Lecture 9

Question 1

Define Sampling versus Census. [2]

Answer 1

• Census: A count of all elements in a population.
• Sampling: Selecting a sufficient number of elements to draw conclusions about the entire population.

Question 2

Advantages of Sampling [3]

Answer 2

• Cost & Time: It is cheaper and faster to collect and process data from a sample than a whole population,.
• Destructive Testing: Essential when the object is destroyed during inspection (e.g., testing bullets or eggs).
• Detail: Allows for more detailed information to be obtained from each unit.

Question 3

Elaborate on the Sampling Process (Population vs. Sampling Frame). [2]

Answer 3

Population: The totality of people, events, or things the researcher wishes to investigate.

Sampling Frame: The source list from which the sample is drawn. It is crucial that this frame is representative of the population to avoid error.

Question 4

Stratified Sampling: [2]

Intrastratum (Within) and Interstrata (Between):

Answer 4

• Intrastratum (Within): Homogeneity (elements within a stratum are similar).
• Interstrata (Between): Heterogeneity (strata are different from each other).

Question 5

Cluster Sampling [2]

Intercluster and Intracluster

Answer 5

• Intracluster (Within): Heterogeneity (elements within a cluster are diverse).
• Intercluster (Between): Homogeneity (clusters look similar to one another).

Question 6

Define Systematic Sampling [2]

Answer 6

Procedure: Involves selecting every nth element, starting with a random choice between 1 and n.

Bias: A specific risk is systematic bias if the list of elements is not arranged randomly (e.g., if every 10th person on a list is a supervisor, and you sample every 10th person, you get only supervisors).

Question 7

Probability vs. Non-probability Sampling.

Answer 7

In Probability Sampling, elements have a known and non-zero chance of being chosen.

In Non-probability sampling, the chance is unknown

Question 8

4 Major Probability Types:
1. Simple Random Sampling
2. Systematic Sampling
3. Stratified Random Sampling
4. Cluster Sampling.

Answer 8

• Simple Random Sampling: Every person is picked entirely by chance, like pulling names out of a hat.
• Systematic Sampling: You pick every $n^{th}$ person from a list (e.g., every 10th person) after a random start.
• Stratified Random Sampling: You divide the population into groups (like age or gender) and pick randomly from each to ensure all are represented.
• Cluster Sampling: You divide the population into sections (like zip codes), pick a few sections at random, and survey everyone inside them.