Unit 12: Sampling. When enough is enough.
45 Probability Sampling
Ok, I’m just going to say it: The way we (researcher types) use probability and random in this context is downright confusing. Colloquially, we (regular humans) tend to use random to mean that something is really out there, or without thought, or, frankly, we’re kind of confused as to WHY…how….what…HUH?!? Like perhaps this TikTok complication that came up when I searched for random funny videos[1]. On the flipside, random, RESEARCH RANDOM is like the opposite of all this! It is absolutely mindful, systematic, purposeful, etc. It has a reason, an important reason, and a process, etc. This is definitely one of those things that students screw up on exams – I think mostly when they don’t read or come to class because I harp on this whole random thing pretty hard – so when you’re reading your exam questions related to randomness think RESEARCH RANDOM, not…well…the other (more popular) kind.
And now…you know what’s coming, right?
Learning Objectives
Understand what probability sampling is, what methodology (that we discuss) generally uses it, and a handful of different mainstream strategies.
Probability Sampling
Probability sampling is a type of quantitative, or measurable, method of finding a sample that represents a population. Probability or random sampling uses methods to systematically select candidates. Ideally, each member of the population has an equal chance of being chosen for a study, making it as representative of the population as possible. A researcher interested in generalizable results may lean towards these types of sampling methods.
Methods of Sampling:
Random Sampling: Having a system to ensure an equally possible outcome/most representative of the population
Types of Random or Probability Sampling (a selection. Yes, there are more…):
Systematic
Stratified
Proportional Stratified
-
-
-
-
-
Simple random sampling is where we select a group of subjects (a sample) for study from a larger group (a population). Each individual is chosen entirely by chance and each member of the population has an equal chance of being included in the sample.
- Example: Assigning numbers 1 to 10 to participants and then drawing a number from 1 to 10 out of a hat to select candidates for your study.
-
-
-
-
-
-
-
-
- Every nth person is chosen; This could also be done with every third name in the phonebook — every 3rd person could be selected (the 3rd, 6th, 9th, 12th, 15th, and so on).
- Another Example: Giving a survey to every fourth customer that comes into the movie theater.
- Every nth person is chosen; This could also be done with every third name in the phonebook — every 3rd person could be selected (the 3rd, 6th, 9th, 12th, 15th, and so on).
-
-
-
-
-
-
-
- This is when you break people into groups, called strata. From there, the researcher picks randomly from within the strata. This is one of the more accurate types of probability sampling.
- Some examples of strata might me sex, generational group, or economic background.
- This is when you break people into groups, called strata. From there, the researcher picks randomly from within the strata. This is one of the more accurate types of probability sampling.
-
-
-
Proportional Stratified Sampling
-
-
-
-
- It is the same as stratified except the groups (strata) are proportional to the population the researcher is trying to emulate.
- Example: In the U.S., in the 85 and older age group, women outnumber men by a ratio of 2-to-1 (4.0 million to 2.0 million). So, your sample should represent that with a 2-to-1 ratio of women to men.
- It is the same as stratified except the groups (strata) are proportional to the population the researcher is trying to emulate.
-
-
-
-
-
-
-
- The people within the sampling frame are broken into random groups (usually geographic) and candidates are chosen randomly. This is a less accurate method when it comes to accurately representing a population, but it works well with a large or fairly unknown population.
-
Example: A researcher wants to survey the academic performance of high school students in Spain. She can divide the entire population (population of Spain) into different clusters (cities). Then the researcher selects a number of clusters depending on her research through simple or systematic random sampling.
-
- The people within the sampling frame are broken into random groups (usually geographic) and candidates are chosen randomly. This is a less accurate method when it comes to accurately representing a population, but it works well with a large or fairly unknown population.
-
-
-
ok, last one, REALLY – but it’s by the Pew Research Center so how can I resist?
Got ideas for questions to include on the exam?
Click this link to add them! [this course element is paused because ya’ll aren’t submitting many questions…]
…Unit 1…Unit 2….Unit 3…Unit 4…Unit 5…Unit 6…Unit 7…Unit 8…Unit 9…Unit 10…Unit 11…Unit 12…Unit 13…Unit 14…Unit 15…Unit 16…
Unit 12: Sampling. When enough is enough.
- Listen, I spent way too much time looking for an acceptable video. I don't like the ones where people get hurt, which is a mainstay of "funny" (I guess it's the mom in me), and heaven knows that I probably have no idea what ya'll young whippersnappers find funny these days. I ended up choosing this video because TikTok (took a chance) and because it's titled "hilarious clean tiktoks that made me giggle in class." IN CLASS. I'm being ironic, obviously. ↵
a type of quantitative, or measurable, method of finding a sample that represents a population
having a system to ensure an equally possible outcome/most representative of the population
Simple random sampling is where we select a group of subjects (a sample) for study from a larger group (a population)
Every nth person is chosen
when you break people into groups, called strata
It is the same as stratified except the groups (strata) are proportional to the population the researcher is trying to emulate
when people within the sampling frame are broken into random groups (usually geographic) and candidates are chosen randomly