two alternatives for each item; usually a point is given for the selection of one the alternatives. / questions reliability (chance plays more of a role and lack of precision): tend to be reliable, therefore, less precise than some other formats.
what is an example of a dichotomous format?
true or false; yes or no
what are advantages to dichotomous format questions?
true or false are very simple, easy administrations, quick scoring, require absolute judgement.
what are disadvantages about dichotomous format questions?
encourages students to memorize material, making it possible to perform very well on content they don't understand.
dichotomous items have advantages for?
personality tests with sub scales; makes scoring easier, all the tester needs to do is count the number of items a person endorses from each subscale.
what is a multiple choice/polychotomous format?
resembles the dichotomous format except that each item has more than two alternatives; good domain coverage, popular method of measuring academic performances, easy to score, takes a little time to take the test.
what are distractors (plausibility)?
must determine which of several alternatives is "correct." distractors are incorrect choices; reliability of an item is not enhanced by distractors, ineffective distractors hurts the reliability of the test.
ineffective distractors do what?
may hurt the reliability of the test.
what is a likert format?
requires that a respondent indicate the degree of agreement with a particular attitudinal question. e.g., 1 to 5 scale items, "I am afraid of heights" --> respond with strongly disagree, disagree, neutral, agree, strongly agree.
what is item analysis?
a general term for a set of methods used to evaluate test items, is one the of the most important aspects of test construction.
what is item difficulty?
defined by the number of people who get a particular item correct; needs a range.
what is an example of item difficulty?
if 84% of the people taking a particular test get item 24 correct, then difficult level for that item is .84 (84% people got that item correct).
the higher the proportion of the people who get an item correct then?
the easier the item is.
what is item discriminability?
high vs low scores; discriminating between people with high scores and people with low scores. (e.g., using point biserial correlations [who is doing well on the overall test]).
what is the item characteristic curve?
Graph showing item difficulty vs. person test score. x axis = test scores and y axis = item difficulty (proportion/percentage) of test takers who get the item correct.