Week 2, Day 4: Research Methods
1. Correlation Research
a. Correlation: The degree to which two variables are related.
b. Positive Correlation: As one variable increases, the other variable increases.
c. Negative Correlation: As one variable increases, the other variable decreases.
d. Illusory Correlation: When there appears to be a correlation, but the two variables don’t actually influence each other.
i. Example: In June, July, and August, the number of ice cream
sales goes up, as does the number of murders.
In actuality, these are both most likely caused due to the
increase in weather. When the weather is hot, people like to eat cold things. Additionally, hot weather can make people more
aggressive and angry.
ii. Example: Pirate shortages cause Global Warming.
Obviously, the fact that pirates are becoming fewer has nothing to do with the Earth getting hotter.
e. Pros of Correlation Research:
i. Helps you to predict behaviors.
ii. It’s a good stepping-stone for research.
f. Cons of Correlation Research:
i. It can lead you astray.
ii. You cannot infer causality.
2. Descriptive Research
a. Naturalistic Observation: A scientist, or team of scientists, who observe organisms in their natural environment. Don't forget about the age old question of What ability does microscope give a person's naked eye?
b. Case Study: When a scientist, or team of scientists, ‘zoom in’ and study only one individual.
c. Validity: Does this measure what it’s supposed to measure?
i. External Validity: How true to real life circumstances is this study designed?
ii. Internal Validity: Ability to control variables.
1. Naturalistic Observation is high in External Validity and
low in Internal Validity.
2. Case Studies are high in Internal Validity and low in
d. Generalizability: To what degree does this apply to the population? e. Pros of Descriptive Research:
i. Able to study unusual circumstances or phenomenon.
ii. Able to study scenarios you can’t typically study in a laboratory. f. Cons of Descriptive Research:
ii. Subjective to bias.
iii. Cannot infer causality.
3. Experimental Design
a. Must have:
i. Independent Variable: A thing that you can change.
Week 2, Day 4: Research Methods
ii. Dependent Variable: Thing that you measure.
1. Example: In a study of the effects of sleep on the scores
of exams, hours of sleep would be the independent If you want to learn more check out In what way is homology an evidence of evolution?
We also discuss several other topics like When does founder effect happen?
variable and the test scores would be the dependent
iii. Control Group: A group that you can use in comparison to your independent variable.
iv. Random Assignment (whenever possible): A group chosen by random. By doing this, you should be able to cancel out any pre existing variables that may influence your study. Additionally, it helps psychologist to understand the Placebo Effect (when
someone gets an effect solely because it is the effect they
expected to get.).
b. Experiementor Bias: When the person performing the study is expecting a certain outcome and that expectation influences their interpretation of the results.
c. Double Blinded: When both the experimentor and the subject(s) are blind to the treatment conditions.
This is preferred in experiments because it helps to eliminate the Placebo Effect and Experimenter Bias. Don't forget about the age old question of What do you call the ability of an organism to survive and contribute its genes to the next generation?
We also discuss several other topics like Which anatomy refers to the structural changes caused by a disease?
Don't forget about the age old question of Who is the proponent of law of definite proportions?
d. Pros of Experimental Design:
i. The only research method of which you CAN infer causality.
ii. High internal validity.
e. Cons of Experimental Design:
i. Difficult because of ethics. (Can’t test the effects of cocaine on the brain by assigning one group to do cocaine for three years, for example).
ii. Low external validity.