Video Summary - ¿Qué es el MUESTREO y TIPOS DE MUESTREO? | Metodología Básica y no tan básica #habiaspensado

Main ideas / concepts explained

Sampling: choosing a limited number of elements from a larger set to conduct a study.
Population: the entire group of elements that share at least one characteristic (e.g., people, organizations, institutions, etc.). It is the target to which results are generalized.
Sample: a subset of the population used to make inferences about the population.
Statistical inference / generalization: interpreting the sample results as applying to the full population (assuming the sample is representative).
Why sampling is used: researchers often lack the time and money to observe/test everyone in the population.

Probabilistic sampling: each population element has the same (known) probability of being chosen.
Non-probabilistic sampling: elements do not have equal probability; selection depends on researcher judgment/criteria.

No universally “best” sampling method: different methods serve different purposes and depend on available resources (time/money).
Sample size principle: larger samples generally reduce inference errors; often 100–200+ is preferred (though not a fixed rule).

Define what you want to learn about:
- Identify the population (the full group of interest).
Select a smaller group:
- Choose a sample from the population.
Collect data from the sample.
Infer/generalize:
- Use sample results to draw conclusions about the whole population, assuming adequate sampling quality.

Goal: allow estimates of how sample statistics differ from population statistics.

Create/obtain a complete list of all elements in the population.
Select participants using random selection (e.g., lottery method, random draw, or software).
Example described:
- If selecting 500 students from a school of 10,000:
  - Write all student names,
  - Randomly draw names one by one,
  - Continue 500 times.

Divide the population into smaller groups called strata.
Within each stratum, apply simple random sampling.
Strata differ from each other by a characteristic, while sharing characteristics within each stratum.
Example given:
- Want spending on books by university students at a public university.
- The university has majors (e.g., philosophy, psychology, engineering).
- Create strata by major, then randomly sample within each.

Divide the population into clusters.
Randomly select clusters.
Use:
- All elements inside the chosen clusters.
Two-stage cluster sampling (as described):
- Stage 1: randomly select clusters
- Stage 2: randomly select participants within each selected cluster
Clusters can be based on characteristics like age, sex, location, salary, etc.

Goal: faster and often easier, but requires more caution because representativeness is less guaranteed.

Choose participants based on the study objective.
Select only those best suited to the research aims.
Example described:
- Market research targeting people 25–40, who shop for household items, use a specific cooking oil, and live in certain neighborhoods of Mexico City.

Use prior knowledge about population composition to set quotas.
Select participants so sample traits match the population proportions.
Example described:
- If the population is 45% male / 55% female, the sample should reflect the same proportions.

Select participants who are easily accessible to the researcher.
Example described:
- Students using accessible classmates.
Pros mentioned:
- Very fast, cheap, commonly used.
Cons mentioned:
- Often unrepresentative, so results must be interpreted cautiously.

Used when the population is unknown or hard to reach.
Process:
- Ask the first participant to recommend others who match study requirements.
- Those recommended participants recommend others, continuing like a snowball.

No specific named speakers are identified in the subtitles.
The video is presented as an explanation by the channel’s narrator/host (unnamed in the provided subtitles).