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;
15 Cards in this Set
- Front
- Back
|
3 characteristics of a well-designed experiment |
1. vary at least one independent variable 2. assign subject to conditions 3. control extraneous variables |
|
|
1. vary at least 1 IV |
-at least 2 or more levels: qualitative vs. quantitative -types of IVs: environmental, instrumental, invasive -experimental vs control group |
|
|
environmental IV difference |
manipulation of Ss physical/social environment (ex. noise level, temperature) |
|
|
instrumental IV difference |
difference in verbal instructions given to Ss |
|
|
invasive IV difference |
creating physical changes in Ss body (ex. more or less sleep) |
|
|
2. assign subjects to conditions (levels) |
must first randomly sample Ss want Ss to be equivalent --> random assignment = used to equate different experimental conditions before manipulations --- random assignment equates 2 groups on every dimension imaginable - increases internal validity |
|
|
simple random |
every Ss has an equal chance of being in one of levels of the experiment |
|
|
matched random |
researcher tries to increase similarity among groups ----> try to identify similar individuals to place in different conditions in order to minimize chances of person confound |
|
|
person confound |
happens in non-experimental research (ex. income) seems to cause something because ppl high or low in variable also happen to be high or low in some individual difference variable (ex. education lvl) that is associated with outcome variable of interest (ex. IQ scores) -eliminated with random assignment |
|
|
within subjects / repeated measures design |
no random assignment Ss is in all conditions of experiment |
|
|
pros & cons of within/repeated measures design |
PROS -more powerful -more generalize than between sub -can measure direct changes -fewer Ss, testing same people CONS -order effects: order in which one is exposed to IV's can affect results |
|
|
3. Control extraneous variables |
variance total = systematic/between groups Ss + error/within SS sources of systematic/ between groups Ss variance = treatment variance (try to max) and confound variance (eliminate) factorial design |
|
|
sources of error variance |
individual differences transient states environment differential treatment |
|
|
factorial design |
2 or more IVs manipulated -main effect = action of single IV -interaction = test for relationship between effects of different IVs -total # groups = times all IVs together |
|
|
6 commandments of being a good experimenter |
1. suave, professional 2. nice friendly 3. educated 4. honest 5. good liar 6. attentive |
