stats midterm 2

Categorical Variables

Variables that describe groups or categories, summarized using counts, proportions, and visual tools like bar graphs.

Quantitative Variables

Variables that take numerical values where arithmetic operations make sense, summarized using means, medians, standard deviations, and graphs like histograms.

Statistical Goals in Experimental Design

To compare background variables between treatment groups, aiming to fail to reject the null hypothesis for group similarity.

Observational Units

The basic unit of analysis, such as one shift in a study.

Explanatory Variables

The variable that is manipulated in a study, such as 'Gilbert on shift?' (yes/no) in the shift example.

Response Variables

The outcome variable that is measured, e.g., 'At least one death?' (yes/no).

Two-Way Tables

A method used to summarize data when both variables are categorical, displaying counts across categories.

Difference in Proportions

A statistical measure comparing proportions between groups.

Randomization Test

A test that assumes data assignments are random under the null hypothesis and evaluates extreme values in simulated distributions.

Null Hypothesis (H₀)

The default hypothesis that indicates no effect or difference between groups.

Confounding Variables

Variables that may influence both explanatory and response variables, complicating causal interpretations.

Validity Conditions for Two-Sample Z Tests

Each group must have at least 10 successes and 10 failures for comparison of proportions.

Random Assignment

A process used in experiments to randomly assign participants to different treatment groups, enabling causal inferences.

Software Output in t-tests

Key outputs include sample means, standard deviations, standard error of difference, test statistic, p-value, and confidence interval.

P-value Interpretation

The probability of observing the data, or something more extreme, assuming the null hypothesis is true.

Matched Pairs Design

A study design where subjects are paired based on shared characteristics, controlling for variability.

One-Sample Paired t-test

A statistical method used to compare means from the same group at different times or conditions.

Confidence Interval (CI)

A range of plausible values for the population parameter, indicating where the true mean difference is likely to lie.