Formal and Informal Assessments: Advantages and Disadvantages

Introduction

Modern assessments are so numerous that it is extremely difficult to classify them closely. These assessments have as their primary function the measurement of the results or effects of instruction and learning.  They are intended to test primarily classroom learning.  Educators generally classify assessments into two general categories:  formal and informal.  Formal assessments tend to use standardized forms and many are published by large testing companies while informal assessments are usually created and/or performed by classroom teachers. However, some authors believe that classroom tests can be formal – if they are professionally prepared and meet applicable testing standards.
Formal assessments, as contrasted to informal assessments, are those that have been designed and validated by professional test makers for specific purposes such as measuring academic achievement or literacy levels.  These tests have been pilot tested and administered to representative populations of similar individuals to obtain normative data.  So the main advantage of any formal assessments is its high reliability coefficient and good validity, since they have been tested on representative sample populations.  Hence, normally they can be administered in many different settings and still produce reliable information.

Nowadays, we have two major types of formal assessments that are used to measure student abilities and achievement.  They are the norm-referenced and criterion-reference tests.  The norm-referenced tests attempt to evaluate a particular student’s performance by comparing it to the performance of some other well-defined group of students on the same test (Gage & Berliner (1998). So, while norm-referenced tests measure student performance against that of other student’s, criterion-referenced tests measure it against some agreed- upon level of performance or criterion.  Normally the content and skills and measured on criterion-referenced tests are much more specific than on norm-referenced tests.  So, obviously each provides different types of information for teachers to use.

If the teacher is interested in how his or her students compare to students elsewhere, results from norm-referenced tests are obviously called for.  Norm-referenced assessments therefore have the following advantages: (1) allow comparisons within a particular school, district, or state.  For example, achievement levels in all sixth grades in a particular district might be compared with those from other districts, and (2) high estimates of reliability and validity because the norms are based on large populations. The norm-referenced tests, however, also have disadvantages: (1) doesn’t tell how well a specified set of school or teacher objectives are being accomplishes, and (2) will not tell how students are currently doing in comparison to past performance on locally derived objectives.

Sometimes an educator is not concerned with how well a student performs compared to other students, but whether the students exhibits progress in learning.  So, the teacher utilizes criterion-referenced tests which offer the following advantages: (1) provide information about a student’s level of performance in relation to some specified body of knowledge or list of agreed-upon objectives, (2) information obtained is better for diagnosing student difficulties and for assessing the degree to which school wide or system wide purposes are being achieved.   The results of criterion-referenced tests, however, also has disadvantages: (1) do not allow for comparisons of students in a particular locale with national norms, so criterion-referenced test do cannot produce meaningful comparisons or standards (Mehrens & Ebel, 1997); (2) lack uniform standards, and the interpretation of the scores is only as good as the process used to set the proficiency levels (Hambleton et al., 1998).

Whereas testing specialists have the major responsibility for formal assessments, classroom teachers are responsible for informal assessments.  In general, informal assessments are usually referred to as teacher-made tests or classroom tests.  These tests have not been tested on several sample populations and therefore are not accompanied by normative data.   Four types of informal (alternative) assessments are: multiple- choice tests, matching type tests, true-false tests, and essay tests.

The multiple-choice test commonly consists of providing students with three types of statement: a stem, which poses a problem or asks a question, the right answer, which solves the problem or answers the question correctly, and distracters, several statements which are plausible, but wrong. Multiple-choice test is considered by most test specialists to be the best type of informal assessment as if appropriately written can have these advantages: (1) tap some types of higher-level thinking and analytical skills, and (2) has high reliability and objectivity.  But it also offers the following disadvantages: (1) good multiple-choice items are difficult to write as plausible distractors could be tricky or trivial, and (2) increases the effect of guessing especially if not carefully constructed.

A modified but more difficult form of multiple-choice test is the matching-type test in which alternatives are listed in another column instead of in a series.  In a matching test, there are usually two lists of related facts which are to be matched by means of symbols, with numbers of letters to indicate their proper relationships.  Matching test has the following advantages: (1) covers a fairly large amount and variety of content, and (2) questions are easier to construct than multiple-choice questions, however, since only one response item has to be constructed for each stem.  But according to Gronlund (1998), two disadvantages with matching tests are: (1) finding

homogenous test and response items that are significant in terms of objectives and learning outcomes, and (2) require recall rather than comprehension and more sophisticated levels of thinking, but such items are hard to construct.

            One means of recognizing previously learned facts is the true-and-false, or plus-minus, or right-wrong test.  True-and false tests are made up of a number of statements, some of which are true and others, false. Students taking this test designate which statements are true by answering T, Yes, or True, the false ones are designated by F, No or False.  According to Ebel (1993), the main advantages of true-false items are: (1) their ease of construction and scoring, (2) can cover a large content area, and a large number of items can be presented in a prescribed time period. But critics assert that true-false items have almost no value because of these disadvantages: (1) it encourage, and even reward, guessing, and measure memorization rather than understanding, and (2)  tend to elicit the response set of acquiescence, that is, the response of people who say yes (or “true”) when in doubt (Gage and Berliner, 1998).

To learn how a student thinks, attacks a problem, writes, and utilizes cognitive resources, essay tests are utilized.  Some authorities advocate using words such as “why,” “how,” and “what consequences,” while other test specialists urge words such as “discuss,” “examine,” “explain,”  “identify,” “compare,” and “contrast” in constructing the essay questions (Gage and Berliner, 1998).  Essay tests offers the following advantages: (1) used effectively for determining  how well a student can analyze, synthesize, evaluate, think logically, solve problems, and hypothesize, (2) the ease and short time involved in constructing the essay questions.  The major disadvantages are: (1) the considerable time to read and evaluate answers, and (2) the subjectivity of scoring or grading bias.

References

Ebel, R.L. (1993).  Essentials of Educational Measurement, 3^rd ed. Macmillan, New York.

Gage, N.L. and Berliner, D.C. (1998).  Educational Psychology, 4^th ed., Mifflin, Boston.

Gronlund, N.E. (1998).  How to Construct Achievement Tests, 4^th ed., Prentice-Hall,

       Englewood Cliffs, NJ.

Hambleton, R. et al. (1998).  “Criterion-Referenced Testing and Measurement: A Review of

       Technical Issues and Developments,” Review of Educational Research, paper presented at

       the annual meeting of the American Educational Research Association, New Orleans.

Mehrens, W.A. & Ebel, R.L. (1997). Measurement in Education, 4th ed., Prentice-Hall,

       Englewood Cliffs, NJ.