Formation : Rasch in measurement in the health sciences
Se former avec
UNIVERSITE PIERRE ET MARIE CURIE
- Renseignements :
- Durée : 2 Jours
- Type : En centre
- Diplômant : Oui
-
Prix H.T. € :1700.00
- Objectifs :
- This workshop will explain how to analyze questionnaire ordinal response
data, when this is built to measure latent unidimensional traits. Participants
will learn how to input data, conduct analyses and interpret output. Rasch
analysis can be applied to assessments in a wide range of disciplines,
including health studies (Quality of Life, Rehabilitation, Cognition, Patient
Satisfaction, Pain Evaluation, etc …), education, psychology, marketing,
economics and social sciences.
Many assessments in these disciplines involve a well defined group of
people responding to a set of items for assessment. Generally, the
responses to the items are scored 0 1 (for two ordered categories); or 0, 1 2
(for three ordered categories); or 0, 1,2 3 (for four ordered categories) and
so on, to indicate increasing levels of a response on a given variable such
as health status or academic achievement. Totaling the scores of the items
to give a single score for a person implies that the items are intended to
measure a single variable, often referred to as a unidimensional variable.
The Rasch model is the only item response theory (IRT) model in which
the total score across items characterizes a person totally. It is also the
simplest of such models having the minimum of parameters for the person
(just one), and just one parameter corresponding to each category of an
item. This item parameter is generically referred to as a threshold. There is
just one in the case of a dichotomous item, two in the case of three ordered
categories.
- Public visé :
-
Physicians, biologists, pharmacists, veterinarians, economists,
psychologists, sociologists, behavioral or scientists working in
health/environmental science fields (clinical, epidemiological or biological)
should apply. Master degree level in Biostatistics will be appreciated;
nevertheless, the minimum required is a basic level in statistics, (CESAM).
The appreciation of this level is left to the chairman
- (i) Overview of classical unidimensional measurement models (1h);
(ii) Backward Cronbach Alpha Curve (2h);
(iii) Exploratory and Confirmatory Analysis(2h);
(iv) Overview of the dichotomous Rasch model and the ICC (2h);
(v) Study of elementary tests of fit to the ICC curve (2h);
(vi) Overview of the polytomous Rasch model for partial credit items (4h)
(vii) Concept of DIF (2h);
(viii) A unified approach to assessing and accounting for DIF (3h);
(ix) Analysis of multiple choice items (2h);
(x) Linking and the study of extended responses in ordered categories (2h);
(xii) Introductory sessions on unidimensional unfolding, pairwise preferences and
paired comparisons, test equating, latent regression (2).
(xii) Analysis of real data sets with the interactive facilities on the program
RUMM2020 for Rasch analysis of data. (6h)
- The program will be organized in two sessions of 2 days each followed
by one practical session of 1 day. Session 1 covers chapters 1 to 6,
session 2 chapters 6 to 11 and session 3 chapter 12.
All lectures will involve real examples from the health sciences
alternating with presentation of theory and practical results.
The last session will involve practical work with participants’ own data
sets to learn, to use and take advantage of the facilities in RUMM2020*.
In addition to being given RUMM2020 which will be current till June
30, 2008, participants will be provided with lecture notes and with
reprinted articles on the application of Rasch models.
Participants should bring their own PC laptop. RUMM2020 platform
requirements are:
- Win2000/ME, WinXP and WinNT;
- Pentium preferred.
- 64M RAM is recommended for good performance.1
- 28M RAM is recommended for large data files
The instructional format will combine lecture, question and answer, and
small group instruction. Participants will have an opportunity to analyze
their own data sets.
Number of participants: 6 to 20.
25/04/2008 au 25/04/2008 (75)
17/04/2008 au 18/04/2008 (75)
