IB Psych Experiment Test Terms/Definitions

theory

an explanation for certain behaviors

study

- tests the validity of a theory

- can SUPPORT or CHALLENGE a theory, not disprove

TEACUP

T - testable; ability to be proven false

E - empirical evidence; not anecdotal, can be replicated and is reliable

A - application; can be applied to different situations

C - construct validity; variables can be clearly defined, can be reliably measured

U - unbiased; self-explanatory

P - predicts behavior; predicts something about human behavior

quantitative research methods

- emphasizes objective measurements

- determines cause-effect relationship

independent variable

causes a change in another variable (manipulative)

dependent variable

measured after manipulation of the IV

operationalized

either the IV or DV (or both) written in a way that can be clearly measured

true experiment

randomly allocates participants - lessens bias, more of a lab setting

quasi experiment

no random allocation - grouped based on existing traits/behavior

natural experiment

more of an observation of what happens rather than an experiment - outside a researchers control

experimental hypothesis

- expectation of a study

- predicts the relationship between the IV and the DV

null hypothesis

- states that the IV has no effect on the DV

- change in DV is due to CHANCE

extraneous variables

other variables that may impact the results of the study

confounding variables

- a type of extraneous variable

- affects both the IV and DV

demand characteristics

- expectancy effect: expects the hypothesis and acts to "help" by giving the right answer

- screw you effect: acts to destroy credibility of the study

- social desirability effect: answers to make them look good

researcher bias

when a researcher has expectations and only sees what they're looking for

participant variability

a limitation when the characteristics of a sample (often similar) affect the dependent variable

artificiality

when a situation is so unlikely the findings have to be questioned on validity

bidirectional ambiguity

NO cause-effect relationship but another responsible variable

repeated measured design

- one sample that receives all conditions

- participants only compared to themselves (strength)

- may demonstrate order effects and have confounding variables (limitation)

- use counter-balancing to control internal validity

order effects

boredom, fatigue, practice effect

counter-balancing

each sample group starts with opposite conditions and then switches

independent samples design

- one sample randomly allocated to one condition

- order effects are controlled (strength)

- lower participant variability (limitation

matched pairs design

- participants pre-tested and then allocated

- lessens participant variability

opportunity sampling

- "convenience sampling"

- not actually convenient but used when random sampling is impossible

- sometimes termed "accidental sampling"

purposive sampling

- not random

- when you need to test a certain trait

snowball sampling

- technique to recruit participants

- word-of-mouth

- typically for rare diseases/disorders that have their own community, participants hard to locate otherwise

stratified/quota sampling

- divide sample into distinctive groups based on traits

- draw samples from those groups and split them into controlled or agent groups

ethics

the correct rules of conduct necessary when carrying out research

6 key ethical themes in research

1. informed consent

2. freedom from coercion

3. protection from physical/psychological harm

4. protection of confidentiality/anonymity

5. risk-benefit rule

6. debriefing

informed consent

participants have to consent to be in a study

inform on:

- purpose

- right to decline

- potential risks and benefits

- confidentiality (limits)

- incentives and participants' rights

freedom from coercion

- coercion present when researcher has power over participant

- no restrictive communication

- no sense of powerlessness or questioning of self

- no strong emotional arousals or intimidation

protection from physical/psychological harm

- no physical harm (TEMPORARY pain is okay is risk-benefit rule is approved)

risk-benefit rule

- all previous themes may have their line crossed depending on the nature of the study

- any risk requires justification and must be approved by ethics board

debriefing

- a "double-check" at the end of the study

- everyone is okay, everyone's data can be used

- minimize any harm that was done during the study (if applicable), clear up misconceptions

deception

only permissible when:

- research is IMPORTANT (risk-benefit)

- there are NO ALTERNATIVES to running the experiment

- NO FORESEEABLE HARM to participants

- must be explained during debrief if used