Introduction to Measurement TheoryThis book is intended to serve as a text and reference book for people who are using or constructing psychological tests and interpreting test scores and scales. It is designed for people who understand collage algebra and who have some famuliarity with elementary statistics, for those lacking this familiarity or desiring a review. |
From inside the book
Results 1-3 of 16
Page 80
... Spearman - Brown formula will be large if the test halves are highly correlated and small if they are not . The halves will correlate highly only if they measure traits that are the same or that are highly correlated ; thus , the Spearman - ...
... Spearman - Brown formula will be large if the test halves are highly correlated and small if they are not . The halves will correlate highly only if they measure traits that are the same or that are highly correlated ; thus , the Spearman - ...
Page 85
... Spearman - Brown Formula : The General Case Another method for calculating a test's reliability is the Spearman - Brown formula , which utilizes information about the reliability of parallel components of the test . The Spearman - Brown ...
... Spearman - Brown Formula : The General Case Another method for calculating a test's reliability is the Spearman - Brown formula , which utilizes information about the reliability of parallel components of the test . The Spearman - Brown ...
Page 88
... Brown formula can underestimate the test reliability . For example , suppose that a ten - item test has a reliability of 0 . The Spearman - Brown formula would estimate that doubling the length of the test with a parallel component ...
... Brown formula can underestimate the test reliability . For example , suppose that a ten - item test has a reliability of 0 . The Spearman - Brown formula would estimate that doubling the length of the test with a parallel component ...
Contents
Introduction | 1 |
Classical TrueScore Theory | 56 |
Reliability | 72 |
Copyright | |
9 other sections not shown
Other editions - View all
Common terms and phrases
assumptions base rate binomial-error model biserial calculate ceiling effect Chapter classical true-score theory confidence intervals correlation coefficient criterion scores criterion-referenced tests cutting score dichotomous discrimination equal error of measurement error scores error variance exam examinee's examinees example expected factor analysis Figure formula scores frequency distribution grade scores homoscedasticity information function interpretation item difficulty item-characteristic curve large number latent latent-trait models latent-trait values level of measurement linear logistic model mean measurement theory method normal distribution normal-ogive number of items observed scores observed test scores observed-score variance obtained p₁ parallel tests phi coefficient point-biserial correlation population prediction probability procedures produce Rasch's ratio raw scores regression line relationship sample scale values Section selection Spearman-Brown formula standard deviation standard error standardized scores statistical Table test developer test reliability test user test's total test scores trait value transformation true scores true-score variance validity coefficient variable X₁ Y₁ στ