STATISTICAL METHOD
Rasch Analysis
Statistical Consulting for Measurement Scales
We test whether your items form a measurement scale under the Rasch model, reading fit statistics in context, checking whether response categories work, and reporting person and item measures with their precision and any group differences.
No Commitment Required
Strictly Confidential
In-House PhD-Level Statisticians
Fixed Project Fee
Same-Day Response
METHOD OVERVIEW
What Is Rasch Analysis?
Rasch analysis applies the one-parameter logistic family of measurement models, in which the probability of a response depends only on the difference between a person’s trait level and an item’s difficulty, and all items share one discrimination. When the data fit, persons and items sit on one logit scale, the total score is a sufficient statistic for the person measure, and comparisons between persons do not depend on which items were used, and vice versa, the property called specific objectivity. The analysis examines item and person fit, category functioning, local dependence and whether items behave the same across groups. OnlineSPSS, a specialist statistical consultancy for academic, clinical and professional research, provides Rasch analysis within your study design, with the relevant checks, effect estimates, uncertainty measures, interpretation and publication-ready reporting.
WHEN TO ENGAGE
When Is a Rasch Model Appropriate?
A Rasch model fits a scale when the aim is a unidimensional measure with items held to a standard, and these are the situations in which researchers most often ask us to apply it.
Constructing a Measurement Scale A questionnaire or test is being built, and the team wants items and persons placed on one scale, with items that do not fit the model identified.
Rating Categories That May Fail Ordered response options, such as five-point agreement, may not be used as intended, so category thresholds must be checked and categories sometimes collapsed.
Scores Used in Group Comparisons Person measures will be compared across groups or time, which requires checking item fit and differential item functioning before any comparison is made.
Reviewing a Completed Analysis A Rasch model was run on default fit statistics, and a reviewer has doubted the reliability. Independent Statistical Review looks over the thresholds, fit and dependence.
ANALYTICAL APPROACH
How We Apply the Rasch Model
Within the family of Structural & Latent Variable Modeling, the Rasch model is a standard that items are asked to meet, not a curve fitted to any data.
The Model as a Standard Under the Rasch model, items are held to equal discrimination, so misfit signals a problem with the item or the sample. Misfitting items are examined for content issues before any is removed.
Fit Statistics in Context We read infit and outfit mean-squares alongside item content and the principal components of the residuals, never as pass-fail rules, since their expected range shifts with sample size and targeting.
Thresholds and Dependence We inspect category thresholds for disorder and residual correlations, because near-duplicate items inflate apparent reliability and distort fit. Categories are collapsed only when content supports it.
Targeting and Group Equivalence Person-item maps show whether items cover the sample's range, and DIF is checked before groups are compared, with person measures reported with their standard errors, not as exact values.
DELIVERABLES
What You Receive
As part of our Quantitative Data Analysis service, the scale analysis is delivered with the estimates, checks and files needed to report and defend it.
Data & Variable Review Item direction, response categories, category frequencies, extreme scores and missing responses are reviewed, and persons with all-correct or all-zero responses are handled explicitly.
Method-Specific Checks Item and person fit, threshold ordering, local dependence, residual dimensionality and DIF are examined, with the effect of removing misfitting items on the conclusions reported.
Statistical Results Item difficulties and thresholds with standard errors, person measures, person and item separation reliability, and DIF contrasts with their confidence intervals.
Publication-Ready Tables & Figures A person-item map, category probability curves and item fit tables, formatted to APA 7, COSMIN for health measurement instruments or the journal's requirements, ready to insert.
Interpretation The scale is described by what it measures, where it targets and how well it separates persons, with reliability read for this sample and not assumed for the instrument.
Reproducible Analysis Files R scripts using TAM or eRm, Stata do-files, Mplus inputs or SAS code are provided where applicable, so every estimate can be traced and rerun if the data change.
ALTERNATIVE METHODS
When Another Method May Be More Appropriate
We choose the method by how the items are meant to be used. A fixed measurement standard suits Rasch work, and these methods fit other aims and philosophies.
01
Item Response Theory (IRT)
When item discrimination or guessing may vary and fit to the data, not a fixed standard, should drive the decision, two- and three-parameter models are available.
02
Exploratory Factor Analysis (EFA)
When it is unclear that the items form one dimension at all, exploring their structure comes first, since a Rasch model presupposes a single trait.
03
Reliability Analysis (Cronbach's Alpha)
When the only question is whether a summed score is internally consistent, a reliability coefficient answers it, without testing whether the items form a linear measure.
04
Likert Scale Analysis
When ordinal ratings are summarized and compared as they are, Likert methods answer the question and no interval measure on a logit scale is required.
CASE EXAMPLE
How We Work in Practice
An illustration of how a scale evaluation moves from a planned model to a reportable, defensible result.
ILLUSTRATIVE EXAMPLE · PALLIATIVE CARE MEASUREMENT · UNITED STATES
A hospice organization's quality unit wanted to evaluate a 12-item family caregiver burden questionnaire with five response options before using it in routine reporting.
The plan named a rating scale Rasch model, with the partial credit model as a comparison, and fixed the rules beforehand: categories could be collapsed only when thresholds were disordered and the content supported it, and items would be removed only after review of their wording. Two adjacent response options were rarely used and showed disordered thresholds. We collapsed them under that rule and reported both versions. Two items with correlated residuals were examined for overlapping content, and differential item functioning by caregiver relationship was tested before any group comparison.
Outcome: The unit received item and person fit tables, thresholds before and after collapsing, a person-item map, separation reliability, DIF contrasts with 95% confidence intervals and the code.
This example does not describe a real client or project. We never share details of actual client work, in line with the confidentiality we promise every researcher. This illustrative example is included to show the type, scope, and standard of work involved in this kind of analysis.
SOFTWARE
Statistical Software for Rasch Models
Rasch models are estimated in R with TAM or eRm, while Stata, Mplus and SAS fit the Rasch form as a constrained one-parameter model, with code provided.
RESEARCH DISCIPLINES
Rasch Measurement Across Research Fields
Rasch measurement appears wherever questionnaires and tests are constructed or refined, each field with its own scales, populations and reporting conventions.
A Defined Process for Your Analysis
We review your dataset, research question and design before any work begins, so the scope, deliverables, timeline and fixed project fee are clear from the outset.
01
Share Your Research Materials
Send your dataset, research question, study design and any existing analysis plan or output. We will let you know if anything else is needed to assess the project.
02
Receive Your Assessment
Your statistically trained Research Coordinator reviews the project and confirms the required analysis, deliverables, timeline and fixed project price before work begins.
03
Analysis & Delivery
Once you confirm the scope, an in-house statistician completes the agreed analysis and delivers the outputs, interpretation and reporting files.
RESEARCHER VOICE
What Researchers Say
Feedback from a researcher who wanted the reliability estimate checked against the scale and rating design.
“I wanted a second set of statistical eyes on our Rasch Analysis, mainly because of whether the reliability estimate matched the scale or rating design. The review was practical and specific, and the final results were much easier to explain to the rest of the research team. The final output was concise, clearly interpreted, and ready to use in our research reporting.”
RESEARCH RESPONSIBILITY
Statistical Support Without Taking Over Your Research
We provide the statistical work for your scale analysis, and the study, its meaning and its authorship stay entirely with you.
Agreed Statistical Support We provide the agreed methodological, analytical and statistical reporting support for your scale analysis, within the scope confirmed before work begins.
No Intellectual Contribution OnlineSPSS claims no intellectual contribution to your research. Our work is statistical support, and authorship remains yours.
Your Research Decisions The construct, the decision to revise or retire an item, interpretation and conclusions remain with you, and our statistical interpretation stays within what the model and design support.
Confidential Handling Research data and project details are handled confidentially and are not published. An NDA is available where required, and can be arranged before sensitive materials are shared.
FREQUENTLY ASKED QUESTIONS
Questions About Rasch Measurement
Short answers to what researchers ask before a scale analysis, from item fit to confidentiality and the assessment process.
Should I use the Rasch model or a two-parameter IRT model?
The choice is partly philosophical and partly practical. Rasch work treats equal item discrimination as a requirement of measurement and removes or revises items that violate it, while two-parameter models let discrimination vary and fit the model to the data. Rasch needs smaller samples and keeps total scores sufficient, but it may discard informative items.
Does an item that misfits always have to be deleted?
No. Infit and outfit statistics depend on sample size and on the targeting of the sample, so we read them with residual patterns and item content, not as pass-fail cutoffs. Misfit can come from wording, local dependence or an unrepresentative sample. We examine the reason first, and removal is a decision made with you.
Why were my response categories collapsed or flagged?
Disordered thresholds usually indicate that respondents did not use the options in the intended order, often because adjacent categories are seldom chosen or hard to distinguish. Collapsing can restore order, but it changes the scale, so it needs a substantive rationale and a comparison with the original categories rather than a purely statistical decision.
Can you review an analysis that has already been completed?
Yes, we can review a completed Rasch model before submission or a response to reviewers. We check the model choice, item and person fit, category thresholds, local dependence, the DIF procedure and how reliability was reported, and confirm clearly what stands or needs to change before resubmission.
How do you protect my dataset and research confidentiality?
Your dataset, output files and any protocol or analysis plan are handled confidentially, and project information is not published. An NDA is available where required and can be arranged before sensitive materials are shared. Read more about confidentiality and data security.
What happens when I request an assessment, and how long does it take?
You share your research question, design and relevant materials, and a statistically trained Research Coordinator reviews the request and responds the same day. The analysis, deliverables, timeline and one fixed project fee are then confirmed before any work begins, so the timeline follows from the agreed scope.
Still have a question? Find more answers in our full FAQ.
NEXT STEP
Need Support With a
Rasch Model?
Share your research question, study design and data. We assess whether a Rasch model fits your scale and confirm the approach before any work begins.
No Commitment Required
Strictly Confidential
Fixed Project Fee
In-House PhD-Level Statisticians
Same-Day Response
Serving Researchers Since 2014