Psyc 321_02 methods of_science
Download as PPTX, PDF1 like226 views
This document discusses several key methods of science: naturalistic observation which involves non-intrusively observing populations without influencing them; correlational approaches which relate variables mathematically without implying causation; experimental methods which test hypotheses using experimental and control groups; and operational definitions which clearly define variables. It also discusses independent and dependent variables, confounds, validity, reliability, and the scientific attitude which values falsification, testability, skepticism, and parsimony.
Psyc 321_02 methods of_science
- 1. Research Methods Topic 2: The Methods of Science 10/1/2014 CEDP 321/322 Ryan Sain, Ph.D. 1
- 2. Naturalistic observation Who is Ford Prefect? What is the Prime Directive? If you want to know about something, without influencing the process itself (well as much as possible anyway) Record everything! Non-interference Useful when you know little about a population or phenomenon OSS 10/1/2014 CEDP 321/322 Ryan Sain, Ph.D. 2
- 3. Correlational approach This is a method of relating two or more variables – mathematically (we get a degree of relation) It does not have anything to do with causation Ice cream and crime; shoe size and IQ; socks and serial killing. There is no attempt at manipulation What is the value? It gives us clues as to where cause MIGHT exist between variables. Positive and negative correlation. 10/1/2014 CEDP 321/322 Ryan Sain, Ph.D. 3
- 4. The experimental method Hypothesis ◦ Is the idea or statement being tested in an experiment or study. Experimental group ◦ The group receiving the treatment Control group ◦ The group we are comparing to – typically receives nothing. 10/1/2014 CEDP 321/322 Ryan Sain, Ph.D. 4
- 5. Operational definitions • Define stopping a car. • Define overeating. • Define exercising. • Define happiness. • (measurement) The goal with an operational definition is to make the instance so clear that anyone can observe it and that we all observe the occurrence at the same time. (reliable) • (experimental) Defining the procedures in an experiment clearly enough to be followed by anyone that has the means to do so. 10/1/2014 CEDP 321/322 Ryan Sain, Ph.D. 5
- 6. Independent Variables Something that has an effect on something else This is the variable of interest in a scientific study To be experimental: ◦ At least one group gets the IV (experimental) ◦ At least one group does not (control) Examples 10/1/2014 CEDP 321/322 Ryan Sain, Ph.D. 6
- 7. Dependant Variables These are the things in a study that are effected by the IV This is what we measure Examples The difference between your groups on the DV is the treatment effect. 10/1/2014 CEDP 321/322 Ryan Sain, Ph.D. 7
- 8. Confounds Things that COULD explain the treatment effect that are not part of the treatment. There are an infinite number of these. We minimize them by introducing experimental control. But if we control too many things – the findings do not represent the real world. Must find a balan1c0/1e/20.1 4 CEDP 321/322 Ryan Sain, Ph.D. 8
- 9. Validity A key to any experimenting is the concept of validity We are doing what we intend to do. We are measuring what we intend to measure. ◦ True and capable of being supported. Internal validity ◦ Is the study sound enough to detect the effect we are predicting and trying to measure! ◦ Are there confounds? These threaten internal validity. External validity ◦ Generalizability. ◦ Do the findings represent the real world or are they only a reflection of the laboratory conditions? 10/1/2014 CEDP 321/322 Ryan Sain, Ph.D. 9
- 10. Reliability The same group of people experiencing the same IV should produce the same results on a DV. Repeatability Treatment reliability Measurement reliability A valid measure is reliable – but not necessarily the converse. 10/1/2014 CEDP 321/322 Ryan Sain, Ph.D. 10
- 11. Scientific attitude Falsification ◦ We seek to show what is not the case! ◦ It only takes one instance to destroy a 1000 years of common held belief. Testablity ◦ The thing we are interested in studying must be testable! Skeptical ◦ We should be skeptical about our results Parsimony ◦ Two explanations for a given event, the more simple is likely correct. 10/1/2014 CEDP 321/322 Ryan Sain, Ph.D. 11