Scale Development: Theory and Applications by Dr. Robert F. DeVellis

By Dr. Robert F. DeVellis

A best-seller in its First Edition, Scale improvement: conception and purposes, moment version has been commonly up-to-date and revised to handle alterations within the box and issues that experience grown in significance because the First Edition. extensively followed for graduate classes in departments reminiscent of Psychology, Public health and wellbeing, advertising, Nursing, and schooling, this ebook will end up precious to utilized researchers around the social sciences. New to the Second Edition are figures and sensible information for college kids, a brand new part on face validity (Chapter 4), a considerably extended presentation of issue research (Chapter 6), a brand new bankruptcy (7) on merchandise reaction conception (IRT), insurance of qualitative techniques, and concerns concerning differential merchandise functioning (Chapter 8).

Show description

Read or Download Scale Development: Theory and Applications PDF

Best methodology books

Deleuze and Guattari: an introduction to the politics of desire

This available ebook examines significantly the writings of Deleuze and Guattari, clarifying the information of those notoriously tough thinkers with no over-simplifying them. Divided into 3 sections - wisdom, strength, and Liberation of hope - the ebook offers a scientific account of the highbrow context in addition to an exhaustive research of the major issues informing Deleuze and Guattari's paintings.

Modern Regression Techniques Using R: A Practical Guide

In a manner that's refreshingly enticing and readable, Daniel B. Wright and Kamala London describe the main helpful of those ideas and supply step by step directions, utilizing the freeware R, to research datasets that may be positioned at the books’ web site through the SAGE homepage. strategies coated during this e-book comprise multilevel modeling, ANOVA and ANCOVA, course research, mediation and moderation, logistic regression (generalized linear models), generalized additive versions, and powerful equipment.

The daily power game

Politicians, social scientists, marketers, exchange unionists, church leaders, philosophers, we all in truth have triggered such colossal obscure­ ness and confusion in regards to the time period 'power' that this may rarely be attributed to mere likelihood. it appears, there's quite a bit at stake, at any time when we predict approximately energy or are concerned, that it can be worthy our whereas to maintain the concept that blurred.

Philosophy and Non-Philosophy

Each one iteration invents new practices and new writings of philosophy. Ours must have been in a position to introduce yes mutations that may at the very least be an identical with these of cubism, summary artwork, and twelve-tone serialism: it has simply partly performed so. yet in spite of everything the deconstructions, after Wittgenstein, Heidegger, and Derrida, this call for takes on a unique size: What can we do with philosophy itself?

Extra resources for Scale Development: Theory and Applications

Example text

Core on some scale) to other variables. It is the extent to which a measure "behaves" the way tl1at the construct it purports to measure should behave with regard to established measures of otbcr constructs. So, for example, if we view some variable, based on theory, as positively related to constructs A and B, iccgatively related to C and D, and unrelated to X and Y, U1en a scale that purports to measure that construct should bear a similar relationship to mcas\JTCS of those constructs. Jn olhcr words, our measure should be positively correlated with measures of constructs A and B, negatively correlated with measures of C and D, and uncorrelated with measures of X and Y.

However, U1c similarity between the numerical values of the criterion and the predictor measure prior to an appropriate transformation would have nothing to do wi~h the degree of validity. core on some scale) to other variables. It is the extent to which a measure "behaves" the way tl1at the construct it purports to measure should behave with regard to established measures of otbcr constructs. So, for example, if we view some variable, based on theory, as positively related to constructs A and B, iccgatively related to C and D, and unrelated to X and Y, U1en a scale that purports to measure that construct should bear a similar relationship to mcas\JTCS of those constructs.

However. there are other reliability estimates that apply the same sort of logic to a single set of items. , a scale that docs not have any alternate form), (b) divide that set of items into two subsets. and (c) correlate the subsets to assess reliabi li ty. 40 SCALE DEVELOPMENT A reliability measure of this type is called a split-lwlfreliability. Split-half reliability is really a class rather than a si ogle type of computational metbO

Download PDF sample

Rated 4.79 of 5 – based on 20 votes