DEFINITION
For each different subject, NAEP creates a scale score based on statistical procedures called item response theory (IRT). IRT is a set of statistical procedures useful in summarizing student performance across a collection of test exercises requiring similar knowledge and skills. All NAEP subject area scales are produced using these procedures. In Math and Reading, NAEP creates a scale ranging from 0-500. In Science and Writing, NAEP creates a scale ranging from 0-300.
A correlation is a statistical measure of similarity between at least two given sets of data. StateMaster's
correlations compare two variables from our database and reveal statistical relationships between them. The percentages
you see represent the strength (or likelihood) that a change in the topic variable is matched by a change in the listed
variables below it. But remember: These correlations do not imply causation, that is, one does not necessarily cause
the other. Also, not all variables contain all states, rather subsets of states matched together.
Outliers have been removed only where they are outside 3 standard deviations of the mean.
Only variable pairs where at least 15 states match for each have been considered.
Strength is given by the correlation coefficient (R squared). It is the fraction of variation in Y that can be attributed to the variation in X. 100% signifies a perfect fit (R squared of 1). The top 50 such stats are displayed
DEFINITION: For each different subject, NAEP creates a scale score based on statistical procedures called item response theory (IRT). IRT is a set of statistical procedures useful in summarizing student performance across a collection of test exercises requiring similar knowledge and skills. All NAEP subject area scales are produced using these procedures. In Math and Reading, NAEP creates a scale ranging from 0-500. In Science and Writing, NAEP creates a scale ranging from 0-300.
Subscribe to our Education feeds (stats:
del.icio.us