L2: Sampling Flashcards
Sampling
selecting units from a population to estimate its characteristics
Population
Aggregate of units under examination
Why sample?
very large population or if investigation is destructive, expensive, time consuming, and labourious
Random Sampling
units are selected randomly (purest form of probability sampling, each member has an equal chance of being selected)
Stratified Sampling
When population is split into several groups of units, called strata, based on characteristics. Units are randomly selected from each stratum.
Systematic Sampling
First unit selected randomly, then consecutive units are chosen at specified intervals.
Authoritative Sampling
Person makes an educated guess on which units are best to sample
Criteria for selecting sampling plan
objectives, cost, environment, patterns of environmental contamination, site considerations, public concerns
Parameter
refers to population characteristic under examination (mean, sd, etc)
Statistic
refers to sample parameters used to estimate population characteristics (sample mean, sample error)
Bias
refers to how far the average statistic lies from parameter it’s estimating (Error that occurs when estimating a quantity)
Sampling Distribution
Probability distribution, under repeated sampling (numerical quantity calculated from data values in sample)
Probability Distribution
Mathematical definition used to ascertain probability of population parameter to acquire specific value or lie within specific range of values (discrete, continuous)
Distribution
normal, t, F, x2, exponential, lognormal, bionomial, poisson
Random Sampling (rectangular plot)
choose 2 random numbers from table (U1, U2), use formulas x=XU1, y=YU2
Random Sampling (Circular Plot)
choose 2 random numbers, multiply by 360 and R (A=360*U1, r=RU2)
Random Sampling (Composite field samples)
choose 10 lab samples from N number of field samples, pick random number from table t, use formula; chosen sample = t*N+1
Random number tables
table of random numbers (choose random starting value; move horizontally to the right, skipping alternate numbers or move vertically down, picking every 5th number)
Hot Spot
contaminated area/pocket in defined environment
Why is systematic grid sampling preferred for locating hot spots
efficient in finding patterns and locating rare events; flexible and more defensible sampling designs
Hot Spot sampling assumptions
assume hot spot is circular or elliptical, samples are taken on square/rectangle/triangular grid, distance between grid points are larger than sampled area
Hot Spot considerations
hot spot shape, length of longest axis of hot spot, level of certainty to confirm there is no hot spots at site
Grab Sample
sample collected at specific time and place
Composite Sample
mixture of samples collected at different times from same sampling point (useful for observing average concentrations, saves time and money)
