Psychometric Study of the admission examination 2008-1 to the Universidad Nacional Mayor de San Marcos

Abstract

In the subsequent exploratory descriptive study, we formulated the following investigation question: Does the 2008-I admission exam of the Major National University of San Marcos, cover all the psychometric requirements demanded by an evaluation of that kind of nature? The General Objective of the investigation is to: Identify the basic psychometric characteristics which the 2008-I admission exam of the Major National University of San Marcos has presented, consisting in the reliability of the test, the validity of the items and their analysis. The investigational design corresponds to the simple descriptive form and has permitted us to analyze the psychometric perspective of the results obtain from 18,588 postulants in the 2008-I admissions process. The instrument utilized in the investigation is the 2008-I admission exam of the UNMSM, which consisted of a total of 100 binary and dichotomic items in multiple choice form, consisting in five alternatives. Half of the Items of the exam, that is 50 questions, correspond to the area of verbal and numeric abilities, each subarea consisting of 25 items. The other part of the test is composed by 50 items that of which corresponds to the different areas of knowledge which have a direct relation to the different study departments of the Major National University of San Marcos has to offer. The statistic process was done taking into account the descriptive and exploratory analysis to describe in a detailed manner the psychometric analysis of the admission exam. The results obtained from the investigation of the 2008-1 process of admission of the UNMSM showed a normal distribution, permitting us to utilize parametric statistics. The results corroborate the consistency between the questions employed by the exam. The internal consistency from the reliability analysis indicate reasonable alpha coefficients in all six areas that were examined, which grants us permission to inferior the existence of reliable results. Amongst the most important conclusions stemming from the following study we can mention: 1) The 2008-I admission exam, en its entirety has presented an acceptable level of reliability en both areas it is comprised of, which are abilities and knowledge. 2) The validity analysis of the items which make up the test, indicate that the exam measures en its entirety what it is set out to measure. 3) The different levels of difficulty in the items are concentrated in the intermediate and hard levels, leaving a small group of questions which are estimated as easy. 4) The item analysis reveals that the questions from the exam have presented in its majority a positive discrimination, that is that the postulants with higher scores on the entire exam answered correctly to more items that of students who obtained lower scores on the exam. 5) In general, one can affirm that the 2008-I admission exam has satisfactory complied with the psychometric requirements necessary for an exam of this nature.