Development and validation of a direct current circuit analysis post-test in electronic engineering technology.

摘要

Scope and method of study. The purpose of this study was to construct and validate a post-test in DC Circuit Analysis for electronic engineering technology taught at the level of programs accredited by ABET (Accrediting Board for Engineering and Technology). The test was validated, in four stages, by the usage of a modified Delphi study using a panel of nine experts from eight different ABET accredited schools in Kansas, Missouri, and Texas. In the first stage the experts chose the topics that should be covered in DC Circuit Analysis. In the second, seven of the nine experts ranked the topics in order of importance by a four point Likert scale. In the third stage seven of the nine experts were given a copy of the test and were asked if the questions properly covered the topics chosen for the test. In the fourth stage the test was administered to 46 students in Kansas, Missouri, and Texas. Item analysis was performed by calculating the intercorrelation between test items, the item discrimination index of each test item, and the difficulty index of each test item.; Findings and conclusions. The test covered: (1) Ohm's law, (2) Kirchhoff's voltage law, (3) Kirchhoff's current law, (4) power, (5) parallel resistor circuit analysis, (6) series-parallel resistor circuit analysis, (7) series resistor circuit analysis, (8) scientific notation and metric prefixes, (9) the current divider rule, (10) the definition of current I = Q/t, (11) the voltage divider rule, (12) conductance, (13) Thevenin's theorem, (14) mesh analysis, (15) capacitance, (16) Coulomb's law, (17) DC meter movements, (18) the definition of energy P = W/t, and (19) capacitor static DC circuits. The test had a Kuder-Richardson KR-20 reliability coefficient of.8044. The mean score of the 46 examinees was 21.48, the median score was 22, the standard deviation was 4.73, the scores ranged from 8 to 30 on a 34 question test, and the standard error of measurement was 2.09. With the Fisher's Exact Test the probability was 1.000 that a significant difference between the scores of the upper quartile of the examinees and the lower quartile on every test question. The Fisher's Exact test provided a probability of 1.000 that there was a significant difference on the test as a whole. This provides evidence of the construct validity of the test. A survey of a panel of experts provided evidence of content validity of the test.