LS Regression
Definitions
Bias
Experiment Design
More Definitions
100

r2 is also called this.

Coefficient of Determination

100

Using a regression line to predict a result far outside of the interval used to obtain the line.

Extrapolation

100

What kind of sample is this?

To conduct a study on depression among the elderly, a sample of 30 patients in one nursing home was used.

Convenience Sample

100

An observed effect so large that it is very unlike to not happen by chance.

Statistically Significant

100

A group of individuals that does not receive treatment in an experiment.

Control Group

200

What is the interpretation of the standard deviation of residuals?

Average distance of the observed values from the predictions.

200

A sample of size n that is chosen in such a way that every group of n individuals in the population has an equal chance to be selected as the sample.

Simple Random Sample (SRS)

200

What kind of sample is this?

To judge the appeal of a proposed television sitcom, a random sample of 10 people from each of three different age categories was selected and those chose were asked to rate a pilot show.

Stratified Random Sample

200

An experiment design where experimental units are assigned to the treatments completely by chance.

Completely Randomized Design

200

A scatterplot of the residuals against the explanatory variable.

Residual Plot

300

A scientist is studying the relationship between hours of sunlight and the height of flowers.

If the coefficient of determination is 0.53, what does that mean in the context of the study?


53% of the variation in the height of flowers can be explained by its linear relationship with hours of sunlight.

300

A group of individuals that are located near each other.

Cluster

300

What kind of sample is this?

Subscribers to the magazine Sound Alive were assigned numbers. Then a sample of 30 subscribers was selected by using a random number table. The subscribers in the sample were invited to rate new compact disc players for a “What a Subscribers Think” column.

Simple Random Sample

300

An experiment in which neither the subjects nor those who interact with them and measure the response variable know which treatment the subject received.

Double-blind Experiment

300

When some members of a population cannot be chosen in a sample.

Undercoverage

400

If the coefficient of determination of a set of data is 0.67, what is the correlation?

0.819

400

The smallest collection of individuals to which a treatment is applied.

Experimental Units

400

What kind of sample is this?

Randomly select some MMH facilities from each of five geographic regions, and then take random samples from each of these hospital discharge lists.

Stratified Random Sample

400

A group of experimental units that are known to be similar in some way that is expected to affect the response to the treatments.

Block

400

When random assignment of experimental units to treatments is carried out separately within each block.

Randomized Block Design

500

If the residuals for five data points are 4, -3, 6, 5, -7, what is S?

6.71

500

When two variables are associated in such a way that their effects on a response variable cannot be distinguished from each other.

Confounding

500

What kind of sample is this?

Randomly selected some MMH facilities and from each of the facilities, ask every person from a discharge list their satisfaction level.

Cluster Sample

500

A single subject receives both treatments in a random order. The response to each treatment is recorded.

Matched Pairs Design

500

A variable that has an important effect on the variables in a study, but is not listed as an explanatory variable.

Lurking Variable

M
e
n
u