Overview
Election polling is the process of collecting the results of a survey of voters with the intent of making predictions about an upcoming election. It is an example of Random Sampling and Polling, but comes with a set challenges that are specific to elections.
Types of Samples
There are various ways to sample a population that are designed to overcome the challenges of random sampling. (see below)
- Simple Random Sample - the basic and simplest form of sampling where the sample is a set of n units from the original population where the probability of any set is the same as any other set of the same size.
- Stratified Random Sample and Aggregated - stratified sampling occurs when the original population is segmented into strata first, and then random samples are taken from each strata.
- Cluster Sample
Challenges
- Limited Sample Size
- By definition, a poll computes a set of statistics from a sample that only represents a small sub-population of the true
population. As such, numbers such as averages will in general not equal the true population average.
The standard way of dealing with the differences between the poulation and average the sample average is to use the tools of statistical inference to create confidence intervals around the estimated value in order to understand how much confidence can be given to the computed statistics. - Selection Bias Selection Bias occurs when the sample that is drawn to represent a population does not match the population in question. As a general rule, the sampled population will never match the population at large due to random factors, however, selection bias is usually concerned with systemic factors which tilt every sampled population away from the true population.