AP Psychology Unit 0

Operational definition

Definition of HOW the variables in the research will be tested

Theory

Explanation which predicts events or behaviors

Validity

It measures what it set out to measure

Reliability

It produces similar results = consistent

Population

All individuals in the group to which the study applies

Sample

Subgroup of population that represents the population

Random selection/sampling

Every member of population has equal chance of being selected (ex. random number generator)

Stratified sampling

Dividing population into homogenous subgroups so that the sample better represents the population

Experiments

Attempt to prove CAUSATION by allowing researcher to manipulate one or more variables and measure outcome

Lab vs Field Experiment

Highly controlled vs Realistic

Random Assignment

Each participant has equal opportunity of being placed into any group → controls confounding variable

Experimental group

Group that receives the independent variable/treatment

Control group

Group that does NOT receive ANY treatment = measuring stick

Confounding variables

Any variable that might affect the dependent variable

Participant/Response Bias

Acting in a way which the experimenter wants them to act

Demand Characteristic

The clues participants discover about the purpose of the study and which affects their behavior

Social Desirability

Acting in a way which reflects well upon them

Experimenter bias

Unconscious tendency for researchers to treat members of experimental & control group differently to increase the chance of confirming their hypothesis

Single blind procedure

Participants don’t know which treatment group they’re in → control participant bias & demand characteristics

Double blind procedure

BOTH participants and experimenter don’t know → control experimenter bias

Placebo effect

The belief that the pill will work, having a psychological effect

Hawthorne effect

The fact that group has been chosen has affected the way they performed 테스트 참여자라는 이유로 평소와 다르게 행동

Correlational Research

Examines the “RELATIONSHIP” between two variables, NOT causation

Quasi experiments

Experiment where participants are NOT randomly assigned to study different in groups that have preexisting differences that CANNOT be controlled (ex. gender, age)

Survey

Using eu

Naturalistic Observation

Watching participants in their natural environment WITHOUT any manipulation

Case study

In-depth examination of rare phenomenon that occurred with an individual, group, or situation; CANNOT be generalized

Frequency Distribution

Gathering data to indicate how often a score occurs; frequency polygons & histogram

Measures of central tendency

Describes the average or most typical scores for set of research data or distribution: mean, median, mode

Negatively skewed

Low score outlier →

Positively skewed

High score outlier → mode/median/mean

Normal distribution

A symmetrical, bell-shape that represents occurrence of all scores in a given set of data

Measures of variability

Measures how spread the scores are: range, standard deviation, variance

Z-score/standard score

Distance of a score from the mean in units of SD

Percentile

Distance of a score from 0 (ex. 98 percentile = 상위 2%)

Correlation

Numerical relationship between 2 or more variables

Positive correlation

Presence of one thing predicts the presence of the other

Negative correlation

Presence of one thing predicts the absence of the other

Correlation coefficient (r)

Strength of correlation that varies from -1 to 1 (closer to 0 = weaker)

Scatterplots

Used to illustrate strength and direction of correlation

Statistical significance

Whether finding is result of systematic events, NOT random chance, which allows the researchers to infer whether the data can be generalized to population

P value

Error % = percentage that findings occurred by chance; smaller the better and it can NEVER be 0 unless there’s confounding variable

Meta-analysis

Combining the results of individual research to reach an overall conclusion

Null hypothesis

Independent variable has NO effect on dependent variable

Alternative hypothesis

Null hypothesis is false

Type I error

When researcher believes that finding is statistically significant, when it is due to some random fluctuation

Type II error

When researcher believes that a finding has appeared due to random fluctuations, when it IS statistically significant

American Psychological Association (APA)

They established the ethical guidelines

Institutional Review Board (IRB)

Any research must first propose the study to this board

Informed consent

Ethical guidelines that participants have to agree to be part of the experiment and are aware of what may take place

Debriefing

Ethical guidelines that participants must be allowed to view the results and informed of what researchers were hoping to accomplish

Confidentiality/Anonymity

Ethical principle of protecting the privacy and personal information of research participants