Rasch Model 

The Rasch model (Rasch analysis) is a mathematical model developed by George Rasch. It is used for analysing data from assessments to measure variables such as abilities, attitudes, and personality traits. Rasch models are particularly used in psychometrics, the field concerned with the theory and technique of psychological and educational measurement. In addition, they are increasingly being used in other areas, including the health profession and market research because of their general applicability.

The model is considered to be the “gold standard” against which outcome measures can be calibrated (Bond, 2001). When data from the outcome measure fit the Rasch model, this implies that the tool possesses construct validity and the total score is statistically sufficient - that is no other information is required to assess the patient on that construct.
In addition, Rasch analysis provides more information about the ordering of the scores for each item, whether the scale measures a single dimension, whether the ratio of between-item difficulty is consistent across different groups of patients and whether there is bias for any item responses among subgroups of patients.
Rasch analysis has the potential to transform the ordinal scores of the questionnaire into a linear, interval-level scale so that valid parametric statistical analysis can be applied when evaluating treatment effectiveness in clinical trials.

References

Rasch G. Probabilistic models for some intelligence and attainment tests.
Bond TG, Fox CM. Applying the Rasch model: Fundamental measurement in the human sciences. London: Lawrence Arlbaum Associates 2001.

Any comment about this page?
Your feedback is appreciated. Please click here.

Follow & Share Scientific Spine 

To join Scientific Spine mailing list, click here.

You are here: Scientific Spine > Statistics Guide > Rasch Model