Scribd
Upload
216 views

2.3 Sampling Techniques

This document discusses different sampling techniques used in statistics. It defines key terms like population, sampling frame, and sample. It then describes several sampling methods: simple random sampling, systematic sampling, stratified sampling, cluster sampling, voluntary response sampling, and convenience sampling. For each method it provides an example and explains how it works and when it would be appropriate to use. It also discusses factors to consider for random sampling like nonresponse rates and sample size.

Uploaded by

AL Rajhi Zakaria
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
216 views

2.3 Sampling Techniques

This document discusses different sampling techniques used in statistics. It defines key terms like population, sampling frame, and sample. It then describes several sampling methods: simple random sampling, systematic sampling, stratified sampling, cluster sampling, voluntary response sampling, and convenience sampling. For each method it provides an example and explains how it works and when it would be appropriate to use. It also discusses factors to consider for random sampling like nonresponse rates and sample size.

Uploaded by

AL Rajhi Zakaria
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 20

2.

3 Sampling Techniques
MDM4U Shaun Le Conte

Firstly, some terms to know!

Population all members of a group under study Sampling Frame all members of a population who actually have a chance of being included in the sample

e.g. you may want to take a random sample of Sarnia, Ontario, Canada residents, but if you are calling people from the phone book, only those who have a listed phone number and are home during the time that you plan to call are part of the sampling frame

Populations

Population: all objects under study. Usually, we do not have complete knowledge of a population and therefore we seek to gather samples in order to make inferences about the population

E.g. true unemployment % among the population is unknown, what you see in the media is an estimate

Sampling a Population

A large population may be impractical and costly to study so a sample is used instead A sample is the part of the population that is actually selected to be studied

image from:

http://mips.stanford.edu/public/classes/stats_data_analysis/lesson_1/pop_sam.mov

What is random?
Can you tell the difference between random and non-random? Look at the next slide, filled with 1 and 0s and decide whether the top or the bottom row is random

Which section of 0s and 1s is the randomly generated section?

Sampling Techniques

Simple random sample Systematic sample Stratified sample Cluster sample Voluntary response sample Convenience sample

A. Simple Random Sample


Key point: Every member of the population has an EQUAL chance of being selected.

Examples: picking names out of a box, randomly generating numbers from a calculator, using a random number table

B. Systematic Sample

Select members of the population at regular intervals Interval, k = population size, N / sample size, n Example: 25 students, sample size 5, select every 5th from the class list Before proceeding, either the list must be randomized or a seed value must be generated, which is the starting point in between 1 and the sampling interval (inclusive) Without the above step, the first 4 people in the list would have no chance of being included in the survey Refer to Example 2 on page 115.

C. Stratified Sample

Used when information is known about the proportions of different, mutually exclusive groups, or strata, within a population
e.g. age groups

Stratified sampling can help in reducing sampling error A sample is randomly collected from each strata. The proportion of each strata in the sample should reflect the proportion of each strata in the population Ensures that subgroups are represented in your sample Examples: - comparing male students and female students - comparing younger students and older students Refer to example 3 on page 115 116.
The difference between the sample statistic and the population parameter

page 117

Answer to #4

D. Cluster Sample

The entire population is divided into groups (known as a clusters) and a random sample of these clusters is selected. Cluster sampling works best when clusters do not differ significantly All members in the selected clusters are included in the sample. Examples: - studying job satisfaction at McDonalds by surveying all employees at a few locations (cluster sample) rather surveying a few employees at all locations (simple random sample) - randomly selecting city blocks and surveying all households on the block

Comparing stratified sampling and cluster sampling

Cluster sampling: a random sample is drawn from the population of clusters. Only the selected clusters are analyzed.

Main advantage: reduce costs by sampling a smaller geographical area.

Stratified sampling: a random sample is drawn from each of the strata. Analysis is performed on elements within strata.

Main advantage: take advantage of prior knowledge of a population to reduce sampling error

F. Voluntary Response Sample

The sample is made up of members in the population who choose to participate in the study. Examples: SMS surveys, radio shows in which listeners call in, questions posted to message boards Are participants representative???

G. Convenience Sample

Sampling whoever is accessible. Example: asking your next door neighbors or your classmates. Major problem: the sample may not represent the population (not representative). Estimates of the population parameter are less valid than with other types of sampling

Voluntary vs Convenience Sampling


Both are considered non-random sampling There is an important distinction between the two types In voluntary sampling, people will willingly seek to take part in the sample

In convenience sampling, the surveyor seeks out others and asks them to participate those asked depend on who is around at that time and place, and most approachable (selection bias)
Voluntary sampling is often considered the least representative method of sampling since those who volunteer tend to hold strong opinions for or against an issue

Some points to consider when doing any sort of random sampling


When you have determined who it is that has been selected to be part of your sample, you must of course find and contact that person! If the person declines to participate, they can't be forced but the surveyor should ask if they could talk at a later time (and pinpoint a time) The more a surveyor fails to collect samples from those on the to be surveyed list, the less random it will become

Sample size?
Generally: as many as possible, provided that the quality of sampling remains constant Ultimately it will depend on the constraints: budget, labor, time. The larger your sample, the more likely your statistic will be a good estimate of the population parameter

You might also like