Bias Mini quiz Cards 10.23.26

Bias

Systematic tendency for a sampling method to overestimate or underestimate a true population value.

Population

The entire group a researcher wants to study.

Sample

The individuals actually selected and observed in a study.

Parameter

A numerical value describing an entire population.

Statistic

A numerical value calculated from a sample.

Representative Sample

A sample that reflects important characteristics of the target population.

Generalization

Using sample results to draw conclusions about a larger population.

Random Sampling

Selecting individuals by chance to reduce selection bias and support generalization.

Simple Random Sample (SRS)

A sample in which every possible sample of a given size has an equal chance of selection.

Undercoverage Bias

Occurs when some population members have little or no chance of being selected.

Voluntary Response Bias

Occurs when people choose to participate, often favoring those with strong opinions.

Nonresponse Bias

Occurs when selected individuals do not respond and responders differ meaningfully from nonresponders.

Response Bias

Occurs when responses are systematically inaccurate because of wording, interviewer influence, memory, or social pressure.

Wording Bias

Occurs when a question's phrasing nudges respondents toward a particular answer.

Convenience Sampling

Selecting individuals who are easiest to reach, potentially producing a biased sample.

Overestimation

When a method tends to produce a sample statistic higher than the true population parameter.

Underestimation

When a method tends to produce a sample statistic lower than the true population parameter.

Direction of Bias

The tendency of a method to push an estimate above or below the true population value.

Sampling Variability

Natural differences between sample statistics caused by chance selection.

Random Sampling vs. Bias

Random sampling reduces selection bias but does not guarantee a perfectly representative sample.

Large Sample Size

A larger sample can reduce random variability but does not automatically eliminate bias.

Stratified Random Sample

Divide the population into groups and randomly sample individuals within each group.

Cluster Sample

Randomly select entire groups and survey all individuals within the chosen groups.

Systematic Sample

Select individuals using a fixed interval, such as every 10th person.

Convenience Sample Example

Surveying students who happen to be in the cafeteria.

Voluntary Response Example

An online poll where anyone chooses whether to submit a vote.

Undercoverage Example

Surveying only club members to estimate the opinions of all students.

Nonresponse Example

Randomly selecting 100 people but receiving only 40 completed surveys.

Response Bias Example

People underreporting socially undesirable behavior in a survey.

Wording Bias Example

Asking, "Should the school waste money on unnecessary programs?"

Overestimation Example

Surveying enthusiastic fans to estimate the popularity of a product among everyone.

Underestimation Example

Respondents systematically reporting fewer hours spent on social media than they actually spend.

Generalization from a Random Sample

Results can generally be generalized to the population from which the sample was randomly selected, subject to study limitations.

Generalization from a Convenience Sample

Results cannot confidently be generalized to the entire target population because selection may be biased.

Random Sampling vs. Random Assignment

Random sampling supports generalization; random assignment supports cause-and-effect conclusions.

Can Random Sampling Still Be Biased?

Yes. Nonresponse, undercoverage, response bias, and chance variation can still affect results.

Does a Larger Sample Eliminate Bias?

No. Increasing sample size reduces random variability but does not necessarily correct systematic bias.

Identifying Overestimation or Underestimation

Determine who is overrepresented or underrepresented and how that affects the measured variable.

When Is Bias Direction Unknown?

When the scenario does not provide enough information to determine whether the estimate tends high or low.

Effect of Leading Questions

They can influence responses and cause response bias.

Effect of Excluding Low Values

If excluded individuals have lower values than the rest of the population, the sample average tends to be too high.

Effect of Excluding High Values

If excluded individuals have higher values than the rest of the population, the sample average tends to be too low.

Effect of Strong Opinions in a Poll

People with strong opinions may participate disproportionately, making results unrepresentative.