230 likes | 361 Vues
Group Differences: 2+ Independent Variables. 2+ IVs Factorial (between) Mixed (between & within). Main Effects. Imagine a study about eyewitness recall of crime. IV #1 = two levels (violent versus non-violent) An Independent t-test.
E N D
Group Differences: 2+ Independent Variables 2+ IVs Factorial (between) Mixed (between & within)
Main Effects • Imagine a study about eyewitness recall of crime. • IV #1 = two levels (violent versus non-violent) • An Independent t-test. http://www.upa.pdx.edu/IOA/newsom/pa551/lectur12.htm
Main Effects • Imagine a study about eyewitness recall of crime. • IV #1 = two levels (violent versus non-violent) • An Independent t-test. • IV #2 = two levels (training to remember, no training) • Another independent t-test. http://www.upa.pdx.edu/IOA/newsom/pa551/lectur12.htm
Main Effects • Imagine a study about eyewitness recall of crime. • IV #1 = two levels (violent versus non-violent) • An Independent t-test. • IV #2 = two levels (training to remember, no training) • Another independent t-test. http://www.upa.pdx.edu/IOA/newsom/pa551/lectur12.htm
Main Effects • Imagine a study about eyewitness recall of crime. • IV #1 = two levels (violent versus non-violent) • An Independent t-test. • IV #2 = two levels (training to remember, no training) • Another independent t-test. http://www.upa.pdx.edu/IOA/newsom/pa551/lectur12.htm
Main Effects • Imagine a study about eyewitness recall of crime. • IV #1 = two levels (violent versus non-violent) • An Independent t-test. • IV #2 = two levels (training to remember, no training) • Another independent t-test. http://www.upa.pdx.edu/IOA/newsom/pa551/lectur12.htm
Main Effects • Imagine a study about eyewitness recall of crime. • IV #1 = two levels (violent versus non-violent) • An Independent t-test. • IV #2 = two levels (training to remember, no training) • Another independent t-test. http://www.upa.pdx.edu/IOA/newsom/pa551/lectur12.htm
Main Effects • Imagine a study about eyewitness recall of crime. • IV #1 = two levels (violent versus non-violent) • An Independent t-test. • IV #2 = two levels (training to remember, no training) • Another independent t-test.
Interaction • Imagine a study about eyewitness recall of crime. • IV #1 = two levels (violent versus non-violent) • An Independent t-test. • IV #2 = two levels (training to remember, no training) • Another independent t-test.
Interaction • Imagine a study about eyewitness recall of crime. • IV #1 = two levels (violent versus non-violent) • An Independent t-test. • IV #2 = two levels (training to remember, no training) • Another independent t-test.
Interaction • Interactions are when the effect of two or more variables is not simply additive.
Interaction • Interactions are when the effect of two or more variables is not simply additive. • Terminology: • Presence of number refers to variable • 2 x 2 = 2 IVs. 2 x 3 = 2 IVs. 2 x 2 x 2 = 3 IVs • Content of number refers to levels within variable • 2 x 3 = 2 levels within IV1, 3 levels within IV2
Interaction • Interactions are when the effect of two or more variables is not simply additive. • Terminology: • Presence of number refers to variable • 2 x 2 = 2 IVs. 2 x 3 = 2 IVs. 2 x 2 x 2 = 3 IVs • Content of number refers to levels within variable • 2 x 3 = 2 levels within IV1, 3 levels within IV2 • Interaction (INT) = F-test across IV • Main Effect (ME) = F-test within IV • Simple Effect (SE) = pairwise between any two points
Interactions • Can have ME separate from INT
Interactions • There are different levels of Interactions. • A “Two-Way” Interaction can occur between any two variables (A-B) • A “Three-Way” Interaction can occur between any three variables (A-B-C) • A “Four-Way” Interaction can occur between any four variables (A-B-C-D)
Interactions • There are different levels of Interactions. • A “Two-Way” Interaction can occur between any two variables (A-B) • A “Three-Way” Interaction can occur between any three variables (A-B-C) • A “Four-Way” Interaction can occur between any four variables (A-B-C-D) • If you have 2 variables, then possibility of: 1 Two-Way (AB)
Interactions • There are different levels of Interactions. • A “Two-Way” Interaction can occur between any two variables (A-B) • A “Three-Way” Interaction can occur between any three variables (A-B-C) • A “Four-Way” Interaction can occur between any four variables (A-B-C-D) • If you have 2 variables, then possibility of: 1 Two-Way (AB) • If you have 3 variables, then possibility of: 1 Three-Way (ABC) 3 Two-Way (AB, AC, BC)
Interactions • There are different levels of Interactions. • A “Two-Way” Interaction can occur between any two variables (A-B) • A “Three-Way” Interaction can occur between any three variables (A-B-C) • A “Four-Way” Interaction can occur between any four variables (A-B-C-D) • If you have 2 variables, then possibility of: 1 Two-Way (AB) • If you have 3 variables, then possibility of: 1 Three-Way (ABC) 3 Two-Way (AB, AC, BC) • If you have 4 variables, then possibility of: 1 Four-Way (ABCD) 4 Three-Way (ABC, ABD, ACD, BCD) 6 Two-Way (AB, AC, AD, BC, BD, CD)
Between • IV1 (sex: male, female) • IV2 (victim of crime: yes, no) • DV – errors1a
Mixed • IV1 (sex: male, female) between • IV2 (pretest, posttest) within • DV – errors1a