Quick Recap: Populations and Samples

Population

• The entire group of units that we want to draw conclusions about.
• Whatever unit or thing we are interested in, the population would be composed of every one of those units.
• A population does not necessarily refer to people: it can be composed of countries, states, organizations, groups of people, etc.

Sample

• A subset of units taken from the population that we collect data on.
• A sample is representative of the population if it accurately reflects the characteristics of the population.
  • In other words, a sample is representative of the population if it looks very similar to the population.

Measuring populations and samples

• Parameter
  • A measure of a characteristic of a population (ex. mean, median, mode, standard deviation, etc.)
  • The “true” value we want to know
  • A fixed measure that does not change

• Statistic
  • A measure of a characteristic of a sample (ex. mean, median, mode, standard deviation, etc.)
  • Unique to each sample, so it changes depending on the sample

Inference

• Drawing conclusions about populations using samples.
• Can only be done with samples that are representative of the population.
• Can only be done with samples that are large enough.

Types of samples

• • Nonprobability samples
    • Not everyone in the population has a probability of being selected into the sample.
    • Are not representative of the population.
    • Cannot use to make inferences about population.
  • Probability samples
    • Everyone in the population has some probability of being selected into the sample.
    • Can create samples that are representative of the population.
    • Includes various types of samples, most well-known is the simple random sample (here referred to as “random sample”).

Random sample

• • Each unit in the population has the same probability of being selected to be in the sample
    • Each unit has an equal chance of being in the sample
  • Gives us samples that are representative of the population
    • Sample will look very similar to our population across various characteristics
  • We can use random samples for inference!
  • Funny video linkLinks to an external site.

Dig Deeper

Populations, samples, and inference

• Galderisi, Peter. Understanding Political Science Statistics: Observations and Expectations in Political Analysis. Routledge, 2015. Chapter 6.
• More nuanced discussion about populations and samples: https://conjointly.com/kb/sampling-in-research/ (Links to an external site.)
• Explanation of sampling distributions: https://conjointly.com/kb/sampling-statistical-terms/ (Links to an external site.)
• Explanation of sampling distributions: https://stattrek.com/sampling/sampling-distribution.aspx?tutorial=AP (Links to an external site.)

Types of samples

• Introduction to different types of probability samples: https://conjointly.com/kb/probability-sampling/ (Links to an external site.)
• Introduction to different types of nonprobability samples: https://conjointly.com/kb/nonprobability-sampling/