Chapter 1 Stats Review

Exam is September 29th @ 6:30

Cases/observational units

The objects on which one takes measurements or records characteristics.

Variable

The measurement or characteristic being recorded for each

subject.

Population

A collection or set of all subjects of interest in a study.

Sample

A subset of the population for which data will be collected for study.

Statistical Inference

Uses the data in the sample to draw conclusions or make predictions about the entire population.

Treatment Group

Patients that received the treatment.

Control Group

Patients meant for the sake of comparison.

Descriptive Statistical Methods

Used to summarize the data collected on a sample.

Inferential Statistical Methods

Used to draw conclusions or make predictions about characteristics of the population if interest based on the information in a sample from that population.

Numerical Variable

A quantitative variable that can be expressed as a number and represents a measurable or countable quantity.

Categorical Variable

A qualitative variable that assigns observations to different groups or categories based on qualitative properties.

Continuous Numerical Variable

A quantitative variable that can take on an infinite number of values within a given range, including decimals and fractions.

Discrete Numerical Variable

A quantitative variable that can only take on specific, separate, and countable values, often whole numbers, but sometimes including a finite number of decimal places within a given range

Nominal Categorical Variable

A type of data that consists of discrete categories with no inherent order or ranking.

Ordinal Categorical Variable

A type of data that consists of discrete categories with an intentional order or ranking.

Explanatory Variable

A variable that is used to potentially explain a causal effect.

Response Variable

A variable that is potentially effected in a casual way by an explanatory variable.

Observational Study

An experiment that researchers may choose to conduct.

Sampling bias

A systematic defect in the sampling method that cause the sample to not look like the population on average since some members of the population have higher chances of being selected than others.

Non-response Bias

Those that respond to a survey may be different than those that do not respond to a survey.

Response Bias

Tendencies for participants to respond inaccurately or falsely to a survey.

Convenience Sample

Individuals who are easily accessible are more likely to be included in the sample.

Voluntary Response

The sample consists of people who volunteer to respond because they have strong opinions on the issue.

Observational Studies

Data is collected in a way that it does not directly interfere with how the data arises and is generally used to establish an association between the explanatory and response variables. Inferring causality is not possible via observational studies. 

Simple Random Sampling

Each case in the population has an equal chance of being included in the sample, and the selection of cases or independent.

Stratified Sampling

The population is divided into strata that are made up of similar cases. We take a simple random sample from each stratum.

Cluster Sampling

The population is divided into clusters that are usually not made up of similar cases. We take a simple random sample of clusters, and then sample all observations in the cluster.

Multistage Sampling

The population is divided into clusters that are usually not made up of similar cases. We take a simple random sample of clusters, and then also take a simple random sample of observations in that cluster.

Experiments

Treatments are assigned and it is then randomized. It is easier to determine causality from an experiment than from an observational study.

Controlling

When assigning treatments to cases, researchers try to control any other differences in the group/

Randomization

Individuals are randomized into treatment groups to account for variables that cannot be controlled. 

Replication

Treatments are assigned to multiple individuals.

Blocking

If there are variables that are known or suspected to affect the response variable, first group subjects into blocks based on these variables. Then randomly assign treatments with each block.

Placebo

Patients are something looking similar to the treatment but isn’t. Allows for researchers to separate the effect of participating in the study from the effect of the treatment. 

Double-blind Experiments

Neither doctors, patients, nor researchers are aware of what treatment patients are receiving.