Unit 1 - Analysis of Data : Sampling Methods

Question 1

Random sampling

Answer 1

Every item in the population has an equal chance of being chosen. You assign a number to each item and then generate random numbers in order to choose a sample (ignoring any repeated numbers).

Question 2

Stratified sampling

Answer 2

The population is divided into categories (strata) by age, gender, social class, etc. Then a random sample is generated from each group. The size of each sample is in proportion to the size of each category in the sample.

Question 3

Cluster sampling

Answer 3

The population is divided into groups (clusters). A random sample of clusters is chosen and every item in the chosen cluster is surveyed. The smaller the size of each cluster, the less likely the sample is biased.

Question 4

Quota sampling

Answer 4

Specific instructions are given about the how many (quota) of each group within the population must be included in the sample. The actual items chosen are left to the discretion of the surveyor.

Question 5

Sampling frame

Answer 5

Used to identify the population. A list of all items in the population and ensures that all have a chance of being sampled.

Question 6

Random sampling advantages

Answer 6

• genuinely random
• free from bias

Question 7

Random sampling disadvantages

Answer 7

• time consuming
• costly
• needs a sampling frame (need to know everyone)

Question 8

Stratified sampling advantages

Answer 8

• more accurate than random
• can show different tendencies for each group

Question 9

Stratified sampling disadvantages

Answer 9

• time consuming
• costly
• needs a sampling frame
• detailed planning needed

Question 10

Cluster sampling advantages

Answer 10

• cheaper to run
• doesnt need sampling frame

Question 11

Cluster sampling disadvantages

Answer 11

• not random
• could be biased (depending on clusters used)

Question 12

Quota sampling advantages

Answer 12

• cheap
• easy

Question 13

Quota sampling disadvantages

Answer 13

• not random
• double be biased (based on how researcher picks sample)

Question 14

Balance between removing bias and increasing sample size

Answer 14

Too small -> cheaper but could easily be biased
Too large -> less likely to be biased but expensive

Question 15

How to work out stratified sampling

Answer 15

Percentage - divide all different strata’s by the given % to get their number in sample
Subjects - divide total number by sample number. Then divide all strata’s by that answer to find their number in the sample