Research Methods Flashcards
What is an observational study?
It’s a study in which individuals are observed or certain outcomes are measured. NO attempt is made to affect the outcome, so it’s not a “controlled” study
What is the main characteristic of a cross-secitonal study?
It’s just one point in time
What is the pro and con of a cross sectional study?
pro - cost effective and easiest to do
con - doesn’t capture time
What is required for a study to be longitudinal?
two or more time periods
What is the pro and con of a longitudinal study?
pro - more powerful because it captures change and aims for causality
con - more expensive and complex
What are the 3 types of longitudinal studies?
trend
cohort
panel
What does a trend longitudinal study look at?
It measures changes in a whole population over time
What does a cohort longitudinal study look at?
It measures changes and follows a articular population over time, but it samples DIFFERENT students within ONE particular population
(so different people are tested in subsequent trials)
How does a panel study work and how does it differ from a cohort study?
A panel study looks at the SAME people within a particular population every time
What are the benefits to observational studies?
- incorporate probability smapling for generalizability
- build knowledge base and explore new topics
- may allow us to answer questions that controlled studies cannot due to ethical reasons
WHat are the 3 main issues with observational studies?
validity of measures across cultures can be an issue if using surveys
findings are associations or correlations, no causality
often rely on self report which can be biased
Which have higher validity?
qualitative or quantitative data?
qualitative are more valid
quantitative are more reliable
What does nomothetic causation mean? What are the 3 criterion?
nomothetic causation is broad causation - in general we can assume that one thing causes another with the understanding that it doesn’t have to cause it every single time.
- correlation/association
- temporal order…A before B
- NOn-spuriousness
In addition to the correlation, temporal order and non-spuriousness criteria for nomothetic causation, what are two other criteria that are sometimes applied?
plausible and reproducible
What are the three main parts to the classical experiment?
- IV and DV
- Pretesting and post-testing
- Control and Experimental groups
What is the Hawthorne effect?
Just being observed affects behavior
What are the 4 stages of the clinical trial?
phase 1 - safety and side effects in healthy individuals
phase 2 - ideal dosing
phase 3 - experimental treatment vs. controls
phase 4 - continued evaluation of FDA approved therapy
What are the three potential pre-experiemental designs that are sometimes necessary but preferrably not used?
one shot case study - apply stimulus and measure DV afterwards
one-group pretest-posttest - measure DV, apply stimulus, measure DV
static group compartison - two groups, only one gets the stimulus, no pre-test
Controlled studies have excellent ___ validity but issues with ____ validity.
excellent internal validity - good at controlling just about everything
often has issues with external invalidity - can it be generalizable to the real world after they lose control?
What are some potential sources of internal invalidity?
history - current events
maturation - improve with aging
testing - learn to test better just through repetition
instrumentation changes
statistical regression
selection bias
experimental mortality
diffusion of treatment - contamination of control group
compensation incentive bias
compensatory rivalry (control group tries to make up for it)
demoraliation (control group discouraged)
