AP Statistics Unit 3

Population

group being observed

Sample

Part of said group to gain information about the population

Statistic

A value that describes a sample

Parameter

A value that describes a population (-estimated from the statistic)

Observational Study

• no treatments

• studies correlation, but cannot imply causation

• lurking variables make it difficult to establish causal links

• measured in a natural setting

Sample Surveys

• research use questionnaires or interviews to collect information from a sample to learn about the entire population

• Census - A type of survey meant to collect data for ALL individuals in the population.

  • Difficulties — Collecting responses cost resources, and time consuming.

    • Best used when it sample is randomly selected.

    • Bias can be introduced if sample is not randomly selected.

3 points to a good survey

1. Speak the audience’s language

2. Keep it simple

3. Stay unbiased

Advantages of Surveys

• Various forms of collecting data:

• Efficient for collecting data from a large population in a timely manner

• Can be designed to focus only on needed response questions

• Applicable to wide range of topics.

Disadvantages of Surveys

• Dependent on respondents’ honesty and motivation when answering

• Can be flawed by bias

Experiments

• Purpose is to study whether a treatment causes a change in the response of an experimental unit.

• Human experimental units are called subjects.

• All experimental units need to believe they are experiencing the same condition.

• The best experiments use random assignment into treatments, are comparative, double—blind, placebo controlled.

Confounding Variable

Lurking Variable

Placebo

“fake” treatment that appears to be the actual treatment, but has no effect.

• Placebo Effect — Favorable response to the placebo.

Single-Blind Study

Double-Blind Study

SB — Researchers are aware which group has placebo. Subj don’t

DB — Both researchers and subjects are not aware which group has the placebo.