Class Notes (837,550)
York University (35,409)
Psychology (4,109)
PSYC 2230 (205)
N/ A (9)
Lecture 9

# RESEARCH METHODS LECTURE 9.pdf

4 Pages
98 Views

School
Department
Psychology
Course
PSYC 2230
Professor
N/ A
Semester
Summer

Description
RESEARCH METHODS LECTURE 9: TUESDAY JULY 24TH, 2012 TOPIC: FACTORIAL DESIGNS / TUTORIAL EXAM: focus on articles in textbook (case studies) know the purpose of the study, what the experimenter did and concluded, 78 questions, 4 choices. This exam is on chapters 5,6,7,8. INTRODUCTION TO FACTORIAL DESIGNS ➔ Overview of complex/factorial designs: main effect and interaction effect ➔ Researchers often investigate the effects of two or more IVs simultaneously ➔ All levels of each IV are combined with all levels of the other IV's ➔ Complexity relies on how the factorial is constructed FACTORIAL DESIGNS ➔ 2 x 2 design (described by number of variables and by how many levels each variable has) So in this case it is two variables with two conditions. First IV factor A: type of question Level 1: misleading and Level 2: unbiased ➔ Factor B: knowledge of the crime Level 1: naive question Level 2: knowledgeable questioner ➔ Each IV here has two levels. Dependent: whether they convict or let the suspects go ➔ It can be modified: by increasing the levels of an IV such as 3 x 2 = 2 IV's one with 3 levels and the other with 2 levels ➔ example: time of day x coffee drinking ➔ Factor: time of day has 3 levels: morning day and night ➔ Factor coffee drinking has 2 levels: cup of coffee or cup of water ➔ Another way is by making it more complex: 3 x 2 x 4 design has 3 IV's ( 3 levels, 2 levels and 4 levels) ➔ Example: time of day x coffee drinking x exam duration ➔ Factor time of day 3 levels: morning day and night ➔ Factor coffee drinking has 2 levels: 2 cups, 5 cups ➔ Factor exam duration has 4 levels: 120 minutes, 180minutes, 240 minutes, and 300 minutes ➔ 1) you can add levels to the study 2) you can add variables to the study 3) you can add both HOW TO DETERMINE THE NUMBER OF CONDITIONS ➔ 2 x 2 = 4 conditions (just multiply the numbers) ➔ 3 x 3 design = 9 conditions ➔ 3 x 4 x 2 = 24 conditions ➔ Number of participants with a between subjects design would become quite large. For instance: 24 conditions that need lets say 10 people per condition is a lot of subjects needed. May not be feasible to include all possible variables. ➔ Know how many IV's, the notation, the DV, how many conditions etc ➔ Example: mood and food deprivation on eating. IV : mood and food which is 2 x 2, DV is eating and..... and it is a between subjects design ➔ Example: hair length on effecting the judgments of a child's personality and intelligence. IV: hair length and gender. DV: personality and intelligence 2 x 2 between subjects design WHY SHOULD WE USEAFACTORIAL DESIGN ➔ We can examine the influence that each factor by itself has o our behaviour, as well as the influence that combining these factors has on the behaviour ➔ Can be efficient and cost effective INTERPRETATION OF FACTORIAL DESIGNS ➔ Two kinds of information: main effect of an IV : the effect that one IV has independently of the effect of the other IV ➔ Design with 2 IV's, there are two main effects ( one for each IV) ➔ Main effect of FactorA (1 IV): overall difference among the levels ofAcollapsing across the levels of B nd ➔ Main effect of Factor B (2 IV): overall difference among the levels of B collapsing across the levels of A MAIN EFFECT ➔ 2 x 3 design IVA and IV B shown B1 B 2 versus A1A2A3 = A1 B1A1 B2 next boxA2 B1A2 B2 and so on. A1 mean A2 meanA3 mean and B1 mean and B2 mean. Look at main effect for Afrom the 3 means you obtained. Do the same for factor B. Look at marginal means. See if they are the same or different. If they are different there could be a main effect in which you would do an F test from stats class ➔ 2 x 2 design Factor B versus FactorAacross the first 2 columns is 1 and 1 and across the bottom 2 is 9 and 9. To get means do 1=9/2 columns and get 5. Go across and do 9=9/2 =9 and 1+1/2=1 as the means. ➔ If the means are the same then there would be no main effect. If factorAgave you the means 30 30 30 there would be no effect. If it was 10 20 30 then yes there would be a main effect. ➔ It is a potential effect. Not certain without stats calculated. INTERPRETATION OF FACTORIAL DESIGNS ➔ Interaction: represent how IV's work together to influence behaviour ➔ Occurs when the effect of one IV differs depending on the level of the second IV ➔ Not ignoring or collapsing the other variable
More Less

Related notes for PSYC 2230
Me

OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Join to view

OR

By registering, I agree to the Terms and Privacy Policies
Just a few more details

So we can recommend you notes for your school.