research methods Flashcards
Internal validity
How legitimate the treatment is
External validity
How generalizable/representable the treatment is
Statistical significance
Difference is not caused by chance
Clinical significance
Difference reflects on clinical impact
Effet sizes
0.1 - 0.3 (weak)
0.3 - 0.5 (moderate)
0.5+ (high/strong)
Completer analysis
Data from only the participants that completed the treatment (overestimates effectiveness)
Correlation research
Investigates relationship between variables without researchers manipulating them
Cross sectional design
Type of correlation researche involving collection of data at one time (no manipulation)
Longitudinal design
Correlation research studying 1 group for a long time
Epidemiological research
Studying the consequences, prevalence (% of pop), incident (# of new cases)
Experimental research
Treatment outcome research
Prevention research
Protecting certain group:
- health promotion (packing labels)
- universal prevention (entire pop)
- selective prevention strategies (at risk groups)
- indicated prevention strategies (people with subclinical symptoms)