1 / 25

PSYC 235: Research Methods

PSYC 235: Research Methods. FACTORIAL DESIGN. Complex Experiments. Some Terminology Factorial Factor Level WHY? Approach the complexity of nature 3 hypotheses for the price of 2. Example: State Dependent Learning. Definition of “state dependent learning”

Download Presentation

PSYC 235: Research Methods

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. PSYC 235: Research Methods FACTORIAL DESIGN

  2. Complex Experiments • Some Terminology • Factorial • Factor • Level • WHY? • Approach the complexity of nature • 3 hypotheses for the price of 2

  3. Example: State Dependent Learning Definition of “state dependent learning” - will this ALWAYS occur, or does “it depend” - what about the effect of time spent studying? - investigating learning with two independent experiments

  4. Investigating question with FACTORIAL design - IV #1: STUDY LOCATION library v. classroom - IV #2: STUDY TIME 1 v 10 hours - DV test score FACTORS and LEVELS FACTORIAL COMBINATION MAIN EFFECTS INTERACTION WITHIN, BETWEEN OR MIXED

  5. Main effect: Does study location have an effect on exam score?

  6. Main effect: Does amount of study time have an effect on exam score?

  7. Interaction: Does effect of location DEPEND upon amount of time studied?

  8. Main effect: - when the effect of one IV is the same at all levels of the second IV - the effect of the two IVs are independent Interaction - when the effect of one IV is NOT the same at all levels of the second IV - the effects of the two IVs are NOT independent

  9. EXPERIMENT: Randomly assign 10 subjects to each condition Give assigned combinations of treatments Record subsequent exam scores and compute means Compute MAIN EFFECTS Testing for interaction: additivity and graphing

  10. Graphing Vertical Axis – dependent variable Horizontal Axis – grouping variable - which one? - rule of thumb

  11. Interaction: Does effect of location DEPEND upon amount of time studied?

  12. Interaction: Does effect of location DEPEND upon amount of time studied?

  13. What would the data look like IF: - you had NO significant main effects - you had NO interaction What would the data look like IF: - you had NO significant main effects - you HAD a significant interaction

  14. A recent study indicates that listening to disco music from the 1970’s causes a strange form of brain damage known as “Saturday Night Fever”. Greatly concerned, being this music was playing during my formative years, I decided to conduct a study investigating the effect in rat subjects. I placed young rats each in one of three rooms. In the first room, a variety of music was continuously played for 24 hours per day excluding anything recorded between the years 1970-1979. After 6 months in this environment, I measured the size of the frontal lobe of these rats. The average size was 70. In the second room, I housed 10 more young rats listening exclusively to the BeeGees. Two of the rats in this condition failed at “Stayin Alive” for the full 6 months, but the average lobe size of the remainders was 10. Finally, I had a control group of young rats that had no music playing. Their lobe size was 40. I did the same thing with a group of old rats, to see if Saturday Night Fever was dependent upon the age of exposure. The experiment was identical to that described above – and lobe sizes were 60 for the group exposed to no music, 30 for those exposed to the non-70’s music, and 30 for those exposed to the BeeGees. Organize the data, calculate main effects, test for additivity, evaluate your hypotheses, graph the data and explain the results to Grandma.

  15. The Burpo Beer company is interested in test marketing their new flavored beers, licorice, pineapple and bubblegum. A group of male college students drank a can of each beer and was asked to rate the flavor on a five point scale where 1 is utterly awful and 5 is delicious. First they drank the licorice beer. The average rating was 1. Next they drank the pineapple beer and gave it a 3 and finally the bubblegum flavor was given a 5. A comparable group of college females were then tested first with bubblegum, which they rated as a 1, pineapple, rated as a 3 and finally licorice which they ranked as a five. The brilliant marketing executive suggests that different ad campaigns be undertaken to specifically target bubblegum flavor to me, and licorice flavor to women. Organize the data, calculate main effects, test for additivity, evaluate your hypotheses, graph the data and explain the results to Grandma. Do you agree with the course of action that the company is planning on? Why or why not?

  16. These 2 guys reared this lion from a baby in England but the authorities would not allow them to keep it once it reached maturity so they were forced to give it up, they took it back to Africa and placed it in a wildlife sanctuary, a year later they went to see it and were told it would not remember them. THIS IS AN ANECDOTE!!!!!http://www.youtube.com/watch?v=oiGKWoJi5qM How would you test for animal remembrance of human caretakers? Could this be operationally defined in such a way as to either verify or refute the experts who say that animals will not remember humans one year later? Do animals feel emotion? How could we test this question?

  17. Suppose that I’m currently conducting a study with a student investigating the effects of predator odors on maze running in rats. After the student provides the rats with extensive daily training in the maze, half of the rats are randomly assigned to be exposed to the odor of a predator and half are not (for 15 minutes). Following this exposure, we take turns conducting four daily tests in which the number of arms in a new maze that the rat correctly visits is counted. On the first day the predator exposed animals visit 4 arms on average compared to the control animals which visit 6 arms. On the second day the exposed animals visit 8 arms to the control groups 10. On day three, the control group drops back to 6, the exposed group visits 4. Finally, on the last day the exposed group visits 8 arms to the control groups 10. Organize the data, calculate main effects, test for additivity, evaluate your hypotheses, graph the data and explain the results to Grandma. Did exposure to the predator have an effect? Did the animals learn?

  18. Real Data Tennison, L.R., Rodgers, L.S., Beker, D.L., Vorobjeva, K.I., Creed, E.T. & Simonenko, A. (under review). Cortisol and symptoms of psychopathology in Russian and American college students. International Journal of Psychology. Method Design 2 x 2 x 4 Mixed 2: SEX: male and female (between) 2: NATION: Russian and American (between) 4: TIIME: upon awakening, 30 minutes later, 4 pm, 10pm (within) Participants Ninety-nine Russian (62 males and 37 females) and 118 American (45 males and 73 females) undergraduate students voluntarily served as participants. The Russian participants were enrolled in the Far Eastern State Transport University, a relatively large, public, urban, university located in Khabarovsk, Russia. The American students were enrolled in the College of Saint Benedict and Saint John’s University, two relatively small Catholic liberal arts colleges that academically function as one institution, located in rural Minnesota. The mean age of these groups was virtually identical (Russian sample M = 20.67, SD = 1.54; American sample M = 20.60, SD = 1.08).

  19. Materials and Procedure All procedures were conducted in accordance with the Helsinki Declaration of 1975, as revised in 1983, and were reviewed and approved by the institutional review board. Each participant was provided with four Salivette tubes obtained from Sarstedt Company, Newton, NC and was instructed to collect four saliva samples on the following day: immediately upon awakening, 30 minutes after awakening, at 4 p.m. and at 10 p.m. Proper procedures for collecting the samples were demonstrated by course instructors and these instructions were also provided in writing to each student when the tubes were distributed. In addition, participants were instructed to refrain from smoking, eating, drinking or brushing their teeth for 30 minutes prior to taking each sample. On the following day the participants returned with their samples and completed a battery of surveys designed to assess symptoms of psychopathology. The packet included standard demographic questions as well as the Hopkins-25, which assesses symptoms of depression and anxiety, a brief checklist of symptoms of PTSD, the Trauma History Questionnaire (Green, 1996), two measures of alcohol-related problems, the CAGE (Ewing, 1984) and the AUDIT (Bohn, Babor, & Kranzler, 1995) and the Brief COPE (Carver, 1997) a survey assessing use of various coping strategies. With the exception of the Hopkins-25 (Hoffman et al., 2006) we were unable to find Russian translations of these instruments in the published literature. Therefore we had the instruments translated and back-translated by individuals fluent in both languages. The Salivette tubes and surveys were first collected in Russia during the final exam period of spring semester classes. The tubes were transported back to the United States and frozen for storage. Because exam stress has been found to alter cortisol levels (Malarkey, Pearl, Demers, & Kiecolt-Glaser, 1995; Loft, Thomas, Petrie, Booth, Miles, & Vedhara, 2007) we collected the corresponding data from American participants during final exam periods in the following two fall semesters. All of the samples were frozen following collection, were thawed and were analyzed using ELISA analysis kits obtained from Salimetrics, LLC, State College, PA.

  20. Tests of Within-Subjects Effects Source Type III Sum of Squares df Mean Square F Sig. time Sphericity Assumed 8.418 3 2.806 59.685 .000 Greenhouse-Geisser 8.418 2.607 3.229 59.685 .000 Huynh-Feldt 8.418 2.808 2.997 59.685 .000 Lower-bound 8.418 1.000 8.418 59.685 .000 time * nationSphericity Assumed .074 3 .025 .524 .666 Greenhouse-Geisser .074 2.607 .028 .524 .641 Huynh-Feldt .074 2.808 .026 .524 .654 Lower-bound .074 1.000 .074 .524 .471 time * sex Sphericity Assumed .162 3 .054 1.149 .330 Greenhouse-Geisser .162 2.607 .062 1.149 .327 Huynh-Feldt .162 2.808 .058 1.149 .329 Lower-bound .162 1.000 .162 1.149 .287 time * nation * sex Sphericity Assumed .155 3 .052 1.100 .350 Greenhouse-Geisser .155 2.607 .060 1.100 .346 Huynh-Feldt .155 2.808 .055 1.100 .348 Lower-bound .155 1.000 .155 1.100 .298 Error(time) Sphericity Assumed 11.000 234 .047 Greenhouse-Geisser 11.000 203.313 .054 Huynh-Feldt 11.000 219.060 .050 Lower-bound 11.000 78.000 .141

  21. Tests of Between-Subjects Effects Measure: MEASURE_1 Transformed Variable: Average Source Type III Sum of Squares df Mean Square F Sig. Intercept 21.020 1 21.020 203.133 .000 nation 1.289 1 1.289 12.460 .001 sex .060 1 .060 .584 .447 nation * sex .212 1 .212 2.044 .157 Error 8.071 78 .103

More Related