Educational Measurement Theory

EDUC 865: Spring 2001

Instructor: Ratna Nandakumar, Ph.D.Lectures: T 4:00-7:00 p.m. 

Office: 213 Willard HallRoom: MDH 220

Phone:831-1635

Office Hrs: Mon: 2:30 - 4 p.m. and by appointment

Email: nandakum@udel.edu

Textbooks:

1.Allen, M. & Yen, W. (1979).Introduction to Measurement Theory.California:Brooks/Cole Publishing Co.(AY)

2.Traub, R.E. (1994).Reliability for the Social Sciences (Vol. 3).California:Sage Publications, Inc. (T)

3.Camilli, G. & Shepard, L. A. (1994). Methods for Identifying Biased Items (Vol. 4). California:Sage Publications, Inc. (CS)

Determination of Grades:

Class Participation20%

Midterm Exam40%

Course Project40%

Overview of the Course:

This course deals with the psychometric issues concerning the construction of an educational/psychological instrument or questionnaire on a domain of interest. The course provides the technical knowledge necessary for test construction and responsible test use. The topics to be covered in this course include principles of test construction, classical true-score theory, reliability, validity, scaling, and procedures to identify biased items. 

The main emphasis of the course will be on the conceptual understanding of the theoretical concepts and the logic that underlie standard test construction and their applications to real data. In this regard, each of these conceptual concerns of high quality of assessment and measurement will be applied and explored in the context of real data collection situations of interest to the students. As part of the course, students will construct a test, refine it, collect data, analyze and interpret the results. 

Prerequisites: EDST 665 or equivalent

Reading Assignments and Lectures:

We will have a lecture/discussion format in the class.It shall be assumed that students have read the assigned readings before the class session.Each of you is expected to contribute to class discussions. 

COURSE PROJECT:

You are required to conduct and present a project in class.There are several possible types of projects that can be conducted.

(1) Test construction, validation--development of a paper-and-pencil test designed to measure one or more attributes or traits (e.g., quantitative ability, anxiety, attitude towards war, quality of instruction etc.).The instrument would be given to a sample of subjects and appropriate reliability and validity questions investigated.

(2) Theoretical paper--an in-depth exploration of some concept or technique in test or measurement theory [e.g., (a) a review and critical discussion of factor analysis, (b) a detailed discussion of multiple regression or validity].

(3) Literature review of the research and methodology associated with some instrument [e.g., (a) an extended review and critical discussion of recent research on the validity and reliability of the MMPI, (b) a review and synthesis of recent research on racial differences in ability, including discussion of the statistical aspects of such research].

(4) Computer project--writing a Fortran or PL-1 program to implement one or more statistical procedures of interest in psychometrics, including an application to real, or artificial data [e.g., program to perform detailed item analysis].Prior programming skills would be necessary to do a project of this type.

If there is some other type of project that you would like to do that does not fall into above categories, see me for possible approval.Project proposals of types (2)-(4) should be very specific and well-formulated.

Project Proposal: A proposal outlining the rationale and methodology of your project is due by March 20.Specially, proposals for test construction should include:

(1) Specification of the attribute, trait, or stimulus domain you want to measure.

(2) Brief review of other research on the attribute or domain (include references).

(3) As much detail as possible on the instrument--instructions, item format(s), etc.

(4) Number and type(s) of subjects to be employed; sampling procedures to be used.

(5) Brief discussion of how reliability and validity will be assessed.

Project Report: Your project should be written and formatted in APA publication style. For example, a test construction project should include the following sections:

(1) Abstract.

(2) Introduction--indicating the nature and scope of problem investigated and citing other pertinent research; statement of hypotheses, if any.

(3) Method--description of (a) subject sample; (b) instrument, scales, etc.; (c) administration procedures.

(4) Analysis--description of statistical measures used, specification of item analyses, how reliability and validity were assessed.

(5) Results/Discussion (may be combined)--presentation of results of study; use tables and figures where possible.Discussion section should go into the meaning and significance of major results and relate these to previous pertinent work; discussion of implications.

(***If your study "bombed" (i.e., results not statistically "large" or significant and/or uninterpretable), include in your discussion educated guesses about what went wrong and what improvements in methodology would be made if the study were to be repeated.)

(6) References/Appendices.

Please turn in typewritten copy of your project report in APA style.Include as appendices correlation matrices, the instrument used, programs, etc.

Grievance Policy:Any reasonable request to regrade a particular item on an assignment or examinations will be considered.However, the instructor reserves the right to regrade the entire assignment or paper.



SYLLABUS

WeekSuggested ReadingsTopics

______________________________________________________________________

Feb. 6AY: Chap. 2Review of basic statistical concepts

Feb. 13AY: Chap. 6Principles of test construction

(Delete Sections 6.4-6.6)

Feb. 20AY: Chaps. 3&4True-score theory and Reliabiltiy

T: Chaps. 2-3

Feb. 27AY: Chaps. 4&5; Sec. 6.4Reliability (contd.) 

T: Chaps. 4-6

March 6AY: Chap. 5Validity

T: Chaps. 7-8

March 13AY: Chap. 5Validity

March 20AY: Chap 9Prediction and Selection 

March 27No ClassSpring Break

April 3MidtermTurn in the project proposal

April 10No ClassAERA/NCME conference

April 17AY: Chap. 7Transforming and equating test scores

April 24assigned readingsScaling

May 1 CS: Chaps. 1&4DIF

May 8SIBTESTDIF

May 15Class Presentations

______________________________________________________________________

NOTE THAT THE SCHEDULE IS ONLY APPROXIMATE.SOME ADJUSTMENTS MAY BE DESIRABLE AS A FUNCTION OF STUDENT BACKGROUND AND INTEREST TO EXPAND OR REDUCE COVERAGE OF SOME OF THE ABOVE TOPICS.