Biostats 2

Regression line

A straight line that summarizes the linear relationship between two numeric variables.

Coefficient of determination

Describes the proportion of variation in y that can be explained by ŷ.

Confounded variables

More than 1 as a group, their effect on the response cannot be distinguished.

Lurking Variables

Individual Listings (ex. diet, lifestyle) that may affect the response variable but are not included in the analysis.

Sampling design

Describes how a sample is obtained from the population.

Statistical

We use these samples to draw conclusions about the population.

Simple Random Sampling (SRS)

A method in which every sample of size has an equal chance to be drawn.

Stratified Sampling

When the population is separated into groups that have a common characteristic called strata.

Strata

Subgroups within a population that are separated based on a common characteristic.

Cluster Sampling

Separates the population into groups in some natural way.

Clusters

Subgroups within a population that are separated.

Multistage Sampling

Takes a SRS from another SRS.

Systematic Sampling

Has individuals lined up in some way, picks a random starting point, and also adds every kth individual after.

Poor Sampling Designs

Convenience Sampling and Voluntary Response Sampling.

Convenience Sampling

Takes individuals who are easily accessible (no clear method).

Voluntary Response Sampling

Allows individuals to choose whether or not they participate.

Normal Distribution

Have the same general shape, defined by two values: mean and standard deviation.

M (Mean)

Determines the location; smaller in the blue curve.

O (Standard Deviation)

Determines the spread/flatness; smaller in the blue curve because more of the data is close to the mean (less spread).

Proportion or probability

The area under the normal distribution curve is 1.

Empirical Rule

68%, 95%, 99% rule for normal distribution.

68%

Data are within 1 standard deviation of the mean.

95%

Data are within 2 standard deviations of the mean.

99%

Data are within 3 standard deviations of the mean.

Impossible

A probability of 0.

Guaranteed

A probability of 1.