Contemporary Educational Psychology/Chapter 10: Teacher-made Assessment Strategies/Selected Response Items

Teachers commonly assess students using questions and items that are multiple choice, matching, and true/false. In selected response items, students choose a response provided by the teacher or test developer, rather than construct one in their own words or by their own actions. Selected response items do not require that students recall information, but only that they recognize the correct answer. Tests with these items are called objective because the results are not influenced by scorers’ judgments or interpretations and so are often machine scored. Eliminating potential errors in scoring increases the reliability of tests but teachers who only use objective tests are liable to reduce the validity of their assessment because objective tests are not appropriate for all learning goals. Effective assessment for learning as well as assessment of learning must be based on aligning the assessment technique to the learning goals and outcomes. For example, if the goal is for students to conduct an experiment then they should be asked to do that rather that than being asked about conducting an experiment.

Common Problems

Selected response items are easy to score, but hard to devise. As a result, teachers often do not spend enough time constructing items. Common problems include these:

1. Unclear wording in the items

• Example: "True or False: Although George Washington was born into a wealthy family, his father died when he was only 11, he worked as a youth as a surveyor of rural lands, and later stood on the balcony of Federal Hall in New York when he took his oath of office in 1789)."

2. Cues that are not related the content being examined.

• A common clue is that all the true statements on a true/false test or the corrective alternatives on a multiple choice test are longer than the untrue statements or the incorrect alternatives.

3. Using negatives (or double negatives) the items.

• Example: “True or False: None of the steps made by the student was unnecessary.”
• A better item: "True or False: All of the steps were necessary” Students often do not notice the negative terms or find them confusing so avoiding them is generally recommended. However, since standardized tests often use negative items, teachers sometimes deliberately include some negative items to give students practice in responding to that format.

4. Taking sentences directly from textbook or lecture notes.

• Removing the words from their context often makes them ambiguous or can change the meaning. For example, a statement from Chapter 3, taken out of context, suggests all children are clumsy “Similarly with jumping, throwing and catching: the large majority of children can do these things, though often a bit clumsily.” The fuller quotation makes it clearer that this sentence refers to 5 year olds: "For some fives, running still looks a bit like a hurried walk, but usually it becomes more coordinated within a year or two. Similarly with jumping, throwing and catching: the large majority of children can do these things, though often a bit clumsily, by the time they start school, and most improve their skills noticeably during the early elementary years” If the abbreviated form was used as the stem in a true/false item, it would be misleading.

5. Avoid Trivial Questions

• Example: "Jean Piaget was born in what year?"

a) 1896

b) 1900

c) 1880

d) 1903

While it important to know approximately when Piaget made his seminal contributions to the understanding of child development, the exact year of his birth (1880) is not important.

Strengths and Weaknesses of Selected Response Items

All types of selected response items have a number of strengths and weaknesses. True/false items are appropriate for measuring factual knowledge such as vocabulary, formulae, dates, proper names, and technical terms. They are very efficient as they use a simple structure that students can easily understand, and take little time to complete. They are also easier to construct than multiple choice and matching items. However, students have a 50% probability of getting the answer correct through guessing so it can be difficult to interpret how much students know from their test scores. Examples of common problems that arise when devising true/false items are in Table 10-3.

In matching items, two parallel columns containing terms, phrases, symbols, or numbers are presented and the student is asked to match the items in the first column with those in the second column. Typically there are more items in the second column to make the task more difficult and to ensure that if a student makes one error they do not have to make another. Matching items most often are used to measure lower level knowledge such as persons and their achievements, dates and historical events, terms and definitions, symbols and concepts, plants or animals and classifications. An example with Spanish language words and their English equivalents is below:

Directions: On the line to the left of the Spanish word in Column A write the letter of the English word in Column B that has the same meaning.

Column A Column B

___ 1. Casa A. Aunt

___ 2. Bebé B. Baby

___ 3. Gata C. Brother

___ 4. Perro D. Cat

___ 5. Hermano E. Dog

                                     F.  	Father

                                     G.  	House 

While matching items may seem easy to devise, it is actually hard to create homogenous lists--lists in which all items are truly comparable. Other problems with matching items and suggested remedies are in Table 10-3. Multiple-choice items are the most commonly used type of objective test items because they have a number of advantages over other objective test items. Most importantly, they can be adapted to assess "higher" levels of thinking such as application, and not simply to "lower" levels such as factual knowledge. The first example below assesses knowledge of a specific fact, whereas the second example assesses application of knowledge.

Who is best known for their work on the development of the morality of justice?

a) Erikson

b) Vygotsky

c) Maslow

d) Kohlberg

Which one of the following best illustrates the law of diminishing returns?

a) A factory doubled its labor force and increased production by 50%

b) The demand for an electronic product increased faster than the supply of the product

c) The population of a country increased faster than agricultural self sufficiency

d) A machine decreased in efficacy as its parts became worn out

There are several other advantages of multiple choice items.

• Students have to recognize the correct answer not just know the incorrect answer as they do in true/false items.
• The opportunity for guessing is reduced because four or five alternatives are usually provided whereas in true/false items students only have to choose between two choices.
• Multiple choice items do not need homogeneous material as matching items do.

But creating good multiple choice test items is difficult and students (maybe including you) often become frustrated when taking a test with poor multiple choice items. Three steps have to be considered when constructing a multiple choice item: formulating a clearly stated problem, identifying plausible alternatives, and removing irrelevant clues to the answer. Common challenged faced in each of these steps are summarized in Table 10-3.

(back to Chapter 10...)