## Psychometric theory |

### What people are saying - Write a review

#### Review: Psychometric Theory

User Review - John Williams - GoodreadsThis book is one of the "gold standard" texts on psychometric theory. While there are many other excellent books on the topic (particularly those delving into single topics, such as factor analysis or ... Read full review

#### Review: Psychometric Theory

User Review - GoodreadsThis book is one of the "gold standard" texts on psychometric theory. While there are many other excellent books on the topic (particularly those delving into single topics, such as factor analysis or ... Read full review

### Contents

chapter | 2 |

TWO scaling models Types of responses required of subjects | 31 |

variance and covariance Variance Correlation and covariance | 104 |

Copyright | |

11 other sections not shown

### Common terms and phrases

actually alternative forms applied approach assumptions attitudes attribute average correlation centroid method Chapter clusters coefficient alpha computed concerning construct validity content validity correct correlation matrix correlations among items covariance cross products developed discriminant discussed distances distribution domain domain-sampling model employed equal estimate example factor analysis factor loadings Figure follows formula guessing Guttman scale highly human abilities hypotheses important individual differences inferential statistics interval scale investigations item analysis item-total correlations linear combination mathematical mean measurement error measures of personality multidimensional scaling multiple correlation multiple-choice normal distribution number of items obtained scores particular personality traits phi coefficients possible predictive problem profiles psychological Q sort ratings ratio scale reliability response styles rotated factors sample self-inventories standard deviation standard scores statements statistical stimuli studies subjects tend theory tion total scores trace lines true scores types usually values variance weights zero