AP Psychology - Methods

Hindsight Bias

Hindsight Bias

The tendency for someone to think they knew a conclusion from research before it was released, even though they did not.

Applied Research

Research to solve practical problems that have real-world applications.

Basic Research

Research into areas that are of interest to psychologists but that do not have real world applications.

Hypothesis

An assertion that there is a relationship between two variables that someone makes as a prelude to performing research.

Can you prove a hypothesis?

No, you cannot prove a hypothesis. Research can only support a hypothesis, but it is impossible to fully assert that a hypothesis is true. If you are doing spaced repetition or learning, note that this is a question.

Independent Variable

The variable that is changed or modified.

Dependent Variable

The variable that changes as a result of changes to the independent variable,

Theory

A claim that is backed by evidence from research (although admittedly some examples of this in psychology lack research).

Operational Definition

A definition of something in terms of how it will be used in the procedure and how it will be measured.

Validity

How well the research measures what the researcher set up to measure.

Reliability

How consistent the results are, even when done by different researchers (with the same procedure)

Participants

Subjects in research.

Sampling

The process wherein participants are selected.

Sample

The group of participants

Population

Anyone who could be selected to be in the sample for the study.

Representative Sample

A sample that is representative of the population the participants were selected from.

Random Sampling or Random Selection

The process of selecting participants wherein everyone in a population has an equal chance of being selected.

Stratified Sampling

The process of selecting participants in which participants are randomly selected, but the number of people of different demographics are standardized to prevent inequality. For example, if a population is 1/3 White, 1/3 Black, and 1/3 Asian, then a study of that population with participants picked this way should have 1/3 of the participants be of each demographic, and the participants selected randomly for each demographic.

Laboratory Experiments

Experiments conducted in a lab.

Field Experiments

Experiments conducted in the actual world, in uncontrolled environments.

Experiment

One type of research that determines cause-and-effect relationships.

Confounding Variables

Any factors that may affect the dependent variable that are not the independent variable.

Participant-Relevant Confounding Variables

Variables that affect the assignment of participants. For example, if someone picks their group, they may pick their group based on their personality, and if the researcher picks their group, they may either have biases or they may unconsciously assign people to certain groups based on things like when they arrived or where they were waiting for the research to start.

Situation-Relevant Confounding Variables

Variables that impact the situations different groups are put in. Participants should be put in equivalent situations except for their independent variable to minimize this.

Assignment

The process of putting participants into groups.

Random Assignment

A type of assignment in which each participant has an equal chance of being placed in any group.

Controls

Ways that researchers minimize the impact of confounding variables.

Group Matching

A process of assignment in which you divide people into groups and then randomly assign them roles. For example, you could create two different groups based on gender before assigning people to groups.

Experimenter Bias

The unconscious tendency for a researcher to treat members of the control and experimental groups differently to prove their hypothesis.

Double-Blind Procedure

When neither the researcher nor participants can affect the study, most frequently done by not telling the participants of their group and having someone else who doesn’t know their groups interact with the participants.

Single-Blind Procedure

When only the participants are unaware of their group.

Response Bias

When a person asks biased questions because of expectations.

Participant Bias

A participant in a study unconsciously behaving the way the researcher expects them to.

Social Desirability

A response bias wherein people give answers that reflect well upon them.

Experimental Group

The group receiving the independent variable.

Control Group

The group that does not receive the independent variable and is compared with the group that does.

Hawthorne Effect

An effect wherein people behave differently when they know they are being studied.

Placebo Method

A method of studying where both groups are lead to believe they’re the experimental group to account for any psychosomatic responses.

Counterbalancing

A method of research in which all groups are both the experimental and control groups, removing confounding variables.

Positive correlations

The presence of when thing predicts the presence of another.

Negative correlations

The presence of one thing predicts the absence of another.

Quasi-Experimental (Ex Post Factor) Study

A study where most aspects are controlled, but some are left as confounding variables.

Survey Method

A method of research using surveys where there is neither an independent nor a dependent variable and instead asks people questions to determine a relationship between two variables.

Response Rate

The rate at which people respond to requests to fill in surveys. This is usually quite low.

Naturalistic Observation

A method of research in which someone watches behavior without interacting or changing variables.

Case Study Method

A method of studying where a specific person or group of people (with a condition) are studied over a long period of time to get a full picture of their development over time.

Descriptive Statistics

Statistics that describe a type of data.

Frequency Distribution

A distribution of a statistic showing its varying levels of frequency in a population. This is frequently used to make graphs, one example being bell curves.

Mean

A measure of central tendency that is usually referred to as the average of a data set, found by adding all of the values together and dividing the sum by the number of values.

Median

The central value in a distribution, found by counting to the middle value of a data set in chronological or descending order.

Mode

The value that appears most frequently in a distribution.

Extremes Scores or Outliers

A value or set of values that is significantly different from other values in the distribution. This can affect the mean and skew distributions.

Positive Skew

Distribution created when there are very high outliers.

Negative Skew

Distribution created when there are very low outliers.

Range

The distance between the highest and lowest values in the distribution.

Standard Deviation

Average deviation of the mean.

Variance

Average squared deviation of the mean.

Z Score

A measure of distance from the mean in standard deviation. For example, if someone got 115 on an IQ test, the would have a measure of 1, because 115 is 1 standard deviation away from the mean.

~Normal Curve

A bell curve that is symmetrical and has ~68% of people in the first standard deviation, ~95% of people in the second, and ~99% of people in the third. One example is an IQ bell curve.

Correlation Coefficient

The measure of a strength of a correlation, that can range from -1 to +1. -1 indicates a perfect negative correlation, +1 indicates a perfect positive correlation, and 0 indicates no correlation.

Scatter Plot

A chart that graphs pairs of values and depicts a correlation (or lack thereof).

Line of Best Fit/Regression Line

A (linear) line that depicts the general trend shown in a scatter plot to the best extent possible.

Inferential Statistics

Statistics that determine how applicable findings are to the general population.

Sampling Error

The extent to which a sample differs from the population.

p value

The probability that the difference between the control and experimental groups is due to chance (and no/minimal other factors).

Statistical Insignificance

A label giving to data that has a high probability of being caused by chance and not differences between the groups studied. This label is applicable when the p value of a study is higher than 0.05.

Institutional Review Board (IRB)

An institution of the American Psychological Association that determines if a study has ethical violations before research is started.

Coercion

When someone is forced in some way to participate in a study. This is an ethical violation according to the APA.

Informed Consent

Researchers must make sure the participant is aware of the study they are participating in. They can be decieved slightly if necessary for the study, but they cannot be deceived to an extent high enough to cause trauma, and the information given must be similar to the contents of the actual study. A lack of this is an ethical violation according to the APA.

Anonymity

A researcher should maintain privacy for the participants to prevent their data in the study from being connected to their name (to the best extent possible). A lack of this is an ethical violation according to the APA.

Confidentiality

If a researcher is unable to provide anonymity, they should do this, which leaves the source of data in a study indeterminate. A lack of this and a lack of anonymity is an ethical violation.

Risk

Participants should not be placed at significant mental or physical risk. Too much of this is an ethical violation according to the APA.

Debriefing

After the study, participants should be both informed of the study and given a way to access the researcher for findings. A lack of this is an ethical violation according to the APA.