##### Psych 311 Unit 4 study guide

Population versus sample (including reasons to use a sample rather than the whole population)

Population: Overall group of interest to the researcher

Sample: Set of individuals selected from a population, goal is to be representative of a population

Reasons to use a sample rather than the whole population: population is too large, cost, feasibility

Target population: Group defined by research interests

Accessible population: Subset of target population that is accessible to researcher

Generalization, representativeness, representative sample, biased sample, sampling bias and selection bias

Generalization: The process of using the results of a research sample to infer information about a population

Representativeness: How accurately a sample reflects the characteristics of the population

Representative sample: Sample with same characteristics as population, allows the researcher to make generalizations

Biased sample: Sample with characteristics *different* from
population, does not allow the researcher to make generalizations,
caused by Sampling Bias (or Selection Bias)

Selection bias: Participants are selected in a manner that increases the probability of a biased sample

Sample size: law of large numbers, general recruitment guidelines for numbers of participants, practical considerations that affect sample size

Law of Large Numbers: the larger the sample size, the more likely the sample accurately represents the population

Feasibility: What is actually possible?

Level of accuracy: Is there a specific target needed? Example: Poll during an election period, voter preferences (±5%)

Probability sampling methods and **three conditions**

...

Simple random sampling

- Each individual has an equal chance of being selected, selection of one individual is independent of the selection of others
- Process
- Clearly define population
- List all members of population
- Use a random process to select

- Two methods: With replacement and Without replacement
- Strengths: Process is fair and unbiased
- Weaknesses: Need to know all members of population, No guarantee that the sample is representative

Stratified random sampling

- Uses identifiable subgroups of a population, goal is to be representative of subgroups
- Process:
- Identify subgroups of the population
- Select equal-sized random sample of the subgroups, using the same steps as random sampling
- Combine subgroups into one overall sample

- Strengths: Guarantees that each subgroup will have representation
- Weaknesses: Need to know all members of population, selection not really random or independent, overall sample is usually NOT representative of the population

Proportionate Stratified Random Sampling

- Sample obtained by subdividing a population into strata, then randomly selecting from the strata so that proportions in the sample correspond to those in the population
- Process:
- Identify subgroups of the population
- Determine what proportion of the population corresponds to each subgroup
- Select proportioned random samples of the subgroups
- Combine subgroups into one overall sample

- Strengths: Composition of the sample (representative strata) will be proportionate to the population
- Weaknesses: Need to know all members of population, selection not really random or independent, some strata may have limited representation within the sample

Nonprobability sampling methods

...

Convenience sampling

- Participants are selected on the basis of availability and willingness to respond, Population is not known, no random process
- Safeguarding against bias: Can be attempted through selecting a broad cross-section of people, explicitly describe the sample methodology for use in limitations and understanding generalization
- Process:
- Identify individuals who are easy to reach and willing to participate
- Recruit participants

- Strengths:
- Don’t need to know all members of the population
- An easy (and sometimes inexpensive) method for obtaining a sample

- Weaknesses:
- Selection not random or independent
- Result is likely a biased sample

Quota sampling

Ensures that subgroups are represented in a convenience sample, variation can mimic proportionate stratified sampling

Safeguarding against bias: Can be attempted through selecting a broad cross-section, explicitly describe sample methodology for use in limitations and understanding generalization

Process:

- Identify subgroups to be included
- Establish quotas for subgroups
- Identify individuals who are easy to obtain and willing to participate
- Recruit participants

Strengths:

- Don’t need to know all members of the population
- Allows researcher to control the composition of a convenience sample

Weaknesses:

- Selection not random or independent
- Result is likely a biased sample