Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
12 Cards in this Set
- Front
- Back
use correlational design when..
|
IV cannot be manipulated
want to test association between/among variables in everyday life focus is on predictions and outcomes, not casual rel. |
|
use exp design when..
|
IV can be manipulated
underlying casual relationship is goal enable cause and effect |
|
Between subjects Design basics
|
random assign. to control and exp.groups
each group administered different level of IV @least 2 levels of IV.. each participant exposed to only 1 leve of IV Each participant measured once on the DV |
|
Within Subjects Design basics
|
1 group of participants recieves all levels of the IV
random assign.not necessary @least 2 levels of IV..each participant exposed to all levels of IV Measured multiple times on DV "repeated measures" |
|
subject variables
|
cannot be manipulated
|
|
Manipulated variables
|
enviornmental: create different conditions by modifying participants' physical or social enviornment (stimuli, confederates)
Instructional man: create different conditions by varying the instructions taht participants recieve Invasive man: create different conditions by using sugery or drugs to create physical changes in participants bodies |
|
3 possibilities of differences attribute to IV
|
1. no difference (null)
2. difference due to IV 3. difference due to confounds |
|
extraenous variables
|
contribute "noise" and do not affect DV, like confounds
random influences that cause some participants within group to score higher or lower than they normally would 1.stable individual differences among participants 2.unstable indiv. differences among participants 3.differences in how exp. treats participants 4.how participants react to exp.setting 5.measurement error |
|
difference due to chance
|
can never say for certain difference is NOT due to chance
if prob that the difference is due to chance is less than 5%, the difference between the 2 groups is 'statistically sig' |
|
F-ratio
|
between group variance/within group variance=
treatment effect+chance variation/ chance variation chance variation=error variance; due to extraneous variables, not confounds |
|
F-ratioo has no effect
|
when treatment effect=0
larger the F-ratio, more likely there is difference between the 2 groups.. more likely we can reject the Null that groups are not different |
|
To increase size of F-ratio..
|
minimize within group variance
maximize between group variance |