Applying the item response theory to classroom examinations.
OBJECTIVE: The purpose of this research project was to determine if the item response theory (IRT) can successfully be applied to smaller-sized class examinations. METHODS: The Rasch mathematical model (RMM) was selected from the family of IRT models because of its ability to work with smaller sampl...
| Publicado en: | Journal of Manipulative & Physiological Therapeutics Vol. 29; no. 5; pp. 393 - 398 |
|---|---|
| Autor principal: | |
| Formato: | research tables/charts Journal Article |
| Publicado: |
Elsevier B.V.
Jun2006
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=106325160&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 106325160 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 01614754 8GE jtl: Journal of Manipulative & Physiological Therapeutics issn: 01614754 maglogo: N pubinfo: dt: Jun2006 vid: 29 iid: 5 pid: 1265 pub: Elsevier B.V. place: Philadelphia, Pennsylvania artinfo: ui: 106325160 106325160 2009234446 10.1016/j.jmpt.2006.04.006 NLM16762668 106325160 ppf: 393 ppct: 5 formats: fmt: @attributes: type: P tig: atl: Applying the item response theory to classroom examinations. aug: au: Lawson DM sug: subj: Education, Chiropractic Test Taking Data Analysis Software Data Analysis, Statistical Descriptive Statistics Pearson's Correlation Coefficient Rasch Analysis Human ab: OBJECTIVE: The purpose of this research project was to determine if the item response theory (IRT) can successfully be applied to smaller-sized class examinations. METHODS: The Rasch mathematical model (RMM) was selected from the family of IRT models because of its ability to work with smaller sample sizes. Two simulated examinations were created for 100 students by 100-item dichotomous examinations. Examination 2 contained 20 items common with those in examination 1. Examination 1 was systematically exposed to randomly missing student responses and to entire items being removed to determine the robustness of the RMM to missing data. The two examinations were then analyzed with the RMM individually and then in combination. Student scores and IRT measures were compared to determine if the IRT could successfully place the students from the two examinations on the same metric of measure. RESULTS: The student measures were not affected when up to 20% of the student responses were randomly missing. Student measures continued to have high reliability and correlated with full matrix measures for up to 40% of items being dropped from the examination. Student scores and IRT measures correlated highly when the two examinations were combined. CONCLUSIONS: The RMM can be successfully applied to small-sized class examinations, such as those at chiropractic, medical, and other health profession institutions. It is possible to place candidates from different administrations on the same metric of measure if there is as little as a 20% overlap of items between examinations. The RMM could assist faculty in determining if differences in candidate scores are caused by ability or item difficulty. pubtype: Academic Journal doctype: research tables/charts Journal Article ougenre: Article language: English refInfo: holdings: @attributes: islocal: N |
|---|