Term
| Exploratory Factor Analysis |
|
Definition
| Generally executed to explain the relationships among a large number of variables in terms of a much smaller number of constructs |
|
|
Term
|
Definition
| The vector following the accepted trial vector; to find, multiply by root of eigenvalue instead of eigenvalue |
|
|
Term
|
Definition
| Multiply each column by its sum divided by the RootSS of the matrix |
|
|
Term
|
Definition
| First principal component |
|
|
Term
|
Definition
| Measures variance accounted for by an individual component; RootSS |
|
|
Term
|
Definition
| Sum of the squared loadings for an item; measures the amount of variance in that item accounted for by all the extracted factors |
|
|
Term
|
Definition
| That which all the items hold in common |
|
|
Term
|
Definition
| Vector obtained from multiplying a converged characteristic vector by the square root of the eigenvalue |
|
|
Term
|
Definition
| Adjusting axes to fit variable vectors |
|
|
Term
| Orthogonal vs. Oblique Factors |
|
Definition
| Perpendicular and uncorrelated vs. non-perpendicular and correlated |
|
|
Term
|
Definition
| Perpendicular & uncorrelated vs. non-perpendicular & correlated |
|
|
Term
|
Definition
| Replicable, interpretive parsimony |
|
|
Term
|
Definition
| Eigenvalue vs. Component Number |
|
|
Term
| Lower-Order vs. Higher-Order Factors |
|
Definition
|
|
Term
|
Definition
|
|
Term
|
Definition
| Unambiguous quantification of an attribute |
|
|
Term
|
Definition
| In the order of the amount of information that they provide, the scale types are Nominal, Ordinal, Interval, and Ratio |
|
|
Term
| Ceiling vs. Floor of Scale |
|
Definition
| Maximum score vs. minumum score of scale is too easily reached |
|
|
Term
|
Definition
| A significant zero for the scale vs. a significant interval of change |
|
|
Term
| Criterion-Referenced vs. Norm-Referenced measures |
|
Definition
| Raw score vs. relative score |
|
|
Term
|
Definition
|
|
Term
|
Definition
|
|
Term
| Linear vs. Non-Linear Transformation |
|
Definition
| Ordinal and relative retained vs. only ordinal retained |
|
|
Term
|
Definition
| A specific level of an attribute is identified by only one level in measurement; it does not map onto multiple, different levels of measurement |
|
|
Term
| Scale-to-Attribute Correspondence (Clear vs. Fuzzy) |
|
Definition
| Numeric and attribute map uniquely onto each other vs. only numeric maps uniquely onto attribute |
|
|
Term
|
Definition
| Equal to correlation under ideal measurement assumptions |
|
|
Term
|
Definition
| How consistently the test will yield the same results |
|
|
Term
|
Definition
| The degree to which some component of the score does not change between administrations of the measure |
|
|
Term
| Alternate Forms Reliability |
|
Definition
| All of the items are different but about the same thing |
|
|
Term
| Parallel Forms vs. Alternate Forms |
|
Definition
|
|
Term
| Content-Parallel Test Construction |
|
Definition
| Questions are paired, imitative, and randomly assigned |
|
|
Term
| Content-Equivalent Test Construction |
|
Definition
| Questions drawn independently from domain and identified as homogeneous |
|
|
Term
| Coefficient of Equivalence |
|
Definition
| How much two tests have in common regarding the types of questions included |
|
|
Term
| Standard Error of Measurement/Estimate |
|
Definition
|
|
Term
|
Definition
|
|
