Statistics: Sampling Methods - YouTube

Welcome to a lesson on sampling methods. In this lesson

we'll define the different types of sampling methods when conducting a survey or

poll.

The first thing we should do before conducting a survey,

is to identify the population that we want to study. Once we identify the

population,

we need to select a survey method. However, a sampling method is biased

if every member of the population doesn't have

equal likelihood of being in the sample. The natural variation of samples

is called sampling variability. This is an avoidable and expected in

random sampling.

And in most cases is not an issue. So we can't avoid

sampling variability but we should try to avoid

sampling bias. The first method we'll discuss

is a random sample, where a random sample is one in which

each member of the population has an equal probability

of being chosen. As an example, every member of the population has an ID.

A computer generates random ID's, and the members with the

random ID's are surveyed. This is an example

of a random sample. Related to random sample,

is the simple random sample, also known as

SRS. A simple random sample

is one in which each sample of a particular size

and is selected, such that every possible sample

of size n, has the same probability

of being chosen. So notice for a random sample,

we're selecting individual members of the population,

but for a simple random sample, we're selecting a sample size

of n. To help account for variability,

pollsters may instead use what's called a stratified sample.

In stratified sampling, the population is divided into a number of subgroups,

or strata.

Random samples are then taken from each subgroup

with sample sizes proportional to the size of the subgroup

in the population. Stratified sampling can also be used to select a sample

with people in desired age groups, political parties,

sex, etc. So an example of stratified sampling,

if thirty percent of employees in various age groups

given here in this table. In a sample of two hundred employees

they may randomly select thirty employees,

twenty to twenty-nine years of age. Because thirty,

would be fifteen percent of two hundred.

Seventy employees of age thirty to thirty-nine,

because seventy is thirty-five percent

of two hundred. Sixty employees,

age forty to forty-nine, because sixty is thirty percent of two hundred.

And finally, forty employees of age fifty to fifty-nine, because

forty is twenty percent of two hundred. Quota sampling is a variation on stratified sampling,

wherein samples are collected in each subgroup until the desired quota is met.

So once again, as an example, consider the percent of employees in various age

groups

as we saw in the previous slide, also provided here.

In the sample of two hundred employees, the pollsters will randomly select employees to

poll.

However, once thirty employees age twenty to twenty-nine are polled,

employees in this age group would no longer be polled.

Again we stop at thirty employees in this age group because thirty is fifteen percent

of two hundred, the sampling size. The same for the other groups until the quota for each age

group is met.

This method is easier than stratified sampling.

This method is easier than stratified sampling, but does not ensure the same

level of

randomness. Another sampling method is cluster sampling,

in which the population is divided into subgroups or clusters.

And a set of subgroups are selected to be in the sample.

As an example, if Taco Shack, a chain of one hundred-fifty restaurants

wants to survey its customers about a possible new menu,

management might randomly select fifty of the stores,

and conduct a survey. This would be cluster sampling.

Another sampling method is called systematic sampling. In systematic

sampling,

every n'th member of the population is selected to be in the sample.

For example, to conduct a survey of students at a college, a pollster might contact

every fiftieth

name on a student enrollment list.

Systematic sampling is not as random as simple random sampling.

If your name is Victor. If your name is Victor Ruiz,

and your sister is Victoria Ruiz, more than likely

you both would not end up in the sample, because your names would be right next to

each other.

Remember, for a sample not to be biased, everyone should have the same chance at

being selected.

However systematic sampling can yield acceptable samples.

Perhaps the worst types of sampling methods, are convenience

samples, and voluntary response samples.

Convenience sampling is when samples are chosen by selecting whoever is

convenient.

As an example, a pollster may go to a mall and pull the first fifty people

who agreed to participate. This is convenient sampling.

So while this method is usually quick and cheap, it often results in bias.

Voluntary response sampling is allowing the sample to volunteer.

For example, if you email out a poll or survey, those that respont are self-selected

or volunteers to participate. This is voluntary response

sampling. Usually voluntary response samples are skewed towards people

who have a particularly strong opinion about the subject of the survey,

or have a lot of extra time on their hands. I hope you found this overview

of the different sampling methods helpful. Thank you for watching.