[2] Class 24-25 Cross-section Flashcards
Study frame
The focus of the study
Obs studies that examine relationships of health/Dz to other variables of interest at the SAME TIME
2 names:
Cross-sectional studies
Or Prevalnce study
Entire population or a subset is selected for study
“Snap-shot” in time
Information gathered represents what is occurring at a point in time or timeframe a-cross a large population
Mostly conducted over one year period
Cross-sectional study
Study based on Dz status
Case-control study
Brought ppl into study based on exposure or not exposed or b/c they have something in common
Cohort study
Focuses simultaneously on Dz and population characteristics, including exposures, health status, health-care utilization, etc.
Seeks asso. NOT CAUSATION
Generates and test hypotheses
By reptition in different time periods, can be used to measure change/trends (not in same pt,’s
Cross-sectional studies
Wonderful at generating future research hypotheisis:
Generate ideas based on what is seen in large population
Cross-sectional study
What form are cross-sectional studies usually conducted in different aspects of US pop.
2 forms:
Surveys
Databases
2 cross-sectional approaches:
- Collect data on each member of the population
2. take a sample of the population and draw inferences to remainder [generalizable]
City/state uses this type of cross-sectional approach most commonly:
Collect data on EACH member of the population- in MO, in KC, etc.
Most frequent approach in Cross-sectional approaches in huge populations, like the entire nation, would be:
Taking a sample of the population and draw inferences
It would be impossible to collect data on each and everyone in the US for study
Types of sampling schemes:
Probability samples
And
Non-probability
Type of sampling:
Every element in the population has a known (non-zero) probability of being included in sample
Most common form of sampling
Probability samples
Ex of probability sampling
Simple Systematic Stratified simple Stratified disproportionate Multi-stage Cluster multi stage
Assign random numbers then take randomly selected #s to get desired sample size
OR
Assign random number then sequentially-list numbers and take desired sample size from top of listed numbers
Simple random sampling
Assign random numbers then randomly sort these numbers , the select highest (or lowest) number, then systematically, then by a predetermined sampling INTERVAL
Take every Nth # to get desired sample size
Systematic RANDOM sampling
Stratify sampling frame by desired characteristic [ex;gender] then use SIMPLE RANDOM sampling to select desired sample size
Stratified simple random sampling
Disproportionately utilizes stratified simple random sampling when baseline population is not at the desired proportional percentages to referent population
Stratified sample “weighted” to return sample pop. Back to baseline pop.
Useful for over-sampling
Stratified disproportionate random sampling
Which sampling style is used for minority individuals in studies?
Stratified Disproportionate random sampling
Uses simple random sampling at MULTIPLE stages towards pt. Selection
Used in counties, city blocks, zip codes, clinics, hospitals, individual, individual chart
Multi-stage random sampling
Same as multi-stage random sampling but all elements clustered together [at any stage] or selected for inclusion
Ex:
All clinics in a zip code
ALL households in a community
Cluster multi-stage random sampling
“Quasi-systematic’ or ‘convenience’ samples [not really, completely random or fully probabilistic]
Decide on what fraction of pop. Is to be sampled and how they will be sampled
Ex: all ppl w/ last name M-Z, ppl in business asso. , all ppl attending clinic every M/W/F
Non-probability sampling schemes
What is the concern w/ non-probability sampling schemes?
There is some known or unknown order to the sample generated by selected scheme which may introduce bias [selection bias]
What bias is of concern for non-probability sampling schemes?
Selection bias