Files in this item
|(no description provided)|
|Title:||A two-parameter partial credit model|
|Doctoral Committee Chair(s):||Harnisch, Delwyn L.|
|Department / Program:||Education|
|Degree Granting Institution:||University of Illinois at Urbana-Champaign|
|Subject(s):||Education, Tests and Measurements
|Abstract:||The assessment of partial knowledge has a long-term history in psychometrics literature. Many solutions based on classical test theory have been suggested and proposed. Thereafter, due to their failure in satisfying rigorous theory background and precise estimation, new methodologies based on modern test theory are currently being examined.
Since Masters (1982) proposed the Rasch-type partial credit model, it became the best-known model to score persons' partial knowledge. Unfortunately, several weaknesses criticized in Chapter 2 may be occurring in real testing situations. Hence, the step discrimination parameter is taken into account in Masters' partial credit model and the two-parameter partial credit model is proposed.
The maximum likelihood estimation solutions and a FORTRAN 77 computer program, TPPCM, are described and used to estimate such model parameters in Chapter 3. An illustrative example is also shown, analyzed, and discussed in Chapter 4 to show how model parameters are calibrated, goodnesses-of-fit are tested, and information functions are provided.
From the analysis and discussion of this illustrative example, four conclusions can be drawn from the findings of this research as follows. (1) The existence of the two-parameter partial credit model is confirmed. This model becomes an alternative model to score persons' partial knowledge or calibrate any questionnaire or test with ordered-response formats. (2) Step discriminations provide a good help in partitioning person's performance levels, hence the fitnesses of person ability estimates on the map of performance space are easily spotted. (3) Step information functions are uniquely and differently determined from step discriminations of each item. Hence it implies potentials for item and test design, selection, and construction. (4) The two-parameter partial credit model shares the same features of the one-parameter partial credit model, except that of specific objectivity and parameter separability.
Finally, the important features and limitations of this research are discussed. Implications for future research are also mentioned.
|Rights Information:||Copyright 1991 Yu, Min-Ning|
|Date Available in IDEALS:||2011-05-07|
|Identifier in Online Catalog:||AAI9136777|