{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

research_methods_fall_2006

research_methods_fall_2006 - 12.Sept.2006 Methods of...

Info iconThis preview shows pages 1–4. Sign up to view the full content.

View Full Document Right Arrow Icon
12.Sept.2006 Methods of Psychological Research: Inference and Knowledge in Society and Science Caton Roberts, Ph.D. What Is Scientific “Knowledge”? Systematic and Controlled Observation, Inference, Tentative Conclusions (rules, structure) Difference from naïve/everyday knowledge: Conclusions are continually revised/revisited with fresh observations and re-examination of knowledge claims in light of current scientific theories (known is not nearly as known as we think) Critical Thinking: “I can find no evidence that the intensity of a belief is any measure of its validity” Sir Francis Bacon Ideas, theories, and “knowledge” are put at risk of “falsification” through rules of scientific method (confidence usually very high, but should we really be so?) The whole of science is nothing more than a refinement of everyday thinking Albert Einstein Common Research Methods in Psychology: Descriptive Research Methods (4 types, below) Experimental Methods Descriptive Research Methods Case Studies Careful Description based on repeated observations: Freud, Sacks, Clinical studies (medicine, completely in depth look) Naturalistic Observation: Ethology Researchers
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
“Ecological Validity” (Jane Goodall, outside in natural habitat, so there's no academic gap) Surveys (see Hite example below, and discussion of class survey experience) (large numbers) Correlational Methods (describe variables, the level of which two things relate to each other) Correlational Methods as Description A correlation is an association between two variables, expressed as a coefficient of association (r) which varies from +1 to –1 Visualizing associated scatterplots (figure 2.16 page 66) “Perfect” correlations: 1.0, direct diagonal graph upward Strong Correlations: Moderate Correlations: (consider “panel c” as a SAT/Freshmen grade scatterplot. http://www.fairtest.org/facts/satvalidity.html ) r=.47 .47-squared=.22 the r-squared and the “explanatory meaning” of a correlation coefficient square of r= the degree of variation that can be accounted for by the data Hence there are Positive, Negative, and Negligible Correlations Between Variables (without high correlations, there is much room for error) A correlation (r) is a mathematical representation which “describes” the strength of relationship between two variables. Limits of Correlations: Correlational Data can suggest causes but are Insufficient Grounds for Causal Inference CORRELATION DOES NOT PROVE CAUSATION.
Background image of page 2
3 Examples: “Statistical Significance” in Correlational and Experimental Research: Significance and chance findings: significant only if it is unlikely that it was due to anything random. sometimes results will be random and will not be statistically significant. p < .05 only < 5/100 p < .01 only <1/100 p < .001 only <1/1000 “Significant” correlations indicate predictability of variables: Statistical significance vs. practical or theoretical value: doesn't mean it is important Two “problems” with inferences of causation from correlational data 1) the directionality problem
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 4
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}