C

Capability [See, Ability, Syn.: Capacity].

Cause-and-Effect Test is one that requires an examinee to state the presumed cause of a specified effect or vice versa. Usually, each question of this type is provided with a set of options of which one may be found as a right cause or effect. Different tests of such type are available for measuring different levels of abilities.

CEEB Score is an abbreviation of College Entrance Examination Board score. It is one of the derived scores and obtained on the basis of the raw score mean and standard deviation of the group (linear standard scores). CEEB score is equivalent to 100z + 500,

where z = x - x#
sx

i.e., raw score - mean
Standard Deviation

Central Tendency - for the purpose of calculating a representative score for a group, the widely used statistic is the average or the arithmetic mean. Since the value is located at the centre of a distribution, it is known as a measure of central tendency. There are three measures of central tendency, namely mean, median and the mode. In a normal distribution, all these three values are located in the same central position (bisecting the area under the normal curve). [Also see, Mean, Median, Mode].

Certification generally refers to a statement made by an organization or official body that an examinee (or an Institution) has compiled wit or met certain standards of excellence. This is a post-evaluation/examination activity. In other words, it is usually preceded by an evaluation or an examination.

Character - attitudes constitute character. It is an important non-scholastic aspect of the learners' growth. It makes the learner to behave in accordance with society approved ideals.

Characterising/Characterization is one of the affective characteristics. It refers to the formation of character which is the final stage in the affective process, as stated in taxonomy of educational objectives. Once values are hierarchically organized and a basic philosophy is established, then the learner will integrate his own beliefs and ideas or attitudes into that structure. It may take place in the form of completing, requiring, revising, changing, resolving of an action and the like.

Checking is a kind of verification or further testing to ensure the presence or absence of something. In statistical analysis, this refers to the accuracy and precision of an observation, measurement or computation. Checklist and the technique of cross questioning are usually employed for this purpose.

Checklist is one of the evaluation tools generally used for the measurement of non-scholastic aspects. It is simply a list of behaviours or personality descriptions or traits or characteristics that are checked by the teacher. This device can be made more objective if each skill is broken down into its components. Regularity, interest, keenness, etc., are some of the affective non-scholastic aspects of learners' growth which could be effectively assessed through this device.

Chi-square is a test of independence or goodness of fit used by setting up of a contingency table where two or more variables involved. Chi-square as per Henning (1987, p.189) is a powerful non-parametric statistical procedure that is used to test independence of categorical variables or goodness of fit to mathematical expectancy models. Part of its value stems from its not assuming a normal distribution.

Choice is the act or power of choosing a thing from among two or more options or objects or acts, usually after a deep-thinking. In examinations, it usually refers to the choice of options/alternatives. {Also see, Multiple Choice Questions}.

Choice Index (CI) is a measure of the popularity of a question. It is one of the four important measures of question indices. CI is the percentage of total population that attempted the question. A compulsory question attempted by all will have a CI of 100%, others will range from 0 to 100 in a choice type examination of traditional type. The other three indices are Mean Ability Index (MT), Facility Value (FV) and Discrimination Index (DI).

Chronological Age refers to the actual age of a person i.e., the number of years and months a person has lived forms basis for computing progress.

Classroom Test is generally prepared by the teacher himself to meet the requirements of a particular classroom situation. Tests of this type are often referred as 'Teacher-made Tests' or 'Ad-hoc Tests' which, of course, are not standardized ones.

Clerical Test attempts to test one's ability related to filing, simple book keeping, checking and other routine arithmetic operations. Sometimes typography and machine calculations are also included in this test.

Closed Book Examination is one of the forms of the examination which has been in practice since long. It is usually a 2 to 3 hours examination during which students are required to answer a number of questions that are not known in advance. They are also not allowed to consult/refer to books or any other written materials.

Close-ended Question refers to an objective type item having only one pre-determined correct answer. It is also characterized as 'fixed response question'. [Syn. Selection Type Item].

Cloze Test is a special form of completion type question. This type is used to test an examinee's comprehension of written material. Usually, a passage is mutilated by systematic and random deletions, i.e., every fifth or seventh word is deleted. The students are required to 'cloze the gaps' in the passage byguessing the missing words. "Unlike then normal gap filling tests favoured by examiners 'cloze tests' do not pre-calculate the types of words which are to be tested. Also the mutilation of reading texts by 'cloze' methods usually extends over the whole passage, whereas 'gap filling' tests are usually limited to single sentence itemization" (Taylor, 1975).

Clue in testing, refers to any sign or piece of evidence that helps the examinee to develop, choose or to locate the correct answer/option. This is one of the characteristics of objective test items/questions that give unintended assistance for answering them.

Coefficient is a measure of ratio and is used to denote a dimensionless statistic (e.g., the coefficient of product moment correlation). The word 'index' is also sometimes used in place of coefficient. (Also see, Index).

Coefficient of Alienation is the proportion of variance in one test not accounted for by the variance in another test. It is calculated as one minus the square of the correlation between the two tests. If r is the product moment correlation between two variables, the coefficient of alienation k is Ö1-r2. It is equal to the square root of the coefficient of non-determination.

r = N S xy - S(x) S(y) _

Ö{N(Sx2)-( Sx)2} {NSy2-(Sy)2}

where 'x' is one characteristic
'y' is another characteristic.
'N' is the number of pairs.

Coefficient of Determination is the proportion of variance in one test accounted for by variance in the other test. It is computed as the square of the coefficient of correlation between the two tests. If the product moment correlation is 'r', the coefficient of determination is r2. This quantity is also known as 'Index of Determination'.

Coefficient of Equivalence is the product moment correlation coefficient between the scores of two parallel forms of a test when the forms are administered to the same group of individuals within a reasonable time gap.

Coefficient of Internal Consistency refers to an index of the extent to which various parts of a test measure the same function.

Coefficient of Precision is the product-moment correlation coefficient between two truly equivalent forms of a test, both administered to the same group of persons within a reasonable time gap. The coefficient is expected to be high nearer to 1.00.

Coefficient of Stability is obtained when the same test is administered twice to the same individuals It is known as 'Test-retest reliability'.

Cognitive Domain of the Taxonomy of Educational Objectives is associated with knowledge and development of intellectual ability. It is divided into six hierarchical levels, namely knowledge, comprehension, application, analysis, synthesis and evaluation. Some scholars refer the first three levels as Lower Order Abilities and the rest as Higher Order Abilities. Further details of each level have been dealt with under separate entries.

Cohesion means the act of uniting. It is one of the social qualities expected to be developed by a learner especially at the higher education. As this is one of the non-scholastic behaviour to be achieved during the course of this quality needs to be measured.

Communication is the exchange of information to achieve a particular purpose. In communicative assessment, the emphasis will be on the handling of language for interpersonal transactions rather than on the mastery of specific linguistic facts. The criteria for establishing one's communicative value is related to its success and not to the accuracy of the language used.

Communicative Competence Test requires an examinee to use the given language elements for communicative purpose in the extra-linguistic contexts. Here, the focus is given mainly on what is being achieved with the aid of the language rather than on the particular linguistic forms being employed. The test consists of question items pertaining to selection of language elements for contextual use, organization, presentation, etc.

Comparable Scores - scores on different tests expressed on the same scale and have the same relative meaning within some common reference group are said to be comparable scores.

Competence is the ability required for a particular job or vocation such as, arithmetical competence, communicative competence, etc. Specialized techniques of different types are available for assessing one's competence in different areas.

Competition refers to the striving on the part of two or more persons/participants for the same object or post or examination, especially for the goal that is considered superior (e.g., competitive examination). One who competes in a competition is competitor.

Completion Item requires an examinee to recall a word, a number, a phrase or a sentence omitted from a statement and get the blanks filled so that the statement is complete and true. [Also see, Incomplete Statement Item].

Complex Overt Response is the fifth one among the seven hierarchical levels of psychomotor domain of the Taxonomy of Educational Objectives. It is a process of performing a motor act that is complex, i.e., action has to be carried out smoothly and efficiently (after attaining the skill). It has two stages: (1) Resolution of uncertainty and (2) Automatic performance. Since they form parts of the educational objectives, evaluation of the extent of their function becomes necessary.

Composition is considered one of the best devices to measure the complex skill of a student in a particular language. It is an art in which students are expected to gain proficiency besides being evaluated. Although a method of practising the students to improve their proficiency, it is widely used as an instrument for measuring one's ability to organize his thoughts, to choose his vocabulary and to formulate his sentences while writing. Valette (1967, p.157) suggests the following five major types of objectives and appropriate subjects for composition test items :

1. Vocabulary based composition,
2. Grammar oriented composition,
3. 'Point of view' composition,
4. Letter writing conventions, and
5. Thought provoking essays.

Comprehension is the power of understanding. It is the second one among the six hierarchical levels of the scholastic abilities classified under the cognitive domain. This would also form a basis for the development of other higher order abilities, namely application, analysis, synthesis and evaluation. Change, rephrase, restate, transform, etc., are some of the action verbs that would represent comprehension. A number of tools are available for the measurement of this ability.

Comprehension Question intends to measure the power of understanding of an examinee. One's ability to read and understand what is stated or implied in a written passage, is measured through 'Reading Comprehension'. Similarly, his ability to listen and understand a spoken discourse of a relatively lengthy and formal communication such as a lecture, narration or a speech, etc., is measured through 'Listening Comprehension'. Comprehension passages usually include both short answer and multiple facet type questions.

Comprehensive Examination is one that seeks to determine the integrated learning achievement made by an examinee over a period of time, in a relatively large scholastic area. In schools and colleges, such examinations are conducted only at the end of the course or after a period of more than a year.

Comprehensive Speaking Test primary intends to test the oral proficiency of an examinee. It is composed of several sections in which various aspects of the speaking skill viz., pronunciation, stress, etc., are independently measured. This form of speaking test constitutes a separate part in many examinations. Oral examinations, interviews, etc., are synonymous to this.

Computation means the calculation made by using all the four arithmetic fundamentals with numbers and letters. Computation plays an important role in every stage of evaluation (Syn.: Calculation).

Computer Adaptive Testing is a procedure of testing in which computer hardwares and softwares are used to present the test content. It follows the examinee to demonstrate his ability by using the computer aids in the ongoing testing process. Items are usually chosen to match the ability of an individual examinee. Decision point tests, Step ladder tests, Error-controlled tests, Multistage tests, etc., are some of the computer adaptive tests which are in use.

Concurrent Evaluation is sometimes referred as 'monitoring'. It is an internal evaluation carried out as a parallel activity in order to provide constant feedback for the successful implementation of a programme or project. It is a kind of formative as well as diagnostic evaluation being linked with the planning and implementation process. (Also see, Monitoring).

Concurrent Validity - tests are said to have concurrent validity if the scores of these tests highly correlate with the other recognized/established criterion measures.

Conditioning is one of the psychological processes of changing the behaviour of a person.

Confidence Interval or Band is a range of scores around the examinee's obtained score. The standard error of measurement is the basis for forming this range. To construct a confidence interval, the degree of confidence desired is to be selected first. For example, for constructing a confidence interval of 95% level, the Z-score value associated with this level, has to be obtained from a normal curve table (appears in the books on Statistics). The value of that will be 1.96. it is then multiplied by the standard error of measurement say 2.40. Then, the confidence interval is worked out by multiplying 1.96 x 2.40 = ± 4.7. The value ±4.7 will be added to or subtracted from the obtained score, so that it could form a range or band, i.e., 90.3 - 99.7.

Consistency is one of the characteristics of reliability. It means the accuracy of measurement. When a measurement is repeated on different occasions or with different instruments or by different persons, the accuracy of result is expected to be the same. It is said in such case that there is consistency I the measurement and also in the growth of the learners.

Consistency Index refers to the extent to which the members of a group provide the same result (responses) to the same test (task) after a specific period of time.

Constant Alternative Item is an objective test item classified under 'Selection type'. It requires an examinee to choose the right answer from the given two alternatives that remain the same for a series of items. It can take a variety of forms. The most common forms are providing the alternatives in terms of True-False, Yes/No, Agree/Disagree, etc., for a given statement or question. Sometimes, a series of True-False items may follow a passage involving comprehension.

Construct Validity of a test (Ebel, 1979) is the accuracy with which it presumably measures some hypothetical operationally undefined personal characteristics such as intelligence, motivation or creativity. The purpose of this validity is to provide evidence that underlying theoretical constructs being measured are themselves valid. For example, if the measurement of 'intelligence' really measures the intelligence for scientific purposes, then this validity would be more valuable.

Content Ability Table is a list of content areas of teaching units and the corresponding quantum of abilities that are to be developed out of learning these units of study. A model table for a particular branch of study is given below :

The above Table is prepared in various other forms also depending on the course/unit of study and subject. We can have two such Tables - one for scholastic abilities and the other for non-scholastic behaviours for each branch of language study. Once the content-ability Table is ready, the preparation of test specification will be very easy. [Also see, Blue Print].

Content Analysis is an analysis that is used to discover and list the ideas, feelings, truth claims and personal references in a text of communication. It would help prepare content-objective Table which ultimately would be useful to set a stage for an effective evaluation of the instructional materials as well as the achievement of the learners of that material.

Content Validity means logical or curricular validity. A test is said to have content validity only if the items of that test adequately represent both the content and the abilities (objectives) of that unit on which the test is based. The test, in this scene, is giving equal weightage to all the aspects. Various aspects of content validity are: Construct, Curricular, Definitional, Face, Logical, Sampling and so on.

Contingency Table is concerned with two or more groups of people/data and each person/datum falls into one of the two or more categories on another variable also. The figures we have will represent the number of people/data falling into the different categories on each dimension. This Table therefore would help to study the relative frequencies of cases in different boxes and to know what relationship they reveal between the variables. The following are some of the shapes/models of Contingency Table, which are useful for statistical analysis of test results and also for the item analysis   

 

 

 

 

Continuous Assessment/Evaluation is the process of a gradual build up of cumulative judgement about the student's performance, based on facts and instruments, over a period of time, usually the entire duration of the course which the student is studying. In other words, it is a continuous awareness of the teacher about the growth of his students in various aspects of their learning.

Continuum is a continuous variable. The value of it may fall in between any two values even when they are close to each other and it is always possible to have a third value. Continuum is used when it is desired to refer to the variable itself (e.g., Continuum).

Contrast is the comparison in order to show the difference between two things or items of any sort, either simultaneously or one after the other. The verb 'contrast' is often used to set questions requiring the answers of complex nature.

Convergence Cue is one of the several types of cues that may appear in the multiple choice answer options. It would help an examinee to guess the correct answer without possessing the knowledge of information tested. This cue may occur when all the distractors focus on domains, and only right option is associated with all domains. Other cues, such as Response cues, Length cues, Inconsistent distractor cues, etc., may also occur in the objective test items.

Convergent Validity is a kind of empirical validity, by its nature, very close/related to concurrent validity. It is used for the establishment of the validity of measurement constructs. This is usually indicated as a monotrait-mono method coefficient on a multi-trait multi-method correlation matrix. The detailed procedures appear in most of the Statistics books.

Cooperation is an act of working together for a common purpose. It is one of the social qualities classified under the non-scholastic behaviours, which the learners are expected to develop during the course of their study and hence it needs evaluation.

Core Curriculum refers to a 'Central Core' which consists of all the conventional subject-matters such as, Mathematics, History, Languages, etc., as the unifying activity. It is sometimes referred as 'common curriculum' or 'common core curriculum'.

Correction usually means the set-righting of the errors/mistakes. In test analysis, it refers to an adjustment used in the computation of arithmetic mean, standard deviation and correlation coefficient for various distribution of scores by the short-cut methods.

Correction for Attenuation or dis-attenuation is a way of holding reliability constant when making comparisons among correlation coefficients. It is made by dividing the correlation coefficient by the square root of the cross-product of reliability.

Formula for this is,
[Henning, 1987]

rCA = rxy __
rttx rtty

where, rCA = the correlation coefficient correlated for attenuation
rxy = correlation between any two variables 'x' and 'y'
rttx = the validity coefficient for the measure of 'x'
variable
rtty = the validity coefficient for the measure of 'y'
variable

Correction for Chance is the adjustment made in the scores obtained on a test. The adjustment is equal to the amount which presumably might have been due to chance or pure guessing. For example, 25 points might be subtracted from each score obtained on a test of 100 multiple choice items with 4 options each, since ¼ of the answers could possibly be right by chance. The procedure, however, rests on the assumption that guessing and chance are the same. Moreover, by doing so the rank order of the scores does not affect although an individual rank gets affected.

Correlation denotes the inter-dependence between quantitative or qualitative data. It is also used to denote the relationship between measurable variables or ranks. Measures of correlation requires a pair of 'events' arising from a single source. The single source usually refers to an individual. For example, scores obtained by an individual in Tamil and English language tests or the ranks obtained by him in Class X and XII. For making a correlation, the essential requirement is that each pair of observations must stem from the same source. The score on Maths by 'x' cannot be compared with the score on Physics by 'y', since the two sets of scores are coming from two different sources. Therefore, it is essential to identify a common source when a correlation is proposed. There are two widely used methods available for measuring the correlation.

1)      Pearson’s Product Moment Correlation [r]

                        r           =                    N Σ xy  -  Σ(x)   Σ(y)          _

                                                  Ö{N(Σx²)-( Σx)²}  {NΣy²-(Σy)²}
             where   ‘x’  is one characteristic

                                    ‘y’  is another characteristic.

                                    ‘N’  is the number of pairs.

                                   2)      Spearman’s Rank Order Correlation Coefficient (P)

                                   P  =   1  -  _6 Σdi^{2
                                                   n(n2-1)}

                  Where ‘di’  is the difference of two sets of ranks of a student.
‘n’  is the number of pairs (or students)

Correlation Coefficient [See, Coefficient of Correlation].

Count Scoring [See, Enumeration].

Creativity is a presumed faculty (Ebel) that enables some to excel in originating ideas, theories, poems, plays, symphonies and so on. It is one o the personal qualities listed under the non-scholastic objectives of education.

Credit means 'greatness', commendation or honour given for some action, achievement, or quality, etc in the area of evaluation and testing, it usually refers to scores or marks or grades or some points awarded to a student in recognition of his growth or achievement with reference to his learning.

Criterion is a standard on which basis the test has been developed and validated. It usually refers to a characteristic or a combination of characteristics used as a basis for judging the validity.

Criterion Groups are those who determine the discriminating power of a test. They differ widely on the characteristic which the test intends to measure. This group is usually a combination of two sets of students - those who score the highest marks and those who score the lowest.

Criterion Reference is a basis for interpreting test results with reference to predetermined behavioural objectives. "In the basic form of criterion reference, the performance can be assessed on a simple 'Yes-No' dichotomy; the testee either performs the task or he does not. In a more complex form, the criteria can be expressed on an external continuum: 'to take 4 minutes to run a mile'; or according to a verbal descriptor: 'to maintain own side of a professional verbal descriptor: 'to maintain own side of a professional dialogue with occasional mother tongue accent and usage'. We give much emphasis to functional and behavioural criteria. Linguistic criteria may be introduced in so far as language is judged to facilitate task accomplishment" (Carroll and Hall, 1985, p.166).

Criterion Referenced Test is designed to measure one's achievement or performance against a criterion or standard which may be determined as a reflection of attainment of a particular objective. For example,

1) Mahendra has pronounced 80 w.p.m. in reading.
2) Arun has scored 90 marks in Mathematics.

In the above examples, the performance is rated on how best they mastered the objectives. It may however be seen that no reference or comparison has been made to the peers or other candidates. (Syn. Domain Referenced Test, Objective Referenced Test).

Criterion Related Validity is a measure of test standardization. If the results of an examination are used to predict the future performance of students and if those who make high marks are later found to succeed in higher classes better than those who make low marks, then the examination is considered to be a valid one with reference to that particular criterion. [Also see, Predictive Validity].

Critical Score is one that divides the ranked scores into two or more distinct groups with reference to some criterion. The pass-fail division point is an appropriate example of critical score. In one set of scores, there may be several critical scores as each one of them may be several critical scores as each one of them may divide the ranked scores into two or more distinct groups with reference to various purposes. (Syn. Cutting Score).

Cross Check Questioning is one of the methods of verifying the correctness of information/facts furnished by an examinee. It is usually done through an alternative form of question. This practice is followed sometimes in the viva-voce examinations/interviews in order to relate their responses to that of the written examinations.

Cross Validation is a process of testing the quality of a test item or a test with different but similar groups of examinees/testees. In other words, it is a method of determining the validity of procedure adopted for one sample by trying it out on a second sample of the same population. Sattler (1982) defines as "the act or process of verifying results obtained with one group by replication with a different but similar group".

Culture-fair Test is a culture free test. It is designed to minimize prejudices/biases associated with socio-cultural experiences of the examinees. It is intended to indicate one's original intelligence or basic aptitudes which are not influenced by the culture. Such tests are valid anywhere in the world and to any person who undertakes the test.

Cumulative Frequency (cf) of a score means the number of cases up to and including that score. It is usually worked out from the bottom towards the top, in order to give the lowest rank to the lowest score. A model frequency table for 40 scores obtained from language trainees on a dictation test is given below. The row at the right indicates the 'cumulative frequency' (cf). The cumulative frequency (see, Table) of the score 5 from the bottom is 3. Similarly, the 'cf' of the score 7 is 4 and so on.

[For some purpose, 'cf' is worked out from the top of the scale to the bottom].

Cumulative frequency distributions are useful for calculating the median and also for assigning a rank position to each member of the sample.

Cumulative Frequency Table

Cumulative Percentage Distribution refers to the distribution of scores. It is [Henning, 1987] obtained by arranging examinees from lowest to highest scoring, finding the examinee's group rank. Thus person 5 in a group of 10 would score 50. this arrangement would be useful for ready reference.

Cumulative Scale is one in which the items may be arranged in such a way that a testee who responds positively to a particular item also responds positively to all items of lower rank order. Having got all the test items in a single dimension, this scale is widely used as an opinion scale.

Curriculum means the aggregate of courses of study given in a school or college. It consists of the following four dimensions: Content, Methods, Purposes and Evaluation. [See, Core Curriculum].

Curricular Validity is a measure of content validity and refers to the degree to which the test items are representative of the abilities/skills that are supposed to be developed by a given course of study. A number of techniques is employed to find out the curricular validity of a course or material.

Cutting Score is often referred as critical score. It is a dividing point in a series of scores that are arranged in the order of merit. All the scores about that point forma specified group and all below the another. In one series, there may be a number of cutting scores. Example: for a pass in M.A. course, the cutting score is usually 50%, whereas it is 80% for the award of distinction. [Syn.: Critical Score].