Flashcards in Correlational designs Deck (18):
What is an operational confounds?
A measure that measures multiple things (things other than what is intended to be measured)
When do spurious correlations occur?
Spurious correlations occur when two things happen to co-vary but are not related in any obvious way.
J.S. Mill's 3 criteria to infer causation
2. Temporal sequence
3. Eliminate alternative explanations (third variable problem)
What are person confounds?
Individual differences that tend to co-vary; e.g. depression and feelings of loneliness
What are environmental confounds?
Situations that cause multiple differences
What are the methodological sources of confounds?
What do correlational studies look at?
The relationship between measured variables
What can and can't correlational studies establish?
Can establish co-variation
Can establish temporal sequence effectively
Cannot eliminate alternate explanations effectively
All issues re: descriptive studies apply:
- Measurement/testing effects
- Question wording
- Random sampling needed to ensure external validity
What is done in a correlational design?
Not manipulating variables
Multiple dependent variables
- Essentially like a descriptive study, except we are using it at the same time
- In its basic form: 2 DVs, continuous, and for each subject we measure them on both DVs
Why correlational designs?
Some things cannot be manipulated
Some things lose external validity in a drastic way once you use an experimental design
You can make predictions (regression lines)
Hamermesh & Abrevaya?
Study on beauty and happiness
An increase in beauty increases happiness/satisfaction
What are the two forms of criterion validity?
Concurrent validity and predictive validity
What is concurrent validity?
How well can current events be predicted based on the available data
What is predictive validity?
How well can future events be predicted based on the available data
What is epidemiology?
Epidemiology refers to the scientific study of the causes of disease. But when clinical psychologists speak of epidemiological research, they are typically referring to descriptive studies that focus primarily on the prevalence of different psychological disorders within meaningful, well-defined populations
What makes a case study scientific?
- Unless the goal of accounts is the development or refinement of theories of human behavior, they should not be considered case studies
- Scientific analyses often explain things in terms of established scientific principles - those that presumably apply to everyone, given the right circumstances
What are drawbacks of case studies?
- They do not always lend themselves well to the use of operational definitions or statistical analyses
- Do not support cause-and-effect explanations