Evaluating Survey Items and Scales Bonnie L. Halpern-Felsher, Ph.D. Professor University of California, San Francisco.

Slides:



Advertisements
Similar presentations
Survey Methodology Reliability and Validity EPID 626 Lecture 12.
Advertisements

The Research Consumer Evaluates Measurement Reliability and Validity
1 COMM 301: Empirical Research in Communication Kwan M Lee Lect4_1.
Reliability And Validity
VALIDITY AND RELIABILITY
Research Methodology Lecture No : 11 (Goodness Of Measures)
4/25/2015 Marketing Research 1. 4/25/2015Marketing Research2 MEASUREMENT  An attempt to provide an objective estimate of a natural phenomenon ◦ e.g.
Validity and Reliability. Validity Is the translation from concept to operationalization accurately representing the underlying concept. Does your variables.
MGT-491 QUANTITATIVE ANALYSIS AND RESEARCH FOR MANAGEMENT
MEQ Analysis. Outline Validity Validity Reliability Reliability Difficulty Index Difficulty Index Power of Discrimination Power of Discrimination.
Measurement Reliability and Validity
Face, Content & Construct Validity
5/15/2015Marketing Research1 MEASUREMENT  An attempt to provide an objective estimate of a natural phenomenon ◦ e.g. measuring height ◦ or weight.
Measurement. Scales of Measurement Stanley S. Stevens’ Five Criteria for Four Scales Nominal Scales –1. numbers are assigned to objects according to rules.
Reliability and Validity of Research Instruments
VALIDITY OF MEASUREMENT S P M V Subbarao Professor Mechanical Engineering Department Justification for Selection of Concepts to Hardware ?????
RESEARCH METHODS Lecture 18
Reliability and Validity Dr. Roy Cole Department of Geography and Planning GVSU.
MEASUREMENT. Measurement “If you can’t measure it, you can’t manage it.” Bob Donath, Consultant.
SOWK 6003 Social Work Research Week 4 Research process, variables, hypothesis, and research designs By Dr. Paul Wong.
© 2005 The McGraw-Hill Companies, Inc., All Rights Reserved. Chapter 5 Making Systematic Observations.
SOWK 6003 Social Work Research Week 5 Measurement By Dr. Paul Wong.
Psych 231: Research Methods in Psychology
Validity of Selection. Objectives Define Validity Relation between Reliability and Validity Types of Validity Strategies.
Reliability and Validity. Criteria of Measurement Quality How do we judge the relative success (or failure) in measuring various concepts? How do we judge.
Construct Validity and Measurement
Copyright © 2005 The McGraw-Hill Companies, Inc. Permission required for reproduction or display. Personality Assessment, Measurement, and Research Design.
Test Validity S-005. Validity of measurement Reliability refers to consistency –Are we getting something stable over time? –Internally consistent? Validity.
Measurement and Data Quality
EDRS6208 Lecture Three Instruments and Instrumentation Data Collection.
Measurement in Exercise and Sport Psychology Research EPHE 348.
Reliability and Validity what is measured and how well.
VALIDITY, RELIABILITY, and TRIANGULATED STRATEGIES
Instrumentation.
Educational Research: Competencies for Analysis and Application, 9 th edition. Gay, Mills, & Airasian © 2009 Pearson Education, Inc. All rights reserved.
MEASUREMENT CHARACTERISTICS Error & Confidence Reliability, Validity, & Usability.
Copyright © 2012 Wolters Kluwer Health | Lippincott Williams & Wilkins Chapter 14 Measurement and Data Quality.
Technical Adequacy Session One Part Three.
Final Study Guide Research Design. Experimental Research.
Principles of Test Construction
Validity Is the Test Appropriate, Useful, and Meaningful?
Measurement Validity.
+ ©2014 McGraw-Hill Higher Education. All rights reserved. Chapter 2 Personality Assessment, Measurement, and Research Design.
Advanced Research Methods Unit 3 Reliability and Validity.
Developing Measures Concepts as File Folders Three Classes of Things That can be Measured (Kaplan, 1964) Direct Observables--Color of the Apple or a Check.
Political Science 104 Wednesday, October 15 Agenda Reliability vs. Validity Groupwork: Applying what we’ve learned to the newspaper articles Assignment.
Validity and Reliability Neither Valid nor Reliable Reliable but not Valid Valid & Reliable Fairly Valid but not very Reliable Think in terms of ‘the purpose.
Validity Validity: A generic term used to define the degree to which the test measures what it claims to measure.
Presented By Dr / Said Said Elshama  Distinguish between validity and reliability.  Describe different evidences of validity.  Describe methods of.
Research Methodology and Methods of Social Inquiry Nov 8, 2011 Assessing Measurement Reliability & Validity.
Copyright © 2008 Wolters Kluwer Health | Lippincott Williams & Wilkins Chapter 17 Assessing Measurement Quality in Quantitative Studies.
MEASUREMENT. MeasurementThe assignment of numbers to observed phenomena according to certain rules. Rules of CorrespondenceDefines measurement in a given.
PRESENTATION TOPIC : RELIABILITY THEORY PRESENTED BY : ABDUL KHALIQ KHAN PRESENTED TO : SIR MUDASSIR FAROOQI SUBJECT : ADVANCE RESEARCH PROGRAM : MCOM.
Validity and Item Analysis Chapter 4.  Concerns what instrument measures and how well it does so  Not something instrument “has” or “does not have”
DENT 514: Research Methods
Validity & Reliability. OBJECTIVES Define validity and reliability Understand the purpose for needing valid and reliable measures Know the most utilized.
Measurement Chapter 6. Measuring Variables Measurement Classifying units of analysis by categories to represent variable concepts.
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins Chapter 11 Measurement and Data Quality.
VALIDITY What is validity? What are the types of validity? How do you assess validity? How do you improve validity?
ESTABLISHING RELIABILITY AND VALIDITY OF RESEARCH TOOLS Prof. HCL Rawat Principal UCON,BFUHS Faridkot.
Reliability and Validity
Reliability & Validity
Test Validity.
Journalism 614: Reliability and Validity
Introduction to Measurement
Week 3 Class Discussion.
پرسشنامه کارگاه.
5. Reliability and Validity
RESEARCH METHODS Lecture 18
Reliability and Validity
Presentation transcript:

Evaluating Survey Items and Scales Bonnie L. Halpern-Felsher, Ph.D. Professor University of California, San Francisco

Question Evaluation Involves How well questions are understand and how easy they are to answer Affects quality of measurement How well do answers relate to what we are actually trying to measure –Validity and reliability

Standards for Survey Items/Questions Content – are the questions asking the right things? Cognitive – do participants understand and are they able and willing to answer Usability – can the questions be completed easily?

Ways to Test Expert review – have experts in the field review the items and scale to ensure they meet the criteria needed –Great for perspective on what researcher needs, but might not tell about the best way to answer most accurately

Ways to Test Focus groups – hold discussions with people in the sample of interest to inquire as to what terms are used, whether the questions make sense, and whether the content is complete –Great for knowing what people in general think, but not individuals

Ways to Test Cognitive interviews – pilot test the survey and ask questions to samples about what the questions meant, why they gave certain answers, and whether they would change something –Great to tell how individuals understand questions, but might not be generalizable

Looking For: Interviewer: Reads question as worded vs with slight changes vs meaning is altered

Looking For: Respondent: –Interrupts –Asks for clarification –Gives adequate answer –Gives inadequate answer –Answers “I don’t know” –Refuses to answer

Ways to Test Field pretests – pilot test in same manner and to same sample as will do for real survey administration; see what distribution of answers were given; hold debriefings; determine best administration method –Tells how instrument and procedures work under real circumstances and to get summary data; but not flexible to probe and test variations

Ways to Test Randomized or split-ballot experiments – randomly assign different sets of items or wording of items to different groups and compare

Reliable & Valid Measures Reliability: Answers provided are consistent. Validity: Responses relate to some truth concerning what we are trying to describe. Items need to be consistently understood, administered, and communicated.

Reliability

Validity Translation Validity –Does the measurement of the construct reflect well the construct of interest Face validity Content validity

Validity Criterion-related validity –Does the measure behave the way it should given the theory or construct Predictive validity Concurrent validity Convergent validity Discriminant validity

Translation Validity: Face Validity Does the measure “on the surface” or “on the face” seem like a good representation of the construct?

Translation Validity: Content Validity Ensuring that your measure taps into each part of the construct that is necessary to measure

Criterion-Related Validity: Predictive Validity Determine whether the measure can predict something it should predict –E.g., doing well on a certain math exam should predict achievement in engineering

Criterion Validity: Concurrent Validity Can the measure distinguish between groups that it should be able to distinguish between –E.g., assessment of manic-depression should differentiation between manic-depressive patients and schizophrenic patients

Criterion Validity: Convergent Validity Degree to which the measure is similar to (converges with) other similar constructs –E.g., a new measure of IQ would correlate highly with the Stanford Binet IQ test

Convergent Validity: Discriminant Validity Degree to which the measure is NOT similar to (diverges from) other constructs not expected to be similar –E.g., measure of math skills should not be correlated with measure of literacy skills

Reliability Inter-rater or inter-observer Test-retest Parallel Forms Internal Consistency

Inter-rater Reliability Degree to which different raters or observers give consistent estimates of the same issue or phenomenon

Test-Retest Reliability Give same test to same people twice and see how much they correlate

Parallel-Forms Reliability Have 2 related forms and see how they correlate

Internal Consistency Give measure to a sample and estimate how well the items reflect the same construct