SOCI1001: Introduction to Social Research - Sampling Flashcards

Comprehensive flashcards covering the definitions, terminologies, stages, and types of probability and non-probability sampling designs based on introductory social research lecture notes.

How is 'sampling' defined in the context of social research?

The process of deciding what or whom to observe when one cannot observe everything or everyone they want to.

What is a 'sample'?

A subset of a population selected for a study.

What are three primary reasons to use sampling instead of studying an entire population?

Reduced cost, saves time, and greater scope.

What is the difference between a target population and a census?

A target population is the group about which social scientists attempt to make generalizations, whereas a census is a study that includes data on every member of that population.

What is a 'population parameter'?

A number that characterizes some quantitative aspect of a population.

What is the function of a 'sampling frame'?

It is a list of population members from which a probability sample is drawn.

What are the six stages in the selection of a sample?

1. Define the target population; 2. Select a sampling frame; 3. Decide on probability or non-probability sampling method; 4. Plan procedure for selecting sampling units; 5. Determine sample size; 6. Select actual sampling unit.

How does population homogeneity or heterogeneity affect sample size?

If a population is homogeneous, a small sample could represent it; with increasing variability (heterogeneity), a larger sample is needed.

What is the fundamental difference between probability and non-probability sampling?

In probability sampling, individuals are selected by chance (random selection) and every member has a calculable probability of being selected; in non-probability sampling, individuals are selected based on choice.

What are the four main types of probability sampling designs?

Simple random, Systematic, Cluster, and Stratified.

How is a simple random sample conducted?

The total population is identified and listed, members are assigned a number, and a table of random numbers is used to select members so each has an equal probability of selection.

Define 'systematic random sampling'.

A probability sampling strategy in which sample members are selected using a fixed interval, such as taking every $k ext{-th}$ person on a list.

How is the sampling interval ($K$) calculated in systematic sampling?

$$K = ext{Total Population} / ext{Sample Size}$$

What is 'stratified random sampling'?

A strategy where the population is divided into groups or strata (e.g., by sex or age) and members are selected in strategic proportions from each group.

When is 'cluster sampling' typically used?

When researchers cannot get a complete list of individual population members but can get a complete list of groups or clusters in the population.

What are the four types of non-probability sampling designs?

Accidental/convenience, Purposive/judgmental, Quota, and Snowball/Mudball.

What is 'accidental or convenience sampling'?

A method that selects observations based on what is cheapest or easiest, such as interviewing whoever comes along at a library entrance.

Define 'purposive sampling'.

A strategy where cases are deliberately selected based on features that distinguish them from other cases, relying on the researcher's judgment.

What are 'typical and extreme cases' in purposive sampling?

An approach where researchers select average cases or deviant/extreme cases to find out why they deviate from the norm.

How does 'quota sampling' resemble stratified random sampling?

Both divide the population into strata (age, race, gender); however, quota sampling uses convenience or judgmental methods to fill the allocated proportions rather than random selection.

What is the 'snowball or mudball' sampling technique?

A strategy where the researcher starts with one respondent who meets inclusion requirements and then asks them to recommend others.

According to Waldorf (1981), what are the four processes involved in snowball sampling?

Finding, Verifying, Engaging, and Controlling (pacing and monitoring).

Which sampling technique is recommended if a sample frame is unavailable but a representative sample is still desired?

Quota sampling (it is considered the next best option for representation among non-probability techniques).

Which general study design is best paired with probability sampling?

Quantitative studies, because they aim to produce representative samples to make inferences to the population.