RM 12 Flashcards
interaction effects
occurs when the effect of the original variable (such as cell phone use) depends on the level of another variable (driver’s age)
ex: does the effect of cell phones depend on age?
“difference in differences”
difference 1: between cell phone and control conditions
might be different 2: for older drivers than younger drivers
“the effect of one independent variable depends on the level of the other independent variable”
Crossover interaction
lines cross each other parallel
do you like hot foods or cold foods? : it depends
ice cream: don’t like it hot, like it cold
pancake: don’t like it cold, like it hot
spreading interaction
lines are not parallel, do not cross each other: pattern can be described with the phrase “especially”
sandwiches are better when you add ham, especially if there is lettuce
factorial design
2 or more independent variables, referred to as factors are crossed, (multiplied) in order to study each possible combination of the independent variables
cells of a factorial design
the outcome of the factorial design
For 2 independent variable: 4 cells (2x2 factorial design)
participant variables
a variable who’s level is selected, not manipulated (such as age), not truly an independent variable but treated as such for simplicity such as age, gender, ethnicity
difference of the differences
ex: the difference between using cell phone while driving and control conditions is different for older people
factorial designs is to test…
limits: whether an independent variable affects different kinds of people in the same way = form of external validity
they are testing whether the effect generalises
moderators
variable that changes the relationship between two other variables
= independent variable that changes the relationship between another independent variable and a dependant variable
the effect of one independent variable depends on the level of another independent variable
Interpreting factorial results: main effects
each independent variable is tested for main effects
- overall effect of one independent variable on the dependent variable, averaged over the levels of the other independent variable
Calculated by taking the MEAN (also called marginal mean) of each independent variable, always located on the far-right column, or bottom row of a contingency table : marginal means
Researchers use statistics to know if the marginal mean is statistically significant, cause it could look not statistically significant and is
interaction effect
Table : difference between the two levels of the independent variables
there is an interaction effect if the difference between these results is statistically significant
Graphs: if parallèle, no interaction
Bar graphs: if the lines at the top are parallèle, then no interaction
variations on the 2x2 factorial design
a. independent groups factorial design
b. within-groups factorial designs
c. mixed factorial designs
Independent groups factorial design
between subject, in 2x2 there are four seperate groups of participants, one for each condition
within groups factorial design
within group, there is only one group of participants that passes through all the conditions
also called a repeated-measures factorial
requires fewer participants
mixed factorial design
one independent variable is manipulated as independent groups, and the other in manipulated as within-group
There is 2 distinct groups of participants
ex: older age would pass through placebo and taking drugs
young age : placebo and taking drugs
