Health Measurement Scales A Practical Guide to Their Development and Use

by Streiner, David L.; Norman, Geoffrey R.

• ISBN13: 9780198528470
• ISBN10: 0198528477
• Edition: 3rd
• Format: Paperback
• Copyright: 2003-12-18
• Publisher: Oxford University Press

Summary

This is the new edition of the highly successful practical guide for clinicians developing tools to measure subjective states, attitudes or non-tangible outcomes in their patients. It is widely used by people from many disciplines, who have only a limited knowledge of statistics. This thoroughly updated edition of Health Measurement Scales, Third Edition gives more details on cognitive requirements in answering questions, and how this influences scale development. There is now an expanded discussion of generalizability theory, a completely revised chapter on Item ResponseTheory and many revisions are included, based on the latest research findings. These features combine to provide the most up-to-date guide to measuring scale development available. It synthesizes the theory of scale construction with practical advice, culled from the literature and the authors' experience, about how to develop and validate measurement scales to be used in thehealth sciences. The theory goes into issues of reliability, generalizability theory, validity, the measurement of change, the cognitive requirements of answering questions, and item response theory. Practical issues cover devising the items, biases that may affect the responses, pre-testing andweeding out poorly performing items, combining items into scales, setting cut points, and the practical issues of using scales in various ways, such as face-to-face interviews; mailed or telephone-administered surveys; and over the internet. One chapter also discusses some of the ethical issues thatscale developers and users should be aware of. Appendices lead the reader to other readings; sources of already developed scales and items; and a very brief introduction to exploratory and confirmatory factor analysis.

Author Biography

David Streiner was trained in Clinical Psychology, and received his Ph.D. in 1968 from Syracuse University. He is the Assistant Vice President for Research, Director, Kunin-Lunenfeld Applied Research Unit at the Baycrest Centre for Geriatric Care. His primary research interests, broadly speaking, include the psychological effects of medical disorders and treatments, and how to apply psychological ways of thinking to other areas. He is involved in a series of studies looking at the psychosocial effects of environmental hazards, and the relationship between stress and the "sick building syndrome". His other studies investigate the cognitive effects of anti-malarial medications, and factors which predispose travellers to do risky things.
Geoffrey Norman's research falls in three broad domains: 1) research on clinical reasoning, particularly in the relative contribution of rapid processing based on prior experience (so-called pattern recognition) and analytical rules; 2) assessment of students (reliability and validity of various approaches from multiple choice tests to undetected standardized patients; 3) methodology of measurement, particularly as it pertains to the measurement of change. He is

Table of Contents

Introduction	1	(3)
References	3	(1)
Basic concepts	4	(10)
Searching the literature	4	(1)
Critical review	5	(4)
The two traditions of assessment	9	(3)
Summary	12	(1)
References	12	(2)
Devising the items	14	(15)
The source of items	15	(4)
Content validity	19	(3)
Generic versus specific scales and the `fidelity versus bandwidth' issue	22	(1)
Translation	23	(3)
References	26	(3)
Scaling responses	29	(32)
Introduction	29	(1)
Some basic concepts	29	(1)
Categorical judgements	30	(2)
Continuous judgements	32	(21)
To rate or to rank	53	(1)
Multidimensional scaling	54	(2)
References	56	(5)
Selecting the items	61	(19)
Interpretability	61	(5)
Face validity	66	(1)
Frequency of endorsement and discrimination	67	(1)
Homogeneity of the items	68	(5)
Multifactor inventories	73	(1)
When homogeneity does not matter	74	(2)
Putting it all together	76	(1)
References	76	(4)
Biases in responding	80	(22)
The differing perspectives	80	(1)
Answering questions: the cognitive requirements	81	(3)
Optimizing and satisficing	84	(1)
Social desirability and faking good	85	(4)
Deviation and faking bad	89	(3)
Yea-saying or acquiescence	92	(1)
End-aversion, positive skew, and halo	93	(2)
Framing	95	(1)
Biases related to the measurement of change	96	(2)
References	98	(4)
From items to scales	102	(24)
Weighting the items	102	(3)
Multiplicative composite scores	105	(3)
Transforming the final score	108	(1)
Percentiles	109	(2)
Standard and standardized scores	111	(2)
Normalized scores	113	(1)
Age and sex norms	113	(2)
Establishing cut points	115	(8)
Summary	123	(1)
References	123	(3)
Reliability	126	(27)
Basic concepts	126	(2)
Philosophical implications	128	(2)
Defining the reliability of a test	130	(3)
Other considerations in calculating the reliability of a test	133	(4)
Other types of reliability	137	(1)
Different forms of the reliability coefficient	138	(4)
Issues of interpretation	142	(4)
Improving reliability	146	(2)
Standard error of the reliability coefficient and sample size	148	(3)
Summary	151	(1)
References	151	(2)
Generalizability theory	153	(19)
G studies	155	(1)
D studies	155	(1)
Example 1---therapists, occasions, and patients	156	(4)
D study examples	160	(2)
Example 2---items, observers, and stations (the OSCE)	162	(2)
Example 3---econometric vs. psychometric perspectives on the utility of health states	164	(2)
Perspective 1: econometric	166	(1)
Perspective 2: psychometric	166	(1)
Perspective 3: experimental	166	(1)
General rules for generalizability	167	(3)
Nested designs	170	(1)
Error estimates for G coefficients	170	(1)
Summary	170	(1)
References	170	(2)
Validity	172	(22)
Why assess validity?	172	(1)
Reliability and validity	173	(1)
The`types' of validity	174	(1)
Content validity	175	(1)
Criterion validity	176	(2)
Construct validity	178	(8)
Responsiveness and sensitivity to change	186	(1)
Validity and`types of indices'	186	(1)
Biases in validity assessment	187	(5)
Changes in the sample	192	(1)
Summary	192	(1)
References	192	(2)
Measuring change	194	(19)
Introduction	194	(1)
The goal of measurement of change	194	(1)
Why not measure change directly?	195	(1)
Measures of association---reliability and sensitivity to change	196	(5)
Difficulties with change scores in experimental designs	201	(1)
Change scores and quasi-experimental designs	202	(2)
Measuring change using multiple observations: growth curves	204	(5)
How much change is enough?	209	(1)
Summary	210	(1)
References	210	(3)
Item response theory	213	(15)
Item characteristic curves	214	(2)
The one-parameter model	216	(1)
The two- and three-parameter models	217	(1)
Polytomous models	218	(2)
Item fit	220	(2)
Person fit	222	(1)
The standard error of measurement	222	(1)
Sample size	222	(1)
Advantages	223	(1)
Disadvantages	224	(1)
Computer programs	225	(1)
References	226	(2)
Methods of administration	228	(20)
Face-to-face interviews	228	(3)
Telephone questionnaires	231	(3)
Mailed questionnaires	234	(5)
The necessity of persistence	239	(2)
Computer-assisted administration	241	(2)
Using e-mail and the Web	243	(1)
References	244	(4)
Ethical considerations	248	(23)
References	253	(1)
Appendices
A Further reading	254	(3)
B Where to find tests	257	(8)
C A (very) brief introduction to factor analysis	265	(6)
Author Index	271	(6)
Subject Index	277