APDV Unit 3.3

Median is usually in the? Why is it there? In what kind of data is it there?

Due to right-skewed nature of most income data, the median is usally the preferred center

Census

Collects data from ALL individuals in a population

Random sample

Much easier than measuring everyone

Simple random sample (SRS)

Sample in which every group of a given size has an equal chance of being chosen

Cluster random sample

• Population is divided into clusters of individuals that are near one another

• SRS of clusters is taken

• All individuals within each cluster are sampled

Stratified random sample

• Population is divided into strata, based on a similar characteristic

• SRS within each stratum is taken

• Selected individuals are combined into larger sample

Difference between cluster & stratified samples

• Cluster: Start by grouping (ideally, heterogenous)

• Stratified: Start by grouping (ideally, homogenous)

Systematic ransom sample

Randomly chose a start point, then sample at a fixed periodic interval

Primary advantage

Easy to collect sample, especially in situations in which individuals in the population are lined up in some way

What do random samples tend to provide?

Random samples, when well-executed, tend to provide representative samples

A simple random sample (SRS) gives?

A simple random sample (SRS) gives every group of ‘n’ individuals an equal chance of selection

Cluster sampling creates?

Cluster sampling creates groups, then randomly sample within each group

Bias

Measure of accuracy (are you centered at the true value you want to be estimated)

Variation

Precision (How much distance between potential estimates)

Non random sample can lead to?

Non random sample can lead to systemic overestimation 

Simple random sample characteristics

• Unbiased (accurate)

• Moderate variability (moderate precision)

Simple random sample (SRS) advantages

• Unbiased

• Easy to explain

• In certain cases, can be easy to perform

Simple random sample (SRS) disadvantages

• In certain scenarios, can be difficult to implement

• May not be as precise as other methods

Cluster random sample characteristics

• Unbiased (accurate)

• High variability (Not precise)

Cluster random sample advantages

• Unbiased

• Can be easy to perform

Cluster random sample disadvantages

• If clusters are homogenous but very different from one another, can have very high variability

  • May get an estimate that is very far from the “truth”

Stratified random sample characteristics

• Unbiased (accurate)

• Low variability (precise)

Stratified random sample advantages

• Unbiased

• When strata are homogenous, tends to have low variability

Stratified random sample disadvantages

• Can be very difficult to implement

Random sampling tends to?

Random sampling tends to provide unbiased estimates

Cluster random sampling is effective when?

Cluster randoms sampling is effective when clusters are heterogenous and similar to one another

Stratified random sampling is most effective when?

Stratified random sampling is most effective when strata is homogenous