Transcript Design
Research Methodologies in Allied Health Mary C. Haven, MS Associate Dean, SAHP 1 This is how Hubert does research! He jumps right in. 2 Objectives After lecture, reading Section 3 in your text, discussion, and class participation, the student will be able to: Compare and contrast at least four research designs. Given a simple research question, adapt it to a specific research design and explain why this design was chosen. 3 Objectives Determine a control group for a research question. Select a way to randomize research subjects into experimental or control. 4 Experimental Design Researcher Manipulates the independent variable Observes the effects of this manipulation on the dependent variable Measure the effect on the dependent variable 5 Experimental Design Dependent variable will be altered by manipulation of the independent variable. 6 True Experimental Method Manipulation Control Randomization 7 Control Group A group of subjects that experiences the same day-to-day occurrences and influences as the experimental group but does not receive the study treatment. 8 Why am I always picked for the placebo and/or control group? 9 Randomization A random selection means that every subject in the population concerned has an equal chance of being selected for the study sample. 10 Table E.l RANDOM NUMBERS Appendix E Statistical Tables 11 Experimental Method Pre-Test/Post-Test Design 1. Hypothesis 2. Select experimental and control groups by randomization. 3. Do pre-test. 4. Manipulate experimental group. 5. Do post-test. 6. Analyze results. 7. Accept or reject hypothesis. R O R O X O O 12 Research Design Shorthand R = Randomization O = Observation X = Intervention 13 Example with This Class This Class Pre-test 40 30 27 44 32 This Class Control Gr. Control Gr. Post-test Pre-test Post-test 100 40 41 95 50 45 94 96 100 30 22 42 29 35 45 14 Another Example R R R O O O X1 O X2 O O New Treatment Old Treatment Placebo 15 Experimental Research The Classic Design Follow-Up Post Test Only R R O O R R X O O R R O O X X O1 O1 02 02 O O 16 Solomon Four-Group Design R O R O R R X O1 O2 X O3 O4 17 Cohort Designs R1 X O R1 O R2 O R2 X O 18 Factorial Design R X1 R X1 R O X2 O X2 O R O e.g. where X1 = Diet X2 = Drug 19 Harvard Law: Under the most rigorously controlled conditions of pressure, temperature, volume, humidity, and other variable, the organism will do as it damn well pleases. 20 Rosenthal Effect What the researcher expects is somehow conveyed to others, and that is the result found. 21 Hawthorne Effect Because the subject is being observed, the subject changes behavior. 22 Quasi-Experimental Designs Lack of Randomization O X1 O O O Cohort Designs O X O O O 23 Naturalistic Comparison Comparison of two or more groups No manipulation Dependent Variables Large sample size Statistical analysis Results may suggest theory Demographics Psychosocial Biochemical 24 Murphy’s Law of Research: Enough research will tend to support your theory. 25 Correlational Design Hypotheses or guiding question Literature search Select large random sample Take measurements Correlational and regression analysis Interpret results Write conclusion 26 Single Subject Design AB Design 6 5 Phase A B 4 Dep. Var. 3 Phase B 2 1 0 Time Phase A – taking observations B – intervention Phase B – observations after intervention 27 Single Subject Design ABAB Design 6 5 4 Phase A Phase A B B Dep. Var.3 2 Phase B Phase B 1 0 Time Phase A – taking observations B – intervention Phase B – observations after intervention 28 Single Group Time-Series Design Lack of Control – Single Group TimeSeries Design O X O O X O O X O Multiple Pre-Tests O O O O X O1 X O2 X O3 O4 O5 O6 29 Case Study Justify need Review literature State guiding question Obtain data Interpret results Generalize to theory 30 Edington’s Theory: The number of different hypotheses erected to explain a given biological phenomenon is inversely proportional to the available knowledge. 31 Survey Research Identify problem Define population Define variable Collect data Analyze results Make conclusions 32 Methodological Research Examination of need Literature search Define variables Pilot test Discussion of results Application 33 Young’s Law: All great discoveries are made by mistake. Corollary: The greater the funding, the longer it take to make the mistake. 34 Qualitative Field Research Disciplined inquiry Examine personal meanings Of experiences of individuals And actions of those individuals Results may lead to common themes 35 Questions to Consider In the article from the J. of Allied Health What was the design of the study? Can you draw a diagram of the sequence of events of the study? How many groups of subjects were in the study? If there were two or more groups, how did the subjects get into the different groups? Was the method of assigning the subjects to the different groups bias-free? 36