Research Methods Flashcards
p-Value
- indicate significance
- indicate how likely something occurred due to chance (small value means not likely due to chance)
What are the steps involved in designing an experiment?
- selection of experimental and control groups
- random sampling from population (ideally each individual in the population is equally likely to be sampled)
- random assignment to groups (assignments should be double-blind)
- control of extraneous variables (ex. age, gender, ethnicity, SES, education level, etc)
What can be some flaws of experimental designs?
- Researchers didn’t do enough
- sampling that is not random
- unable to control for every single extraneous variable
- control group not selected carefully enough
- not double-blind
Double-Blind
- neither the person placing people into groups nor the participants know which group is which
- good way to avoid the placebo effect
Extraneous Variable
- variables that are not being intentionally studied in an experiment
- considered undesirable variables when they influence the outcome of the experiment
- these variables should try to be controlled for so that individuals are equally distributed among groups
What is the main benefit of experimental designs?
these are the only designs that allow us to infer a casual relationship (ex. anti-depressants lead to better outcomes)
Internal Validity
the extent to which we can say the change in the outcome variable (or depend variable) is due to the intervention
List some common threats to internal validity
- impression management
- confounding variables
- lack of reliability (measurement tools do not measure what they purport to, lack consistency)
- sampling bias
- attrition effects
Impression Management
- participants adapt their responses based on social norms or perceived research expectations
- Ex. self-fulfilling prophecy (occurs when participants can predict how the research thinks they should act so they act in this way), methodology is not double-blind, Hawthorne Effect
Confounding Variables
- extraneous variables not accounted for in the study
- another variable that offers an alternative explanation
- an outside influence that changes the effect of a dependent variable and independent variable
- they can destroy the validity of an experiment by suggesting correlation when there is not any
Sampling Bias
- occurs when selection criteria is not random
- population used for sample does not meet conditions for statistical test (Ex. population is not normally distributed)
Attrition Effects
due to participant fatiguing during study or even dropping out of study
External Validity
the extent to which the findings can be generalized to the real world
List some common threats to external validity
- experiment doesn’t reflect the real world (lab set-ups that don’t translate to real world, lack of generalizability)
- selection criteria (can be too restrictive of inclusion/exclusion criteria for participants leading to a sample that is not representative)
- situational effects (presence of lab conditions change outcome)
- lack of statistical power (ex. sample groups have high variability, sample size is too small)
Non-Experimental Designs
- variables not directly manipulated, lack a control group
- includes: observational studies, ethnographic studies, twin studies and heritability studies, archival and biographical studies, phenomenological studies , case studies, longitudinal studies