteria for making a causal claim
● We infer that event X is a cause of event Y if:
○ X and Y are correlated, or associated (covariance)
○ X comes before Y (temporal precedence)
○ WE can eliminate other possible causes that may be confounded with X (internal
validity)
Language of Claims
● Verb phrases that distinguish association and causal claims
● Association claim verbs: is linked to, is at higher risk for, is associated with, is correlated
with, prefers, may predict, is tied to, are more/less likely to, goes with
● Causal claim verbs: causes, affects, may curb, exacerbates, changes, may lead to,
makes, sometimes makes, hurts, promotes, reduces, prevents, distracts, fights,
worsens, increases, trims, adds
Iclicker Q:
● “ADHD drugs not linked to false memory susceptibility”
○ This is an association claim
● “Two out of five Americans say they worry everyday”
○ Frequency claim
● “Regular exposure to nature changes memory”
○ Causal claim
● “70% of runners are chronically happy”
○ Frequency claim
● “Inconsistent responses from parents are bad for children’s learning”
○ Causal claim (“bad for” is saying that it has an effect)
● “Antidepressant use during pregnancy linked to autism”
○ Association claim (positive association)
How do we evaluate the 3 types of claims?
● A valid claim is reasonable, justifiable, and accurate
○ In other words, the conclusion is appropriate
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Quiz: three necessary criteria for causal claims are: Covariance, temporal precedence, and internal validity: an association claim has: Exercise: 45% of you shake your booty in zumba. Most students don"t know when news is fake, stanford study finds . This is an association claim, frequency claim. As a research consumer, you need to know what type of evidence is convincing. A variable is something that varies or can take on different values. Levels are the values that a variable takes on. A constant is a potential variable that is kept as a single level. Baby makes way down runway towards parent with a toy. Measuring the frames of what babies are looking at. Variables: toy height (low, middle, high), whether babies were walking or crawling (posture variable), proportion of frames (0-1) Age of babies was kept constant, to keep similar developmental model in mind when making conclusions. When we talk about research, we speak in terms of conceptual variables.