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;
69 Cards in this Set
- Front
- Back
Concept
construct |
abstract idea, 'chair'
Highly abstract concepts, leadership and emotions |
|
operationalization
|
Take concept or construct and turn into reality
|
|
Measurement
|
assigning numbers
quantify classify compare |
|
Attribute
|
feature of object along various dimenions
?english |
|
Problems with messurment
|
representation
objectivity correzpkndence |
|
common error sources
|
am i really gonna write this
situatiinal response bias transitiry factors administration variation intstrument clarity reresponse sampling instrument format |
|
response bias
|
recall
option fake good or bad end aversion framr choices healthstate bias |
|
Validity
reliability |
How well messures what we want
accuracy and precision of measure cannot be valid and unreliable |
|
external validity
|
ability of research to be generalized across persons, settings and times
|
|
internal validity
face validity |
degree which treatment alone brings change
look at whether seems like good translation of construct, weak way |
|
Content validity
|
measuring instrument provide coverage of topic
|
|
criterion related validity
|
compare urs against criterion
|
|
crit related
predictive validity |
ability to predict what it should be able to predic
|
|
crit relsted
concurrent validity |
ability to distinguish groups that should be distinguishable
|
|
Crit related
convergent validity |
degree to which operationalization is similar to other ones it should be similar tooooo
|
|
Crit related
discriminant validity |
degree which operationization is nit similar to othrr operatikniship that it should not be simlialr too
who the fuck comes up with this shit |
|
ingernal
construct validity |
cAn inferences be made from operationalitions to constructs they were basd
need high convergent and discriminant validity |
|
threats internal validity
|
maturation
history instrumentation placebo or hawthorne effect dropout bias subject selection bias |
|
reliability
|
stabiliyy
equivalence internal consistency |
|
stability
|
instrument is stable if consistant results w/ same subject
|
|
Equivalence
|
considers how much error may be introduced by diff investigators in observation
|
|
internal consistency
|
consistency among items of a scale
|
|
attitudes
|
enduring patterns of behavior which are believed to be predictable of behavior
|
|
scaling
|
construction of instrument that associates qualitative constructs with quantitative units
aka how much pain scale 1-10 (pain abstract but scale is not) |
|
Thurstone (equal appearing interval scaling)
|
2 answer choices,
usually agree or disagree with large set of statements |
|
Guttman scaling
|
also called cumulative scaling
one dimensional continuum for concept being measured one concept, limited use series of items with increasing intensity add all scores of every item agreed with |
|
Pair comparison scale
|
can compare bunch of items but do it one by one
1 , 2, 3, 4 1v2, 1v3, 1v4 2v3, 2v4 3v4 |
|
rank order scales
|
rank list of stuff say 1-7 for 7 items
each # only once |
|
Q sort technique
|
to discriminate among large groups objects
object sorted into pile of preference, piles rank in order of preference items evaluated relative to one another each statement based on diff issue sort items according to which most agree with +5 and least agree -5.... |
|
Constant Sum scale
|
distribute 100 points into these 5 categories based on importance
|
|
line marking scale
|
where are you on this line,
hate stats-----------------------really hate stats |
|
smiling faces
|
how high are you right now
:( :/ :l : ) ;p |
|
Likert scales
|
provided with range of responses to series of +ve and -ve statements
scores associated with each possible response |
|
Semantic differential scale
|
positions are developed within predefined set of pair polar opposite adjectives
You are more? Honest..............................Dishonest |
|
Stapel Scale
|
Rxn to commercial
+3 +2 +1 Exciting -1 -2 -3 based off one emotion or thingy |
|
IRB what the hell is it
|
group individuals (not all healthcare) that evaluate protocals for studies assess risk and benfits
aka local ethic committee Duties- Protect humor, informed concent, documentation |
|
HISTORY IRB
|
Numemberg code
Declaration of Helsinki National Commission Protect Human Subjects Belmont Report Code of Fed. REgulations Health insurance portability and accountiability act (HIPPA) |
|
Nurenberg code
|
from WW II crimes
most important document Voluntary consent!!!! and other ones |
|
Declaration of Helsinki
|
World medical association
health of pt first consideration |
|
National Commission for the protection of Human Subjects
background |
Tuskeke study-blacks syphalis and didnt get full disclosure
guidelines were developed for diff stuff IRB 1978 |
|
Belmont Report
|
Respect for persons- ackknowledge authority
Beneficence- no harm, more benefits Justice- each person equal share |
|
Code Fed. Regulations
|
Created IRB
Office for Human Research Protection -its job...yea |
|
HIPPA
|
1996- health insurance coverage
2003 securitty and privacy of health data |
|
IRB
Responsibilites |
informed concent
not put ppl at risk |
|
IRB documentation
|
proposal forms first
minutes reviews correspondence |
|
Req for study approval, IRB stuff
|
minimal risk to pt
risk/benefit reasonable written consent data monitoring confidentiality add. stuff for vulnerable ppl |
|
Informed consent
|
legal document
must be done before begin -elements provision essential info -concent must be voluntary children, adolesants, adult all have diff ones |
|
NSU ethics board
|
Chair
Vice chair at least one rep from each college Dr Seamon and Franco |
|
Apllication for IRB/ ethics board useful
|
HIPPA
CITI |
|
IRB test center req types
|
any research w humans
need center rep college lvl review expeditied review full review |
|
Ethical Principals
|
Autonomy
Non-malefience beneficence distributive justice truth telling professional responsibility |
|
Stat Sampling Methods! where in project is it
|
Project starts
hypothesis after literature sampling then methodology and stat analysis |
|
Population vs
Sample vs Sampling |
p- entire set ppl common characteristic
s-subset of pop selected for study s- selecting portion of pop to represent entire |
|
representation
variation |
sample should be highly similar to target pop
variationin should be reflected in sample if in pop |
|
Advantages of sampling
|
Time saving
money saving quality assurance data greater response rate |
|
technical terms
-sampling element -sampling frame |
-each entity from pop that is ultimate sampling object
-complete list of all units from which sample drawn (not sample yet) |
|
Methods of sampling (2)
|
Probability sampling- random selection choosing unit
x-simple random, stratified random, custer sampling, systematic sampling Nonprobability sampling- non random selection x-convenience sampling, quota sampling, purposive sampling |
|
Randomization selection
Randomization assignment |
- each element in pop has equal chance be selected
-allocate subjects to diff. experimental conditions on random basis good for prospective studies |
|
Simple Random Sampling
|
each sample element in frame has EQUAL OPPORTUNITY of chosen once
Procedure->frame->random selection from frame -ve- inefficeient, if frame off big issue |
|
stratified random sample
|
random select from 2 or more STRATA of pop independantly
strata-subdivision of pop by characteristic can be proportional or disproportional possible bias |
|
Cluster
|
multistage w/ large groups selected first then successive subsampling
ex nationwide limit-more sampling errors |
|
Systematic sampling
|
selection where every Nth person is chosen
|
|
Non probability sampling (3)
|
Conviencience
quota purposful |
|
Conveince sampling
|
easist ppl availible
Method- accidental, snow effect, network Limit- weak, lack control |
|
Quota sampling
|
nonrandom with PRESPECIFIED CHARACTERISTIC of sample
lack control variability |
|
Purposive Sampling
|
selected on basis personal judgement on which most productive
Method-personal knowledge on issue of study Limit-lack of controlling variability |
|
Sample Size whats important about it
|
POWER- ability of design to detect relationship tween variables
EFFECT SIZE- (r) stat representation of magnitude of relationship tween variables magnitude of diff tween 2 groups |
|
FORMULA FOR EFFECT SIZE know it
|
r = U1-U2 / SD
U's- diff in pop. mean SD- standard dev. |
|
know how to use the efect size vs power chart, when number is found what must be done?
|
X2
r=.2 small .5- med. .8 large |