class 10 Flashcards
target population
selecting a group of people, events, or behvaiours, or other elements with which to conduct a study
element (aka case)
an individual unit of a population:
-person(subject)
-event
-behaviour
sample
the selected group of elements from which data are collected for a study
target population
the entire set of indviudals (or elements) that met the sampling criteria (the P in PICO)
“PICO”
P-population
I-intervention of influence
C-comparison
O-outcome
target population
group to whom the researcher aims to generalize transfer/findings
accessible population
the portion of the target population to which the researcher has reasonable access
generalizability
the degree to which data are representative of similar phenomena in a population beyond the studied sample
representativeness
-key characteristics of the sample are similar to those of the population
-sample,accessible,population, and the target population are alike in as many ways as possible
representativenes needs to evaluate:
-setting
-characteristics of the subjects: age, gender, ethnicity,income, education
-distribution of values of variables
homogenity
degree to which subjects are similar
heterogeneity
degree to which subjects are different
what do homogeneity and heterogeneity impact
representativeness and generalizability
what is sampling criteria
the characteristics essential for inclusion in the target population
-consist of: inclusion (eligibility) criteria and exclusion criteria (delimitations)
sampling criteria examples
i.e. between the ages of 18 and 45, ability to speak english, diagnosed with diabetes in the last month
sampling error
-when the value of a statistic fluctuates from one sample to another drawn from the population
i.e. difference between the population mean & the mean of the sample
example: study looking at hairloss, ages 18-65 specified
sampling frame
-a listing of every member of the population, using the sampling criteria to define membership in the population
-subjects are selected from the sampling frame
sampling plan
outlines strategies used to obtain a sample (who, where, when, how)
ex: sample will randomly chosen from all patients with the diagnosis of subarachnoid hemorrhage in the past 10 years
probability (random) sampling
random selection methods
-every individual in the population should have an opportunity to be selected for the sample
-advantage: increases respresentativness
-disadvantage: limited feasibility
nonprobability sampling
nonrandom methods
probability sampling (simple):
-elements selected randomly from sampling frame (using computer software, number table, drawing names)
probability sampling (stratified):
-used when certain variables are considered critical to representativness (e.g. age, gender, geographic location)
-elements organized into strata - random selection from each stratum
probability sampling (cluster):
-elements drawn from chosen “mini-representations” or the target population (e.g. cities, provinces)
-can occur in one stage or multiple stages
probability sampling (systematic):
-selecting every nth element from a complete list of the population using a random start point
nonprobability (nonrandom) sampling:
-not every element in the population has an opportunity to be included in the sample
advantage of nonprobability (nonrandom) sampling:
increased feasibility (compared with random sampling) in healthcare research
disadvantage of nonprobability (nonrandom) sampling:
increased likelihood of obtaining a non-representative sample
nonprobability (nonrandom) sampling (convenience):
-elements are enrolled based on availability
-researcher enrolls until desired sample size is achieved
nonprobability (nonrandom) sampling (quota):
-similar to stratified sampling, but inital sample is not random
nonprobability (nonrandom) sampling (network (snowball)):
-initial subjects are enrolled by convenience, then additonal subjects are identified through association (e.g. friends, colleagues)
nonprobability (nonrandom) sampling (purposive):
-researcher selects subjects based on judgement about who will provide the most useful data to answer the research question
factors that influence sample size
-power
-effect size
-type of study conducted
-number of variables
-measurement sensitivity
-data analysis techniques
-frequency of the phenomenon
-cost
power
0.80 is minimal acceptable power level
-capacity of the study to detect actual differences aka capacity of the study to correctly reject the null hypothesis
-# of participants needed to avoid a type ii error in a comparative study
effect size
a statistical expression of the magnitude of the relationship between two variables or the difference between two groups
-0.2=small effect->larger sample needed
-0.5=medium effect
-0.8=large effect->smaller sample adequate
power analysis
-standard power of 0.8
-lvl of significance=0.05,0.01,0.001
-effect size=0.2 small, 0.5 medium, 0.8 large
