Term
|
Definition
|
|
Term
|
Definition
|
|
Term
|
Definition
vary with independent variable and effects the dependent variable
ex: color confounded with sweetness bc color is varying with sweetness
-Always confounded with independent variable |
|
|
Term
|
Definition
doesn't vary with the IV but could effect the DV
-Random assignment prevents thrid variables from becoming confounding |
|
|
Term
|
Definition
causal relationship can be reversed
-Drink alcohol bc depressed or drepres |
|
|
Term
|
Definition
Given two ties theories, go with the simpler one.
-Occam's razor, Lloyd Morgan's canon |
|
|
Term
| Inductive vs. Deductive reasoning |
|
Definition
deductive: start with simple and go to specific
ex: all american beers are gross
-miller high life is gross- all beers must be
inductive: specific to general
bud light is gross
bud light is american
all american beers are gross |
|
|
Term
|
Definition
reliability: do you get the same measurement each time?
validity: does it measure what it is intended to? |
|
|
Term
|
Definition
null: status quo/ opposite of your hypothesis
-IV has no effect
Alternative: means groups are not equal
-IV does have an effect
-observed differences reflect real difference |
|
|
Term
|
Definition
Incorrectly rejecting the null hypothesis when it is in fact correct (gullability)
-assuming a drug your testing actually works when it doesnt |
|
|
Term
|
Definition
| Saying null is ture when in fact it is false (blindness to relationship) |
|
|
Term
|
Definition
alpha of .05 is a balance between type 1 and type 2
if alpha is .50, more likely to do type 1 errors
if alpha is .001 more likely to commit type 2 errors |
|
|
Term
|
Definition
| select randomly from the whole population |
|
|
Term
| stratified random sampling |
|
Definition
| create groups within the population and then randomly select same number from each group |
|
|
Term
| proportional random sampling |
|
Definition
| same as stratified but you match proportion of groups in the sample to that in population |
|
|
Term
|
Definition
convenience: take whoever you can get
ex- supermarket surveys
Quota sampling: create groups with population (make up quota) and then take whoever you can get within these groups |
|
|
Term
| between vs within designs |
|
Definition
between: everybody gets one treatment
within/repeated measures: everybody gets all treatments
-Counterbalancing sometimes needed for within |
|
|
Term
|
Definition
factor: independent variable
level: number of different conditions within a factor
(different conditions of a particular independent variable) |
|
|
Term
|
Definition
Manipulation of the IV
Control of extraneous variables
-Held constant
-Match levels of variable in all conditions
-Random assignment |
|
|
Term
|
Definition
2 or more Independent variables
2x2 designs
quasiexperimental |
|
|
Term
|
Definition
| Overall effect of each independent variable |
|
|
Term
|
Definition
the effect of one independent variable on the DV
-Depends on the level of the other independent variable |
|
|
Term
|
Definition
Organize and summarize data so easier to understand
-describes data in a sample |
|
|
Term
|
Definition
| makes inferences about the population based on sample |
|
|
Term
|
Definition
Measures the strength of relationship between two variables
-value gives strength
pos or neg |
|
|
Term
|
Definition
| tests differences in continuous DV between 2 groups |
|
|
Term
|
Definition
tests differences in categorical DV between 2 groups
-right vs. wrong, yes vs. no |
|
|
Term
|
Definition
| lower p value, higher significance and less likely results occurred by chance |
|
|
Term
| ways to keep science honest |
|
Definition
Peer review
Replication
Database repositories |
|
|
Term
| 3 principles for belmont report |
|
Definition
Respect for person
Beneficence- benefit must out way risks
Justice-equitable selection and recruitment and fair treatment of research subjects |
|
|
Term
| Active vs. passive deception |
|
Definition
Active: tell a lie
Passive: dont tell participants some info |
|
|
Term
| Confidentiality vs. Anonymity |
|
Definition
confidentiality: know youre identification but wont be shared
Anon: have your data but no id |
|
|
Term
|
Definition
| concerned with proper vs improper |
|
|
Term
|
Definition
| only show significant data, hide failures in the file drawer and dont publish them |
|
|
Term
|
Definition
The ethical issue requiring clinicians to provide the best possible treatment for their patients
-Thus limiting research to studies that compare equally preferred treatments |
|
|
Term
|
Definition
| The ethical principle requiring the investigator to provide all available info about a study so that a participant can make a rational, informed decision regarding whether to participate in a study. |
|
|
