Probability Sampling

Probability sampling is random sampling that can be done in three ways:

1) Simple Random Sampling
Samples are drawn randomly from the waste, ensuring that all parts of the waste have a statistically equal chance of being selected. With this method, each member of the population has a statistically equal chance of being selected as a sample, thus reducing bias in the sample. Simple random sampling is the best approach if little or nothing is known about the waste.

2) Stratified Random Sampling
When the waste exhibits a regularly occurring or predictable strata in the waste (e.g. phase liquids [oil & water] or liquid with sludges), samples can be drawn randomly from each of the stratum. This method ensures that all waste within a stratum have a statistically equal chance of being selected for the study. It is important to note that if the sampling strata differ from naturally present strata exhibited in the waste, then stratified random sampling can product less precise results than simple random sampling.

3) Systematic Random Sampling
In this approach, sampling points are predetermined based on the number of samples needed and one randomly selected point as the first sample to draw (e.g. creating a grid of the sampling area and taking a sample from every third grid). This method is useful for sampling wastes that are mobile or dynamic, such as discharge from a process (e.g., taking a total 20 samples every five minutes for one hour). A disadvantage of systematic random sampling are the poor accuracy and precision stemming from unrecognized trends or cycles in the population due to the sampling method.

Accuracy is the measure of how closely the data from the sample is to the true average properties of the waste.

Precision is the measure of variability between the results of sets of samples.