Class Notes. Psychosurgery. Psychosurgery: A Sad History. The Importance of Research Methods. Notable Cases: Rosemary Kennedy

Transcription

1 Chapter 2: Research Methods Psychology: From Inquiry to Understanding 1/e Scott O. Lilienfeld Steven Jay Lynn Laura Namy Nancy J. Woolf C. Brown Unit 2 Class Notes Remember taking notes is like sending a text message to yourself. Do not expect to write down every word. The members of each study group should help their team mates fill in the gaps. You learn best by having to fill in the missing information after the class period is over, but before the day is over. This multimedia product and its contents are protected under copyright law. The following are prohibited by law: any public performance or display, including transmission of any image over a network; preparation of any derivative work, including the extraction, in whole or part, of any images; any rental, lease, or lending of the program. The Importance of Research Methods Protecting ourselves from inaccurate impressions: The Saga Psychosurgery The first controlled lobotomy was performed by the Portuguese physician António Moniz in Moniz won the Nobel Prize for medicine in 1949 for this work. The procedure was brought to the United States by Drs. Walter Freeman and James W. Watts. Psychosurgery: A Sad History Following WWII thousands of returning veterans sought assistance. Some veterans may have been suffering from post-traumatic stress disorder. Prior to the development of a lobotomy was thought to be a reasonable treatment. Lobotomies were most prevalent in the States: nearly WWII veterans received lobotomies. Notable Cases: Rosemary Kennedy Rosemary Kennedy, sister of, was given a lobotomy at age 23 to control her moodiness and growing interest in men. She was reduced d to an infantile mentality. The nature of Rosemary's condition was concealed for years by her father, and was presented to the public as the result of mental retardation. Her sister, Eunice Kennedy Shriver, founded the in her honor in

2 Notable Cases: Howard Dully Howard Dully received a lobotomy at 12, to control his "youthful defiance". He had no memory of the procedure, and did not discover the story of his lobotomy until he was in his 50 s. He noted, "I've always felt different -- wondered if something's missing from my soul. He is shown holding an icepick like instrument similar to the one used for his lobotomy. Psychosurgery Today In 1977, the U.S. Congress created a National Committee for the Protection of Human Subjects of Biomedical and Behavioral Research to investigate allegations that lobotomy techniques produced unethical after effects. It concluded that psychosurgery could have positive effects in some cases. Today about lobotomies are conducted in the US each year. The Psychosurgery Saga Illustrates the in the absence of ethical guidelines, and research methods. Lobotomies were adopted for use on human subjects following the report of only one case study conducted on one chimpanzee. No controlled experiments were ever conducted. : surgeons perceived improvements in the behavior of their patients that supported their belief in the benefits to be derived from the procedure, and downplayed outcomes that didn t. Textbook Assignment Ch2-1 The figure to the left is from our textbook. What phenomenon is being depicted here? Ans: Were the results supporting this procedure more scientific than those supporting prefrontal lobotomies? Ans: your response to your SI. Modern Ethical Guidelines HUMAN RESEARCH Institutional Review Board (IRB) Justification of of subjects afterward Animal Research Institutional Animal Care and Use Committee (IACUC) About 8% of psychological research uses animals The Need for Unbiased Observations It is clear that behavioral scientists need to develop rigorous methods to insure that their conclusions are derived from and scientific observations, and are not contaminated by errors in human judgment. The following exercise illustrates how psychologists conduct experiments. Copyright Allyn & Bacon

3 In Class Experiment In Chapter 3 we will learn that the two sides of the brain are not symmetrical regarding their function. This finding is called. In the following exercise we will test the idea that the two sides of the human face are unequal regarding the expression of human emotion. The Research Process Research scientists would test this idea by posing two alternative hypothesizes. Studies are designed to collect data to determine if the Null Hypothesis is statistically unlikely. Hypothesis: the two side of the face differ in their capacity to express (or conceal) emotion. Null Hypothesis: the two sides are equally expressive. Research Designs In this experiment a computer has been used to split the human face into two images. One image is composed of the left-side and a mirror image of the left half of the face. The second image is composed of the right-side and the mirror image of the right half of the face. Vote for the most expressive side! Class Results The results from our class suggest that the two sides of the human face are equally expressive. T/F Class Results The results from our class rejected the null hypothesis (the two side are equally expressive). Confounding Variable A variable T/F 3

4 Our Class Experiment Has A Confound In Its Design Results from a large sample of students suggest that the left-side of the face is emotionally most expressive. This side of the face is controlled by the right brain. The confound is the possibility that the right brain is also more sensitive to the detection of emotion, and it receives images from the left visual field. Thus, we are uncertain if the left side of the face is more expressive emotionally, or if images in the left visual field are perceived as being more emotionally expressive. How could you change the location of the pictures in the slides to correct for this potential confound? Confer with the members of your study group to propose a solution to the confound. You have 60 seconds to meet. There are at least two ways to resolve the confound. Answer Anatomy of an Experiment Some students score well on exams while others score poorly. Differences between exam scores may not always be due to the student s s ability level. Positive and negative thoughts that arise during the exam may reduce or heighten anxiety, and differences in anxiety may alter test performance. How to Influence +/- Thoughts Give the exam on a computer and present bogus feedback during an exam via computerized testing. Deceive the subjects into thinking they are performing very well or very poorly. E.g. Correct! Not many students select the right answer the first time they are tested on this concept. Hypothetical Experiment In the hypothetical experiment that follows the investigator attempts to improve performance during the exam by giving positive feedback to the students. After a response is submitted, the student is always told that they were correct. A control group 4

5 Random Sample A Experimental Group (positive feedback) Outcome (test scores) Population of Students Independent Variable Dependent Variable Random Sample B Control Group (no feedback) Outcome (test scores) It would be easier for the investigator to assign all the female students to the Experimental Group, and all the male students to the Control Group. Would this be a good idea? Justify your answer. Confer with the members of your study group and develop your response. You have 60 seconds to meet. Answer Research Designs Experiments Independent Variable The proposed cause of change in the dependent variable. Dependent Variable Proposed to be affected by the independent variable. Research Designs Experiments Experimental Group Control Group Experimentation Random Assignment assigning participants minimizes between those assigned to the different groups 5

6 The Dependent variable is the variable that is being altered, changed or manipulated by the investigator. T/F Data Analysis Researchers rely on statistics to analyze the results they collect. Statistics is a branch of mathematics that is used for analyzing research data. Summary Statistics Measures of Mean - score Mode - score Median - score Measures of Variability Range Average Deviation Standard Deviation Central Tendency Mode the most score in a distribution Mean the arithmetic of a distribution obtained by adding the scores and then dividing by the number of scores Median the score in a distribution half the scores are it and half are it Subject Score Sum 1174 Summary Statistics Central Tendency Mean = 1174/11 = Mode = 100 Median = Dispersion Range Standard Deviation 6

7 Subject Sum Mean Summary Statistics Score /X M/ /X M/ Class Notes The average deviation is an intuitively sensible way to characterize dispersion around the mean. The values generated by the average and standard deviation are different numbers, but they tend to be similar to each other. The average deviation is of no value in statistical tests, and it is rarely ever calculated. Standard deviations are a very useful value to calculate. Summary Statistics Measures of Variability Range Average Deviation Σ/X M/ N Standard Deviation Σ/X M/ 2 N Historical Figure Sir Francis Galton, (Charles Darwin s half cousin). Founder of. Discovered that psychological traits were distributed. Proposed that a distribution be described by only two numbers: the and. Invented the concept of and the formula for its computation. Distributions Skewed Distributions The Normal Curve Rectangular Statistical Reasoning** A Skewed Distribution Mode Median One Family Mean Income per family in thousands of dollars 7

8 Measures of Variation** Range: The difference between the highest and lowest scores in a distribution. Standard Deviation Units and the Normal Distribution** Standard Deviation: A computed measure of how much scores vary around the mean. Revised 2006 PSB Rectangular Distribution Which value is the mode in a rectangular distribution? A. Every value within the range of scores in the distribution has the same frequency they are all the mode. B. The median value. Low Scores High Standard Deviation Units and the Normal Distribution** 34% of the scores fall between the mean and 1 SD below or above the mean Class Notes For the purpose of this class round 34.13% to 34%. Round 13.59% to 14%. SD = Standard Deviation 8

9 Standard Deviation Units and the Normal Distribution** What is the approximate percentage of scores that fall between the mean and 1 SD above the mean? 14% of the scores fall between 1 and 2 SD either above or below the mean What is the approximate percentage of scores that fall between 1 SD below the mean and 2 SD below the mean? For his first two college assignments Sam earned a score of 26 in his psychology class and a 32 in biology. In which of these classes did Sam perform best relative to his classmates? 1. Psychology 2. Biology 3. I need to know the mean and SD for the scores in each class to answer this question. Given: Psy: M = 20, SD = 3 Bio: M = 30, SD = 4 In terms of percentile ranks, which score is better, a value of 26 in psychology or a grade of 32 in biology? 1. Psychology 2. Biology Class Notes M = mean You are expected to know how to evaluate scores using percentile rankings. Ask questions now if this is unclear! 9

10 Correlation Sometimes cannot be conducted. The correlation coefficient is a measure of how strongly two variables are. A high association imply causation. Correlation** Correlation Coefficient a statistical measure of the extent to which two factors vary together, and thus how well either factor predicts the other Correlation coefficient Indicates direction of relationship (positive or negative) r = +.37 Indicates strength of relationship (0.00 to 1.00) Correlation Examples of a Scatterplot Scatterplot a, each of which represents the values of variables the of the points suggests the direction of the relationship the of scatter suggests the of the correlation little scatter indicates correlation also called a scattergram or scatter diagram Copyright Allyn & Bacon 2008 Perfect positive correlation (+1.00) Correlation** No relationship (0.00) Perfect negative correlation (-1.00) Scatterplots, showing patterns of correlations Correlation A correlation is an x y plot of how strongly two variables are associated. In a positive correlation (positive slope) as x-values y-values also tend to. In a negative correlation (negative slope) as x-values, y-values tend to. 10

11 Textbook Assignment Ch2-2 In addition to experiments and studies using correlation, two other methods are sometimes used in psychological research. What are they? designs involve watching behavior in real-world settings. designs can provide existence proofs, but are largely anecdotal. Of the two designs noted above, has high external validity, but low internal validity. your response to your SI. Free Transportation At the University of South Alabama a free public transportation system (Jag Tran) was launched just prior to the onset of the rise in the cost of gasoline. It took a while for students to begin to ride the Jag Tran. The association between the number of students riding the Jag Tran and the cost of gasoline is shown in the next slide. Correlation and Causation Correlation and Causation Riders per We eek Jag Tran and the Cost of Gas Is it fair to argue that the increase in the number of Jag Tran riders has caused the increase in the cost of gasoline? A strong correlation or association does not prove!! Dollars per Gallon The Advantage of Experiments Experiments permit researchers to establish cause and effect relationships between dependent and independent variables. Other types of research design may provide greater validity. All psychological research is conducted in laboratory settings. T/F 11

12 THE RULES Cooperative Group Challenge 1. Placebo effect 2. Nocebo effect 3. Double blind design 4. Experimenter expectancy effect 5. Demand characteristics 6. Hawthorne effect Q harm resulting from the mere expectation of harm Q neither researchers nor subjects know who is in the experimental or control group Q phenomenon in which participants knowledge that they re being studied can affect their behavior Q improvement resulting from the subject s expectation of improvement 12

13 Q cues that participants pick up from a study that allow them to generate guesses regarding the researcher s hypothesis Q improvement resulting from the investigator s expectation of improvement 13