Step 1. A Pregnant Construct
A variable you would like to use - a pregnant question, the answer to which is not obvious.
Step 2. The Construct Person Map
Find persons in your inspired imagination who can typify the extremes of your construct (variable). Think up extreme examples. Converse with them in your mind. Jot down how you expect them to manifest and speak their extremeness.
Now think about persons in between, converse with them, put them into their places along your variable (construct). Continue to jot down what you expect them to do and say.
Step 3. The Construct Concept Map
As your conversations with persons along the line of your Map evolve, put together your ideas about the construct that is emerging. Caption the extremes and the intermediates of your construct. Step back and capture its sweep from less to more. What is your name for this variable you are bringing to life?
Step 4. The Construct Question Map
Now operationalize your construct in terms of a hierarchy of lively, well-constructed questions, questions that spring naturally out of your conversations with the persons you imagined to personify and mark out the gradient of your evolving Map.
Step 5. The Construct Item Map
Simplify your rich questions into clear, clean, brief questionnaire items that can be answered: yes/no, true/false, good/bad, agree/disagree, never/always, a lot/ a little, ...
Step 6. The Rating Scales
For each item, explore whether its yes/no, etc. designation, might be usefully answerable in degrees of agreement, frequency, strength, conviction,.... Find the most natural conversation by which to facilitate a person's unconstrained response to each item. Don't force all scales to have the same labels. Should that happen, fine. But it is not necessary and can produce awkward bedfellows.
Step 7. Ask Yourself Out Loud
Read your questionnaire to yourself out loud. Does it sound right, read smoothly, make sense? Does it flow? Can you answer the questions? Do your answers say what you want to say, tell you what you want to know?
Congratulations! Your research has now advanced beyond 99% of existing published papers!
Step 8. Collect Some Relevant Data
Ask, interview, distribute your questionnaire to a few people. Watch them answer. Read their responses. Listen to their words. Revise the questions. Start data analysis immediately.
Step 9. Check Your Scoring Direction
Which end of your construct is "up"? Are all your items scored or rescored in that same direction? Does more raw score mean more of what you are looking for? Perform your first Rasch analysis. Examine the item response/person measure correlations to make sure they are all positive.
Step 10. Fine-Tune Your Rating Scales
Implement your Rasch analysis with the partial credit model so that you can examine the utility of each item's rating scale. Do the person measures advance substantially along with the advancing categories of each item? Examine the category labels. Are "Don't Know" and "Not Sure" really part of the story? Are "somewhat", "a little" and "slightly" really different category levels? Remember, our purpose is to extract meaning from the data, so we must not allow ourselves to be seduced or overwhelmed by misguided notions of being "fair" to meaningless data. Rescore meaningless and contradictory categories as missing data. Collapse adjacent equivalent categories together.
Step 11. Is Anything Measured?
Look at the person and item distribution maps. Is there spread? Do these spreads mark out real differences in terms of the standard errors of the individual estimates? Check your item and person separations. But do not despair when they are low. Separations are meaningless when a distribution is not unimodal and symmetric. Mark your map in standard error units so you can see how far your construct reaches.
Try some rescoring and some item grouping to find the simplest scoring and least groups of items which maintain good item and person distributions, advancing average category measures, and reduced misfit.
Step 12. The Item Calibrating Map
Where do your items locate on the Map. Compare it with your construct concept map. Do the items mark out your intended construct hierarchy? Do they confirm or improve your Construct/Question Maps? Are you learning new and important answers to your pregnant question?
Which items show misfit? Locate the errant responses - do they cluster by items or by persons? Are some items off-construct? Do some persons have a different perspective? Are some items and persons better put to one side for their own special analysis?
Step 13. The Person Measuring Map
Now you are ready to do some person measuring. You have found the best scoring model and the best items by fit and reconciliation with your construct intentions. What story do the people tell? Are there demographic differences between those high and low on the construct? Are there expected differences? Surprises? Is the person measure distribution unimodal, bimodal, skewed? What does this say about the pregnant question? Does it make sense?
Step 14. Explanatory Maps
Now that you have settled your definition of your construct, you are ready to tell others about it. But it is foolhardy to force anyone else to swallow everything you've discovered and documented in one gulp. They won't be able to digest it all at once.
To start your reporting pick a small set of pivotal, well-separated, typical persons (by grade, age, severity, accomplishment) to mark out the "person type" definitions of your construct. The map positions of these person types provide a normative description of your construct. Give these person types usefully descriptive names.
Now pick a small set of key, easy to grasp, well-spaced items (and their rating scale labels) to document what you mean by the succeeding levels of your construct. This provides a criterion definition of your construct. Give these key items their full text.
Perform a Rasch analysis anchored on all items and steps from your complete, definitive analysis, but select for report only the subset of persons you chose to provide the normative definition of your construct.
Then anchor all persons as well and run once more selecting only your construct defining items to provide the criterion definition of your construct.
For these runs, show plots and tables which map and list your construct definitions and normative implications in terms of the full item and person demographic labels you supplied to mark your construct.
Step 15. Report Rosters/Tables
Now that you have made sense out of your construct, you are ready to report and diagnose individual measures. List your persons in a meaningful order. This is rarely entry order. List your items in a meaningful order, with meaningful wording. Produce maps, plots, tables that clearly and powerfully show what you've learned. Present your results clearly so that others can stand on your shoulders, for you have become one of the giants!
Ben Wright, Mark Stone, Matt Enos
Evolution of meaning in practice. Wright B.D. Rasch Measurement Transactions, 2000, 14:1 p.736
|Rasch Measurement Transactions (free, online)||Rasch Measurement research papers (free, online)||Probabilistic Models for Some Intelligence and Attainment Tests, Georg Rasch||Applying the Rasch Model 3rd. Ed., Bond & Fox||Best Test Design, Wright & Stone|
|Rating Scale Analysis, Wright & Masters||Introduction to Rasch Measurement, E. Smith & R. Smith||Introduction to Many-Facet Rasch Measurement, Thomas Eckes||Invariant Measurement: Using Rasch Models in the Social, Behavioral, and Health Sciences, George Engelhard, Jr.||Statistical Analyses for Language Testers, Rita Green|
|Rasch Models: Foundations, Recent Developments, and Applications, Fischer & Molenaar||Journal of Applied Measurement||Rasch models for measurement, David Andrich||Constructing Measures, Mark Wilson||Rasch Analysis in the Human Sciences, Boone, Stave, Yale|
|in Spanish:||Análisis de Rasch para todos, Agustín Tristán||Mediciones, Posicionamientos y Diagnósticos Competitivos, Juan Ramón Oreja Rodríguez|
|Forum||Rasch Measurement Forum to discuss any Rasch-related topic|
Go to Top of Page
Go to index of all Rasch Measurement Transactions
AERA members: Join the Rasch Measurement SIG and receive the printed version of RMT
Some back issues of RMT are available as bound volumes
Subscribe to Journal of Applied Measurement
Go to Institute for Objective Measurement Home Page. The Rasch Measurement SIG (AERA) thanks the Institute for Objective Measurement for inviting the publication of Rasch Measurement Transactions on the Institute's website, www.rasch.org.
|Coming Rasch-related Events|
|June 29 - July 27, 2018, Fri.-Fri.||On-line workshop: Practical Rasch Measurement - Further Topics (E. Smith, Winsteps), www.statistics.com|
|July 25 - July 27, 2018, Wed.-Fri.||Pacific-Rim Objective Measurement Symposium (PROMS), (Preconference workshops July 23-24, 2018) Fudan University, Shanghai, China "Applying Rasch Measurement in Language Assessment and across the Human Sciences", www.promsociety.org|
|July 29 - August 4, 2018||Vth International Summer School `Applied Psychometrics in Psychology and Education`, Institute of Education at the Higher School of Economics, St. Petersburg, Russia, https://ioe.hse.ru/en/announcements/215681182.html|
|July 30 - Nov., 2018||Online Introduction to Classical and Rasch Measurement Theories (D.Andrich), University of Western Australia, Perth, Australia, http://www.education.uwa.edu.au/ppl/courses|
|Aug. 10 - Sept. 7, 2018, Fri.-Fri.||On-line workshop: Many-Facet Rasch Measurement (E. Smith, Facets), www.statistics.com|
|August 25 - 28, 2018, Sat.-Tue.||Análisis de Rasch introductorio (en español). (Agustín Tristán), Instituto de Evaluación e Ingeniería Avanzada. San Luis Potosí, México. www.ieia.com.mx|
|Sept. 3 - 6, 2018, Mon.-Thurs.||IMEKO World Congress, Belfast, Northern Ireland, www.imeko2018.org|
|Oct. 12 - Nov. 9, 2018, Fri.-Fri.||On-line workshop: Practical Rasch Measurement - Core Topics (E. Smith, Winsteps), www.statistics.com|
The URL of this page is www.rasch.org/rmt/rmt141g.htm