Literature.pptx Research Methods in Psychology

Literature.pptx Research Methods in Psychology

1. Literature Review and Hypothesis Formulation

2. HYPOTHESIS General concepts • Is a formal tentative statement of the expected relationship between two or more variables under the study.

3. HYPOTHESIS General concepts •A hypothesis helps to translate the research problem and objectives into a prediction of the expected results or outcomes of the research study.

4. HYPOTHESIS General concepts •A clearly stated hypothesIs includes the variables to be manipulated or measured, identifies the population to be examined and indicates the proposed outcome for the study.

5. HYPOTHESIS General concepts A hypothesis is a statement of a predicted relationship between two or more variables. It comes after much thinking about a research idea. It is the thesis or main idea of an experiment. Y will have a different type of hypothesis for an experiment than for a nonexperimental design, but two kinds of studies have hypotheses.

6. HYPOTHESIS General concepts •The formulation of a hypothesis plays an important role in theory building.

7. Characteristics of Non-experimental HYPOTHESIS A nonexperimental hypothesis predicts how variables (events, traits, or behaviors) might be correlated, but not causally related. For example, the amount of TV viewing will be directly related to vocabulary size, or incandescent lighting in Company A will produce better performance than fluorescent lighting in Company B.

8. Characteristics of Non-experimental HYPOTHESIS Sometimes a nonexperimental design, such as a case study or a naturalistic observation, will not have a hypothesis because it was meant to explore and describe behaviors as they occur naturally. Other nonexperimental designs like correlational studies and quasi experimental studies, include a hypothesis about the predicted relationship between variables.

9. Characteristics of Experimental HYPOTHESIS An experimental hypothesis is a tentative explanation of an event or a behavior. It is a statement that predicts the effect of an independent variable or antecedent, on a dependent variable or measured behavior. For example, cognitive behavior therapy (CBT) produces less relapse than antidepressants.

10. Every experiment has at least one hypothesis. To formulate a hypothesis, the researcher must think about all of the possible factors affecting the variable they wish to study and decide which are important. Hypothesis must meet the criteria of being synthetic statements, that are testable, falsifiable, parsimonious and fruitful. Experimental Hypothesis

11. When we state a hypothesis, we then gather data that either support or contradict it. For this reason, a hypothesis must be capable of being true or false, which is a property of synthetic statements. The statements, “the weight of dieters will fluctuate” is always true, therefore it is not a synthetic statement. This statement is an analytic statement, one that is always true. Also must avoid contradictory statements; these are always false. Why must a hypothesis be a synthetic statement?

12. A hypothesis meets the definition of a synthetic statement when it can be stated in “if…..then” form. If the subject eats candy, then his mood will brighten. This form expresses the relationship between the antecedent conditions of eating candy and the measured behavior of mood. This statement can also be true or false. Synthetic statements

13. An experimental hypothesis is testable when it can be assessed by manipulating an IV and measuring the results on the DV. Without testability, we cannot evaluate the validity of a hypothesis. If a dog’s eyes move behind his lids during sleep, then he must be dreaming. Can we really test this hypothesis? What is testability and why is it important?

14. Statements of research hypotheses must be falsifiable (disprovable) by the research findings. Hypotheses must be worded in a way that failure to find the predicted effect must be considered evidence that the hypothesis is false. If an hypothesis is worded incorrectly, then it may not be falsifiable because any failures to produce the predicted effect can just be explained away by the researcher. “Chocolate is better than vanilla” and “God is real,” are both non- falsifiable statements. In science, you don’t prove hypotheses true, you either support them or prove them false. Falsifiable statements

15. Parsimony means that we prefer a simple hypothesis over one requiring many supporting assumptions. A simple hypothesis allows us to focus our attention on the main factors that influence our dependent variable. If you look at an appealing photograph, your pupils will dilate, or, If you look at an appealing photograph in June and it is warm and sunny out, then your pupils will dilate. Why should hypotheses be parsimonious?

16. A hypothesis should be fruitful, it should lead to other studies. Little Albert example Fruitful

17. The Inductive Model Induction is reasoning from specific cases to general principles to form a hypothesis. The inductive method is often used to create hypotheses. You notice something specific about some people, for instance, that all of your male friends who are hairy have higher sex drives than other men who are less hairy. Explain the inductive model of formulating a hypothesis.

18. You make a conclusion that hairiness must be related to sex drive. This is a theory that you have formed that can now be tested through an experiment. Researchers use inductive reasoning to construct theories by creating explanations that account for empirical data (observations). Induction is the basic tool of theory building. Explain the inductive model of formulating a hypothesis.

19. The Inductive Model- A theory is a set of general principles that can be used to explain or predict behavior. Astrology is an example of a theory. Scientists can use the results of extensive experiments designed to test hypotheses to construct a theory that unifies their findings. For instance, by looking at lots of research on exposure, a theory was formed that the more we are exposed to something, the more we like it. How can we build a theory using induction?

20. The Deductive Model Deduction is reasoning from general principles to specific predictions. It is the opposite of induction. ith deduction, you take a general idea, like a theory, and make prediction about individual specific behavior from this general principle. This approach is used to test the assumptions of an established theory, especially when applied in a new situation.. For example, Baumrind’s Parenting Styles theory – can see if her theory about the effects of authoritarian parenting holds in recent immigrant children. Explain the deductive model of formulating a hypothesis.

21. Combining Induction and Deduction Develop propositions using induction by examining specific cases. Induction makes theories. Then, make predictions using deduction. Deduction tests theories. Walster et al. formulated equity theory based on specific observations (induction) and then tested predictions from this theory (deduction). Equity theory says when we feel over benefited or under benefited in a relationship, we do something to restore a sense of fairness. How can we combine induction and deduction?

22. Building on Prior Research Review research that has already been published. Both experimental and nonexperimental studies can prove helpful. Ex: post facto studies suggested that cigarette smoking was associated with cancer since larger amounts of smokers got cancer than non smokers. This led to a hypothesis of, If a person smokes, then he will get cancer, and experimental studies being done in rats.Some rats, smoked, some didn’t (manipulation of antecedents) then all rate were tested for cancer. What is the most useful way to develop a hypothesis?

23. Building on Prior Research A review of prior experiments helps in five ways: 1. Identifies questions that have not been conclusively answered or addressed at all, research might say that this theory was never tested among black females; only white subjects were used. 2. Suggests new hypotheses 3. Identifies additional variables that could mediate an effect, perhaps authoritarian parenting is mediated by number of extended family members living in home. How does a review of prior experiments help us develop a hypothesis?

24. Building on Prior Research A review of prior experiments helps in five ways: 4. Identifies problems other researchers have experienced 5. Helps avoid duplication of prior research when replication is not intended How does a review of prior experiments help us develop a hypothesis?

25. Serendipity and the Windfall Hypothesis A scientist who is open to unexpected results and who is sufficiently informed can understand the significance of unexpected findings. Serendipity is finding something that was not being looked for. However, a dogmatic scientist would be less likely to “see” or appreciate the significance of serendipitous events. Example Ivan Pavlov. He was studying digestion in dogs, not classical conditioning. He wasn’t even a psychologist. How does serendipity help develop fruitful hypotheses?

26. When All Else Fails. Three promising strategies are: (1) Read an issue of a psychology journal, and try to stick to one or two topic areas. (2) Observe how people behave in public places (3) Choose a real-world problem and try to identify its cause. Byron (1996) wanted to increase female undergraduate condom use. She tested to see what type of attitudes were associated with women not using condoms (feelings of no control) and designed an intervention to reduce this feeling. What are helpful strategies for developing experimental hypotheses?

27. A meta-analysis can provide helpful information about your topic. A meta-analysis is not an experiment, but rather a statistical analysis of many similar studies. Explain the value of a meta-analysis?

28. A meta-analysis measures the average effect size of an independent variable across studies that share similar methodologies. This statistical procedure helps establish the strength and external validity of a causal relationship. Value of a meta-analysis