Exam 2

Measure of central tendency

A typical value for a data set.

Measure of dispersion

How different the data points are from one another.

Mean

Central tendency measure for interval data; this is the average.

Median

Central tendency measure for ordinal and interval data; the middle data point.

Mode

Central tendency measure for nominal, ordinal, and interval data; the most frequent value.

Range

Dispersion measure for ordinal and mostly interval data; the span of the data.

Standard deviation

Dispersion measure for interval data; how much a typical data point differs from the mean.

When to use the median instead of the mean

Use the median when there are extreme data points.

Measure of association

Describes statistical strength of the relationship between two or more variables.

Correlation

Tells you the strength of the relationship between two variables.

Levels of correlation

Weak, strong, or no correlation; can be positive or negative.

Perfect positive correlation 1.0

Strong positive correlation .70

No correlation 0

Moderate negative correlation -0.50

Test of significance

Determines if results could simply be due to chance.

Null hypothesis

The negation of the substantive hypothesis.

Substantive hypothesis

The hypothesis you think is correct.

Criterion of significance level

Allows the evaluation of the p-value and is typically set to .05.

How to arrive at the criterion of significance level

Determining the cost of accepting the substantive hypothesis and being wrong compared to the cost of accepting the null hypothesis and being wrong

Comparing the criterion of significance level to p-value

If p-value < significance criterion, accept substantive hypothesis; if p-value > criterion, accept null hypothesis.

Factors affecting p-values

Larger sample sizes lead to smaller p-values; the stronger the correlation between the variables, the smaller the p-value; the type of test

Statistical significance

Unlikely that results were due to chance; findings are consistent across the population.

Practical significance

Whether the magnitude of the difference is relevant is a function of interpretation

Type I error

Incorrectly accepting the substantive hypothesis.

Type II error

Incorrectly accepting the null hypothesis.

Trade-off between Type I and Type II error

Reducing Type I error by decreasing the criterion increases Type II error. Reducing Type II error by increasing the criterion increases Type I error.