Flashcards
HS 2801 MIDTERM
Cards in this deck (129)
- 1what are the two types of health researchphilosophical and applied
- 2what are the three categories of health research1. clinical
2. population
3. biology (basic) - 3five steps of the research process1. study question
2. select study approach
3. design and collect data
4. analyze data
5. report findings - 4a focus question should consider...EDPs
- 5what does PICOT stand forpopulation, intervention, control, outcome, timeframe
- 6a good research question is...1. real
2. testable
3. generalizable
4. purposeful - 7a good specific objective (aim) is...SMART
- 8what does SMART stand forspecific, measurable, attainable, relevant, timely
- 9what is a study goalsingle overarching idea
- 10what is a theoretical frameworkset of established models
- 11two types of objectives1. sequential
2. independent - 12what were the 4 model examples given in the lecturelalonde, evans and stoddart, eco-social, conceptual
- 13what are the main types of factors1. protective
2. risk (modifiable/non-modifiable) - 14what does FINER stand forfeasible, interesting, novel, ethical, relevant
- 15EDPsexposure, disease, population
- 16what is translational researchbridges the gap bw basic and clinical
- 17what is samplinghow data is collected
- 18whats an exception to samplingcensus
- 19what are the 4 population groups used in sampling1. target
2. source
3. sample
4. study - 20what are the 4 types of probability sampling1. simple random
2. stratified
3. cluster
4. multistage - 21what are the 2 types of non-prob sampling1. convenience
2. purposive - 22what is sample sizenumber of observations in a data set
- 23what is a type 1 erroralpha
false positive
jump the gun - 24what is a type 2 errorbeta
false negative
miss the boat - 25what is the power of a study(1-B)
how much a study is able to detect a true relationship - 26what is probingprompting to provide a better more in-depth answer
- 27what are the two types of surveys1. interview
2. self-admin - 28what is uniformityeveryone and test is done the same
- 29what is habituationtoo many agrees in survey, whats one more lol
- 30what is sampling bias (ascertainment bias)study pop is not representative of the source pop
- 31what is healthy worker biaspop of employed people is usually healthier
- 32what is exclusion biasdifferent eligibility criteria for cases and controls
- 33what is participation rate% of members of a sample pop who are included in a study pop
- 34what is a censuscomplete enumeration of a pop
- 35what are the 5 types of studies mentioned in lecture most often1. case-control
2. cohort
3. experimental
4. qualitative/quantitative
5. cross-sectional - 36what is a confidence intervalstat estimate of the range of likely values of a parameter in a source pop based on the value of that stat in a study pop
- 37a narrow CI indicates...more certainty about the value of the stat
- 38a wide CI indicates...may indicate a small sample size
not as certain - 39a study with a 20% likelihood of type 2 error have a power of...80%
- 40what are the 8 types of questions used in questionnaires1. numeric
2. date
3. yes/no
4. categorical - no rank (nominal)
5. categorical - ranked (ordinal)
6. paired comparisons
7. rank ordering
8. open-ended - 41a categorical question that doesn't using a rank system is classified as...nominal
- 42a categorical question that does using a rank system is classified as...ordinal
- 43what is the likert scaleordered responses to a questionnaire item
- 44what is a filter questiondetermines eligibility to answer certain q's
- 45what are the 4 types of exposure measurement1. available dose
2. admin dose
3. absorbed dose
4. active dose - 46which dose shows actual affects to the target organactive dose
- 47what is countnumber of occurrence of an event
- 48what is a ratio2:5, etc
- 49what is a proportionbetween 0-1, 5/25
- 50what is oddsprobability of an event occurring relative to it not occurring
- 51what is ratespeed of an occurrence over time
- 52what are the 4 mathematical measures of disease occurrence1. PR
2. IR
3. RR
4. OR - 53what are the 2 types of prevalence rates1. point
2. period - 54what is the difference bw incidence and prevalenceI: frequency of disease (tap)
P: population disease status (sink) - 55how does point PR differ from period PRperiod PR measures over a spec. time period (i.e. all of 2023)
- 56what is PRproportion of a pop that has a given disease at a given time
- 57what is IRproportion of a pop at risk that develops a given disease or attribute during a specific time period
- 58what is RRhow many times as likely it is someone who is exposed to something will have that adverse outcome compared to someone not exposed
- 59what measures the strength of an associationRR
- 60what is used to calc RRany measure of disease occurrence
- 61what is an odds ratioodds of a disease in exposed / odds of a disease in unexposed
- 62what are the 3 types of error1. random
2. systematic
3. both - 63what is random error caused bychance
- 64what is systematic error caused byrecognizable source (bias, etc.)
- 65another word for reliabilityprecision
- 66another word for validityaccuracy
- 67what are sources of measurement error1. interviewer
2. participants - 68internal consistencypresent when ALL ITEMS in survey instrument measure various aspects of the same concept
- 69intercorrelationpresent when 2+ related items in survey instrument measure various aspects of the same concept
- 70how do internal consistency and intercorrelation differinternal consistency is present when ALL survey items measure various aspects of the same concept, not just 2+
- 71how is internal consistency testedmeasuring intercorrelation
- 72how do you measure intercorrelation (2 ways)1. cronbach's alpha
2. kruder-richardson formula 20 (KR-20) - 73what does an intercorrelation score near 1 meanminimal random error and highly reliable
- 74what is test-retest reliabilitythe same test performed at a later time yields the same results
- 75what does interobserver agreement describethe degree of concordance among the independent observes for the same study participants
- 76what does kappa stat (cohen's kappa, k) determineif 2 assessors who eval the same study participants agreed more often than expected by chance or if its truth
- 77a k = 0 points to observers agreeing...as often as expected
- 78agreement on 100% of answers bw both observers would yield a k = ....k = 1
- 79if observers agreed more than expected, we would see a k value of...between 0-1
- 80what is content validity (logical)present when subject-matter experts agree that a set of survey items captures the most relevant info about study domain
- 81what is face validitypresent when content experts and users agree that a survey instrument will be easy for study participants to understand and do
- 82what is principle component analysis (PCA)creates more than 1 index variable (components) from a larger set of measured variable
- 83what is construct validitypresent when set of questions in survey instrument measures the theoretical construct that the tool is intended to assess
- 84what is factor analysisuses measured variables to model a latent variable that represents a construct that cannot be directly measured with 1 questions but appears to have a causal relationship with a set of measured variables
- 85what is convergent validitywhen 2 items that the underlying theory says should be related do show a correlation
- 86what is criterion validity (concrete)uses established test or outcome as standard for confirming the utility of a new test that examines a similar theoretical construct
- 87what is concurrent validityevaluated when part in pilot complete both the existing and new tests and the correlation bw the test results is calc
- 88what is predictive validityappraised when new test is correlated with subsequence measures of performance in related domains
- 89what are indicators of the quality of measurementreliability and validity
- 90what is a true scorethe score that a person would receive if the measurement was perfect
- 91what is the range of reliability coeff0.00 to 1.00
- 92a higher reliability coeff means0.00 to 1.00 → higher = better
- 93how does one test stability of a studytest-retest
- 94what is a critical component of test-retesttiming of the following retest
- 95what is internal consistencyestimate of the equivalence of sets of items from the same test
- 96what is the coeff of internal consistencycronbach's alpha
- 97what is cronbach's alphaavg intercorrelations of items and the number of items in the scale
- 98with more numbers in a data set, you would assume cronbach'a alpha to be...higher
- 99what is interrater reliabilityequivalence of ratings when instrument used by different observers
- 100what is cohen's kappaCategorical variables (coeff of agreement)
- 101what is responsivenessability of a measure to detect change over time in the construct of interest
- 102what are three tests of interrator reliabilityKappa (binary), percent agreement OR kendall tau
- 103Generalizability coefficientquantify how much measurement error can be attributed to each potential factor
- 104Measurement of validity of an assessment instrument?Content
Response process
Relationships to other variables
Consequences - 105two misclassification typesdifferential
non-differential - 106what is misclassificationclassification of part. In the wrong exposure or outcome groups
- 107what are the two types of validity1. internal
2. external (generalizable) - 108steps to observing an associationStep 1. Due to chance?
Step 2. Due to bias?
Step 3. Due to confounding?
Step 4. TRUE - 109what are 3 sources of error in validity1. Chance (random sample error)
2. Bias → selection, information
3. Confounding - 110what is random sampling errorvariability in sampling due to chance
- 111what is the source of random sampling errorselection technique where probability of selecting each sample is known
- 112how do you minimize random sampling errorlarger sample size
- 113what are the types of selection bias (3)1. control-selection
2. differential participation
3. differential loss to follow up - 114what are the 5 examples of selection bias as seen in the lecture1. volunteer
2. non-response
3. membership
4. loss to follow up
5. berkson's - 115how do you reduce selection bias?cannot be reduced like rse but can be avoided through careful design and conducting study
- 116what are the two overarching types of bias1. selection
2. information - 117what is bias caused by on a big scalesystematic errors
- 118what is publication biaspapers with better findings get published more
- 119what are the third variables1.. confounder
2. modifier
3. mediator - 120what is a confounderdistortion of actual association due to mixing of effects between the exposure and an incidental variable
- 121what type of error is a confoundersystematic
- 122what is the three step criteria to definding a confounding variable1. Causally associated w outcome
2. Noncasually or casually associated w exposure
3. Not an intermediate in the pathway - 123what are the three ways to identify a confounding variable1. lit review
2. consult experts
3. statistical test, i.e. stratification - 124how does one deal with a confounder in the DESIGN stage1. restriction
2. matching
3. randomization - 125how does one deal with a confounder in the ANALYSIS stage1. standardization
2. stratified analysis
3. include confounder in multivariate regression model - 126what is effect modificationsomething that modifies the effect
- 127how do you deal with an effect modifierjust a fact of life
don't deal, just report and highlight - 128what is mediationsomething on the pathway to the outcome
- 129what are surrogate interviewsinterview w family, spouse, etc after someone has died
could cause inaccurate info
Information
- Description
- HS HS 2801 Flashcards on HS 2801 MIDTERM, created by Holly White on 09/10/2024.
- Created by
- Holly White
- Views
- 1
- Rating