AP Statistics Unit 3 Notes

Sampling Methods

  • Simple Random Sample (SRS):

    • Randomly selected subset of the population.

    • All members have an equal chance of selection.

    • Process:

      1. Define the population and label individuals (assign numbers or use names).

      2. Randomize (random number generator, names in a hat, etc.).

      3. Select members for the sample.

Stratified Random Sampling

  • Split the population into groups (strata) based on shared traits (homogeneous groups).

    • Example: Dividing PrepWorks viewers into subscribers and non-subscribers.

  • Randomly select samples from each group.

    • Ensures everyone in the group has an equal chance of selection.

    • Example: Divide a school by grade level and randomly pick students from each grade.

Cluster Sampling

  • Divide the population into groups (clusters) and randomly select ENTIRE clusters.

    • Important: Select entire clusters to sample.

    • Example: Splitting a city into neighborhoods and surveying everyone in a few randomly chosen neighborhoods.

  • Key Difference:

    • Stratified: Homogeneous groups (e.g., all red or all blue).

    • Cluster: Heterogeneous groups (mixed representation).

Diagrammatic Depiction

  • Stratified: Split into red and blue, then SRS within each.

  • Cluster: Clusters contain both red and blue; entire clusters are selected.

Systematic Random Sampling

  • Select individuals at regular intervals, starting at a random point.

    • Example: Surveying every fifth person entering a school.

Bad Sampling Methods

  • Convenience Sample:

    • Choose people who are easy to reach.

    • Example: Surveying people at a nearby mall.

  • Voluntary Response Sampling:

    • Allow people to choose to participate.

    • People who feel strongly are more likely to respond, introducing bias.

    • Example: Online polls where respondents self-select.

Shortcomings in Sampling

  • Undercoverage:

    • Some groups are left out or underrepresented.

    • Example: Surveying only people with internet access.

  • Non-response:

    • Selected individuals don't or can't respond.

    • Example: People refusing to participate in a phone survey.

  • Response Bias:

    • People give false or misleading answers.

    • Example: A friend reviewing your YouTube channel.

  • Wording in the Question:

    • Poorly phrased or biased questions influence answers.

    • Example: "Do you want to subscribe to Prep's Education to receive $1 million?"

Observational Studies and Experiments

  • Observational Study:

    • Observe and collect data without influencing subjects.

    • Example: Watching people use seatbelts in their cars.

  • Experiment:

    • Manipulate variables or apply treatments to subjects.

    • Observe and measure the effects.

Principles of Experimental Design

  1. Comparison: Compare two or more groups to see the difference in the treatment.

  2. Random assignment: Randomly assign subjects to groups to reduce bias. You can also be randomly assigning treatments.

  3. Control: Keep all variables constant.

  4. Replication: Ensure a sufficient number of subjects.

Key Vocabulary

  • Factor:

    • Explanatory or independent variable (when there are multiple).

  • Level:

    • Specific value or category of the factor.

    • Example: Sunlight (factor), low/high sunlight (level).

  • Confounding:

    • Another variable affects the results, making it hard to determine the true cause.

  • Placebo:

    • A fake treatment where participants may react favorably.

  • Single Blind:

    • Subjects don't know their group, but researchers do.

  • Double Blind:

    • Neither subjects nor researchers know group assignments.

Randomized Block Design

  • Subjects are divided into blocks based on specific characteristics.

  • Each block is randomly assigned a treatment.

  • A block is a group of experimental units with the same characteristic.

    • Example: PrepWorks viewers split into subscribers and non-subscribers.

  • Treatments are then randomly assigned (e.g., using a random number generator).

Matched Pairs Design

  • Subjects are paired based on specific characteristics.

  • Each pair is randomly assigned a treatment.

  • One subject in each pair can be a control.

  • Example: Pairing a male and female to eliminate gender bias; assigning treatments or control status randomly within each pair.