PSY 1001 Chapter 2: Research Methods Part 2 Notes

Theories in Psychology

Theories organize existing knowledge and predict future observations.
Theories supported by substantial evidence become accepted as fact.
Psychological theories often employ abstract concepts or "constructs" to simplify explanations (e.g., memory, anxiety, extraversion).
Operational definitions are used in research to measure constructs. For instance, "memory capacity" can be measured by the number of words recalled from a list.
Some theories are so obvious they are no longer explicitly recognized as theories.
Primacy and Recency Effect: People are more likely to recall items from the beginning and end of a list than from the middle.
Psychological theories are often based on probability rather than certain outcomes, differing from physical theories.
Degradation of Memories: The likelihood of recalling items from a list decreases as time passes.

Observational Research: Objectively observe and record behavior.
Correlational Research: Measure two variables and determine if a relationship exists between them.
Experimental Research: Control or manipulate one variable (independent variable) to determine if it causes changes in another variable (dependent variable).

Acknowledges that individuals are not identical and do not behave identically.
Psychologists aim to describe average or typical behaviors.
Survey Research: Involves asking people about their characteristics, beliefs, and behaviors in certain circumstances.

Descriptive research often focuses on making statements that describe everyone (the population), rather than specific individuals or groups.
Due to the impracticality of surveying everyone, a smaller group (sample) is chosen, with the assumption that the results for the sample apply to the entire population.
Population: The large group of interest.
Sample: The small group that is measured.

If the sample is similar to the population, on average, measurements from the sample can be generalized to the entire population.
Example: A survey question asks, "How would you rate the quality of your Intro Psych lectures (on a scale of 1-10)?" If the average rating is 9.5, the question is whether these results can be generalized to all students registered for the class.

A sample is biased if it is not representative of the population.
Example: Surveying only people who choose to attend lectures may result in a biased sample because they may have more positive attitudes than those who do not attend.
Random Sampling: All members of the population have an equal chance of being selected for the sample.
Random sampling helps create an unbiased sample that is likely to be similar to the population.

A random sample is likely, but not guaranteed, to be similar to the population.
Example: If a class is 50% first-year and 50% second-year students, a random sample of 20 students is likely to have a similar distribution, but could, by chance, be all first-year students.
Statistical analyses can be used to estimate how similar the sample probably is to the population.

Researchers observe people or animals in their normal environments without intervening.
Example: Observing preschoolers on the playground to see how aggressively they behave.
Researchers may conceal themselves or remain visible.

Researchers use coding schemes to categorize and quantify observed behaviors by:
- Operationally defining behaviors of interest.
- Counting the number of times those behaviors were observed.
This produces descriptions of behavior in natural conditions.
Example coding scheme:
- Aggressive behavior = behavior intended to harm another person or damage objects
- behavior that is helpful/cooperative (non-aggressive)

Detailed observation of a single individual, often used when individuals have rare or unusual conditions.
Provides a description of the condition and its development.
Limited usefulness due to the lack of comparative information.
Can help provide direction for future research.

Correlational Research: Identifies relationships between variables.
Allows prediction of one variable's value based on the value of another.
Correlation: Measure two variables (measured variables), compute a correlation.
Produces a correlation coefficient indicating the strength and form of the relationship.

The Pearson product-moment correlation coefficient r quantifies the degree of linear relationship between two variables.
-1 ≤ r ≤ 1
- If r = 1, it is a perfect positive linear relationship.
- If r = -1, it is a perfect negative linear relationship.
- If r = 0, there is no relationship between the variables.

The sign of r (+ or -) indicates a positive (direct) or negative (inverse) relationship.
r^2 (r-squared) indicates the strength of the relationship, with 0 ≤ r^2 ≤ 1.

If two variables (A and B) are correlated:
- Changes in A may cause changes in B.
- Changes in B may cause changes in A.
- Changes in some "third variable" (C) may cause changes in A and B.
- A and B may be related by coincidence.

A study found that people who watched less than an hour of television a day scored higher on a memory test than those who watched more television.
- Does watching less TV cause your memory to improve? What other interpretations are possible?
The same study found that people who read more fiction scored higher on the memory test than those who read less.
- Does reading more fiction cause your memory to improve? What other interpretations are possible?
The number of churches in a city is positively correlated with the number of cars in a city. (Cities with more churches have more cars.)
- Do more churches cause more cars? Do more cars cause more churches? What other factor might cause a city to have more churches and more cars?