Exam 1: Three Claims, Four Variables Flashcards
Variable vs. Constant
A variable has at least two values
a constant remains the same across all participants in a study.
Measured vs. Manipulated Variables
Measured variables are recorded as they occur
manipulated variables (independent variables) are controlled by researchers
Conceptual vs. Operational Variables
Conceptual variables are abstract (e.g., intelligence)
operational variables are measurable (e.g., IQ score).
Three Types of Claims
Claims include:
frequency claims (describe how common something is)
association claims (show relationships between variables)
causal claims (indicate causation).
Types of claims
Frequency Claims
Describe a rate or level of something (e.g., ‘60% of students like coffee’). They focus on a single measured variable.
Types of claims
Association Claims
Indicate a relationship between two variables. Types: positive, negative, zero, or curvilinear associations.
Types of claims
Causal Claims
One variable causes changes in another
Requires an experiment with controlled variables and random assignment.
Positive Association
Both variables increase or decrease together.
Negative Association
As one variable increases, the other decreases.
Zero Association
No relationship between the two variables.
Curvilinear Association
The relationship between two variables changes at different levels.
Making Predictions
The stronger an association, the more accurate predictions based on that relationship will be.
Criteria for Causal Claims
- The variables must be correlated
- The causal variable must come first.
- No alternative explanations should exist.
Construct Validity
Measures whether a variable is operationalized correctly for accurate assessment.
External Validity
Determines how well findings generalize to other populations, settings, or times.
Statistical Validity
Ensures conclusions are statistically accurate, avoiding Type I (false positives) and Type II (false negatives) errors.
Internal Validity
Ensures no confounding variables affect the relationship between independent and dependent variables.
Evaluating Frequency Claims
Consider construct validity (how well the variable is measured) and external validity (generalizability).
Evaluating Association Claims
Assess construct validity, external validity, and statistical validity (strength of relationship and errors).
Evaluating Causal Claims
Requires:
experimentation
random assignment
high internal validity
to rule out confounds.
Type I Error
False positive: Finding an association when none exists.
Type II Error
False negative: Failing to detect an actual association.
Causal Research Methods
Involves independent variables (manipulated) and dependent variables (measured) to establish cause-effect relationships.
