Chapter 2 - Research Methods Flashcards
Between-subject vs Within-subject design?
- Between-subject design: participants assigned to either test (independent variable) group, or control (placebo) group
- Within-subject design: participants act as their own control group; measurement taken before and after independent variable
What are the two characteristics of an experiment?
- Random selection and assignment of participants to an;
- Independent variable manipulated by researcher
What is an independent variable?
A variable that an experimenter manipulates
What is a dependent variable?
The variable that the experimenter measures to see whether manipulation has an effect.
ex, school grades, health, conviction rates, etc
AKA “outcome variable”
Internal validity vs external validity
Internal validity: extent to which we can infer cause and effect
- Low in naturalistic, case studies, correlational design
- High in experimental design
External validity: extent to which we can generalize findings to the real-world
- Can be low in experimental design
- High in naturalistic design
Advantages of experiment
- Manipulate variables
- Show cause and effect
- Conclude that any difference in DV was caused by IV
Disadvantages/pitfalls of experiment
- Placebo
- Nocebo
- Hawthorne effect (reactivity)
- Rosenthall effect (experimenter expectancy=biased outcome)
When would one use a correlational instead of an experiment?
- unethical (pepsi in pregnant women)
- impossible
What is positive correlation?
When both variables move in the same direction (UP or DOWN)
What is negative correlation?
When variables move in opposite directions
What is the range of r values?
-1.0 to +1.0
What are the strengths of some correlations?
(in absolute values + or -) 1 perfect 0.8 strong 0.5 moderate 0.3 weak 0 none
Can you infer cause and effect from correlational studies (IV or DV influence each other)?
No. But it’s possible if the correlation is strong enough.
Mean
Average (sum, divided by n). Type of central tendency (Influenced by outliers).
2, 2, 7, 2, 2, 3 = 18/6
Mean=3
Median
Middle score in an ordered data set; 50th percentile. Type of central tendency.
2, 2, 7, 2, 2, 3 = 2, 2, 2, 2, 3, 7
Median=2
Mode
Most frequent score in the data set. Type of central tendency.
2, 2, 7, 2, 2, 3
Mode=2
Low variability vs high variability?
Low var. = values clustered around middle
High var. = values spread out toward extremes
Range
Difference between lowest and highest value. A way of calculating variability.
2, 2, 7, 2, 2, 3
Range=5
Statistical Significance
- Typically a calculation based on on number of values (n) and standard deviation (SD)
- ex, “95% confidence” “p < 0.05”
Central Tendency
Sense of central score of data where the group tends to cluster
Standard Deviation
Average distance from mean. A way of calculating variability.