R

Random Number Grid is one of the methods used for randomizing the answer scripts of the public examinations in order to ensure that each bundle of answer scripts sent to the examiners are fairly representative of almost all the educational institutions. The following is the model 'Random Number Rid' constructed for randomizing the register numbers allotted to 50 candidates.

The first step is to draw up a grid with 5 rows and 11 columns (so that 55 entries can be made). The empty cells are filled in as follows. The diagonal cells starting from the top left hand corner are filled in with the consecutive in with the consecutive numbers 1, 2, 3, etc. the direction is reversed at the boundaries and filled continuously by jumping over cells already occupied. When the sequence ends at a corner of the grid a new start is made from any vacant corner of the grid. When this sequence also ends the next start is made from the first vacant space in column number 1. Thus, the grid will be completed.

The first step is to draw up a grid with 5 rows and 11 columns (so that 55 entries can be made). The empty cells are filled in as follows. The diagonal cells starting from the top left hand corner are filled in with the consecutive in with the consecutive numbers 1, 2, 3, etc. the direction is reversed at the boundaries and filled continuously by jumping over cells already occupied. When the sequence ends at a corner of the grid a new start is made from any vacant corner of the grid. When this sequence also ends the next start is made from the first vacant space in column number 1. Thus, the grid will be completed.

Random Parallel Tests refer to the two forms of tests which are composed of items drawn randomly from the same population of items. They have the same number of items with equivalent content. Each of the item of both the tests has a stem with the same number of words and same number of multiple choice distractors.

Random Sample is one which is selected from a population in such a way that each member or constituent of the population has equal probability of being included in a random sample.

Range of Scores is a measure of the variability (spread) of a group of scores. It is computed by subtracting the lowest score from the highest one (and this is influenced by extreme values).

Rank is the position of an item or datum relative to others that have been arranged in the order of certain defined character (merit, size, importance, social status, value to the firm, etc.). The highest position is rank-1, the next highest rank is 2, etc.

Ranking is the process of arranging a group of learners in the order of their merit on the (basis of their scores), either from the highest to the lowest or from the lowest to the highest. It is a method of reporting one's performance in comparison with that of his peers. [Also see, Rank].

Rank Order Correlation is one of the methods of obtaining correlation coefficient by assigning ranks to each score and determining the relationship between sets of ranks. It is usually described as Spearman's Rank Order Correlation Coefficient (p). the formula for computing this is,

p = 6S di2/n(n2-1)

where, ' S ' is the symbol representing the sum of the scores.
'di' is the difference of two sets of ranks of a student.
'n' is the number of pairs.

Rapport is the relationship that is built upon mutual trust between the performers and those who plan or assess the performance.

Rasch Model [See, One Parameter (Rasch) Model, Latent Traint (Measurement) Theory].

Rate (n) refers to an assigned position or a relative condition or quality. It may be of grade, class or rank. The one who is rated is called 'ratee' and who rates is 'rater'. Usage of this term is quite common in the affective measures, i.e., the measures of non-scholastic behaviours.

Rating refers to the award of credits, marks or grades in the process of assessing or valuing or reviewing of an answer script/the traits (non-scholastic aspects of growth) of a learner or a textbook or the like. It would also refer to the classification as well according to the grade or rank.

Rating Scale is one of the tools used to estimate the magnitude of characteristics that are present in the student. It helps to make assessment by 'impression'. Rating scale is composed of 3 or 5 point impression-scales of varying magnitudes.

Ratio Scale is one which measures intervals from a real zero point. Zero indicates the absence of a trait. Measures of height, weight, speed, time, etc., from a known starting point are said to be on a ratio scale.

Raw Score is one which is obtained on a test before any transformation to a standard score or derived score is done. It is usually equal to the total number of correct items on an objective test composed of item types, viz., constant alternatives, multiple choice, etc.

Reader Reliability [See, Marker Reliability].

Readiness Test is one that measures the extent to which an individual possesses the capacity (background) for learning some complex (new) subject. [Also see, Aptitude Test, Prognostic Test].

Reading Age refers to the reading ability of a child, indicated as scores in terms of age.

Reading Test attempts to measure the reading skill of an examinee. It includes the testing of Silent Reading, Loud Reading, Fast Reading, Reading with Comprehension, etc.

Rearrangement Item is one of the selection type test items. It requires an examinee to put into order a set of randomly presented information, material, etc. for example, arranging in chronological order, size or order of events, etc. Tests with this kind of items are mostly used in the beginning stages of learning.

Recall/Item Question is one that requires an examinee to recall necessary facts, forms of structures before organizing a response to a question. This would test the comprehension ability. [Also see, Comprehension Question].

Recall Score Method is a scoring procedure for measuring one's 'retention' of his learning in the past. It is the procedure to find out at the time of testing, the amount of reproduction in comparison to the amount of learning [Syn.: Right Associates Procedure].

Recall Test is a test in which the task is to say or write specific items previously studied by the examinee. {Syn.: Simple Recall Test].

Receiving is one of the affective characteristics. It is the first among the five hierarchical levels of the non-scholastic aspects of learner's growth (under affective domain) as stated in the TEO. It refers to the sensitivity of the learners to phenomena and stimuli. It may happen either in the form of sharing, or accepting or listening.

Receptive Skill is the receiving skill. It is a skill of language acquisition, i.e., spoken and written by others and received by an examinee. The term 'receptive' is contrasted with 'productive' which refers to 'speaking and writing'. [Also see, Listening Test].

Recognition Item/Test is one that requires an examinee to select correct answer from among the given alternatives through a process of recognition. This type of items are used in the tests to diagnose the learners' understanding at the initial stages of their learning. Tests composed of such items are called 'Recognition Tests'.

Regression Analysis is a procedure used to find out the statistical relationship exists between two or more test variables. One of the two purposes of this analysis is to provide estimates of values of the dependent variable from values of the independent variable. The device used to accomplish this estimation procedure is the 'Regression line'. The other purpose is to obtain a measure of the error involved in using the regression line as a basis for estimation. For this purpose, 'standard error of the estimate' (or its square) and 'error variance around the regression line' are calculated. These are the measures of the scatter or spread of the observed values of 'y' around the corresponding values estimated from regression line.

Regression Effect causes to the tendency for a re-test score to be closer to the mean than the initial test score.

Regression Line is the line that describes the relationship between two variables. It according to some theorem best fits the relationships actually found. It is because of this reason, the line is often referred as the 'line of best fit'. In other words, it is a line which displays mean values of 'y' for the given values of 'x'. the equation of this line provides estimates of the dependent variable when values of the independent variable are inserted into the equation,

y = a + bx

Where, y = a given score to be predicted on the independent variable.
a = the 'y'-intercept, the point where the line crosses the
vertical ordinate.
b = the 'slope' of the line.
x = a given score being the basis of prediction from the
'independent variable'

The constant in a linear regression equation that measures the slope of the regression line is 'Regression Coefficient'. Any smooth curve (including a straight line) fitted to the means of one set of arrays in a double entry table is 'Regression Curve' . [Also see, Scatter Diagram].

Relevance is the extent to which different items of a test contribute to the purpose of the test by making the testee to employ all the required abilities in answering a question.

Reliability refers to the consistency of a test in measuring whatever it measures. In other words, it yields the same results when administered to the same results when administered to the same individuals on different occasions the reliability estimate (value) usually ranges between 0 to 1. Inconsistency or unreliability of a test usually occurs in three levels:

1) Test items,
2) Test administration, and
3) The marker inconsistency

Reliability Coefficient is the estimate of the coefficient of correlation between the scores obtained by a group of students on the same test which is administered twice to them with a considerable amount of time gap. This estimate (index value) varies from 0.00 to 1.00. Several procedures are available for estimating reliability coefficient, of which some are given below :

1) Equivalent forms
2) Internal consistency
3) Kuder Richardson
4) Reader or marker
5) Split-Halves
6) Test-Retest

Replication refers to the building of as many features into a test as possible of the situations being tested for. It can be best assured by a careful pre-specification of all the salient features in order to include them in the test itself.

Reporting plays an important role in the evaluation and monitoring process, and without which continuous evaluation/monitoring may not be possible. The main purpose of reporting is therefore to provide timely, regular and dependent information about the functioning of the programme, to the implementing authorities in order to get the progress in the right direction.

Representative Sample refers to a sample which is representative of a population with respect to the characteristics important for the purposes under investigation. [Also see, Sample].

Responding is one of the affective characteristics. It is the second among the five hierarchical levels of the non-scholastic aspects of the learners' growth (under affective domain) as stated in the TEO. 'Responding' means 'response' whereby learner does more than just note that the stimuli. It may happen in any of the following forms, viz., complying with, following, commanding, approving, volunteering and so on.

Response in the area of evaluation refers to the reply or answer given by an examinee to a question. It may be either in the form of a word phrase, sentence or an essay. Also, it may either be oral or written.

Response Count refers to the counting of responses to the items for an objective test. It indicates the frequency with which one or more of the answer options were chosen by examinees in a particular group. It is a time consuming part of the item/test analysis in a particular group.

Response Cues are the cues that are to be avoided while preparing multiple choice items. Students who have prior exposure to certain types of questions/examinations may have a chance of developing 'test-wiseness', i.e., capable of selecting the correct option independently of any knowledge of the correct field being tested, simply on observing the response cues provided in the option itself. Therefore, care should be taken to avoid providing such a scope to the testees/examinees.

Response Validity is an another type of non-empirical validity. This is intended to describe the extent to which the examinees are responding as per the examiners/test developers. If the examinees respond in a haphazard manner their obtained scores may not represent their actual ability. If both the instructions and the test format are not clear and unfamiliar to the students, their responses may not reflect their true ability and in this way the test may be said to lack response validity.

Restricted Book Examination is a modified form of open-book examination. In this, a restriction is imposed on the books that can be used during the examination. [Also see, Open Book Examination].

Result is a phenomenon that follows an action or examination. Declaration of results after the final examinations/selection tests, etc., is quite common in educational institutions. In general evaluation, it obviously refers to the conclusion (i.e., consequence of the data obtained from investigation) leading to the value judgement.

Retention Test is designed to measure the students' ability to retain the language acquired. Such test is usually designed for L2 learners. Retention test is often considered as diagnostic because the retention ability has not yet been proved directly proportionate to comprehension. It would be more useful if the retention test items are arranged in the order of increasing difficulty.

Review implies examining something, making a judgement and putting it usually in the written form it is somewhat a less formal approach than criticism in referring to literary works. Review is distinctly different from evaluation also. It is not as extensive and exhaustive as the evaluation. As Rastogi, et al [1970, p.44] puts it, review consists of only checking up a few observable points mostly related to physical aspect and giving global impression about the quality and suitability of the academic content. A reviewer usually goes through the book and starts giving his reactions on the points listed in the review sheet.

Right Associates Procedure is a very commonly used procedure in the study of learning and retention. Item (usually verbal) are presented in pairs for learning, then the first of each pair is presented for a brief time and the learner endeavours to reproduce the second. The score is the number of successes or of retained numbers.

refer to the two forms of tests which are composed of items drawn randomly from the same population of items. They have the same number of items with equivalent content. Each of the item of both the tests has a stem with the same number of words and same number of multiple choice distractors.

Random Sample is one which is selected from a population in such a way that each member or constituent of the population has equal probability of being included in a random sample.

Range of Scores is a measure of the variability (spread) of a group of scores. It is computed by subtracting the lowest score from the highest one (and this is influenced by extreme values).

Rank is the position of an item or datum relative to others that have been arranged in the order of certain defined character (merit, size, importance, social status, value to the firm, etc.). The highest position is rank-1, the next highest rank is 2, etc.

Ranking is the process of arranging a group of learners in the order of their merit on the (basis of their scores), either from the highest to the lowest or from the lowest to the highest. It is a method of reporting one's performance in comparison with that of his peers. [Also see, Rank].

Rank Order Correlation is one of the methods of obtaining correlation coefficient by assigning ranks to each score and determining the relationship between sets of ranks. It is usually described as Spearman's Rank Order Correlation Coefficient (p). the formula for computing this is,

p = 6S di2/n(n2-1)

where, ' S ' is the symbol representing the sum of the scores.
'di' is the difference of two sets of ranks of a student.
'n' is the number of pairs.

Rapport is the relationship that is built upon mutual trust between the performers and those who plan or assess the performance.

Rasch Model [See, One Parameter (Rasch) Model, Latent Traint (Measurement) Theory].

Rate (n) refers to an assigned position or a relative condition or quality. It may be of grade, class or rank. The one who is rated is called 'ratee' and who rates is 'rater'. Usage of this term is quite common in the affective measures, i.e., the measures of non-scholastic behaviours.

Rating refers to the award of credits, marks or grades in the process of assessing or valuing or reviewing of an answer script/the traits (non-scholastic aspects of growth) of a learner or a textbook or the like. It would also refer to the classification as well according to the grade or rank.

Rating Scale is one of the tools used to estimate the magnitude of characteristics that are present in the student. It helps to make assessment by 'impression'. Rating scale is composed of 3 or 5 point impression-scales of varying magnitudes.

Ratio Scale is one which measures intervals from a real zero point. Zero indicates the absence of a trait. Measures of height, weight, speed, time, etc., from a known starting point are said to be on a ratio scale.

Raw Score is one which is obtained on a test before any transformation to a standard score or derived score is done. It is usually equal to the total number of correct items on an objective test composed of item types, viz., constant alternatives, multiple choice, etc.

Reader Reliability [See, Marker Reliability].

Readiness Test is one that measures the extent to which an individual possesses the capacity (background) for learning some complex (new) subject. [Also see, Aptitude Test, Prognostic Test].

Reading Age refers to the reading ability of a child, indicated as scores in terms of age.

Reading Test attempts to measure the reading skill of an examinee. It includes the testing of Silent Reading, Loud Reading, Fast Reading, Reading with Comprehension, etc.

Rearrangement Item is one of the selection type test items. It requires an examinee to put into order a set of randomly presented information, material, etc. for example, arranging in chronological order, size or order of events, etc. Tests with this kind of items are mostly used in the beginning stages of learning.

Recall/Item Question is one that requires an examinee to recall necessary facts, forms of structures before organizing a response to a question. This would test the comprehension ability. [Also see, Comprehension Question].

Recall Score Method is a scoring procedure for measuring one's 'retention' of his learning in the past. It is the procedure to find out at the time of testing, the amount of reproduction in comparison to the amount of learning [Syn.: Right Associates Procedure].

Recall Test is a test in which the task is to say or write specific items previously studied by the examinee. {Syn.: Simple Recall Test].

Receiving is one of the affective characteristics. It is the first among the five hierarchical levels of the non-scholastic aspects of learner's growth (under affective domain) as stated in the TEO. It refers to the sensitivity of the learners to phenomena and stimuli. It may happen either in the form of sharing, or accepting or listening.

Receptive Skill is the receiving skill. It is a skill of language acquisition, i.e., spoken and written by others and received by an examinee. The term 'receptive' is contrasted with 'productive' which refers to 'speaking and writing'. [Also see, Listening Test].

Recognition Item/Test is one that requires an examinee to select correct answer from among the given alternatives through a process of recognition. This type of items are used in the tests to diagnose the learners' understanding at the initial stages of their learning. Tests composed of such items are called 'Recognition Tests'.

Regression Analysis is a procedure used to find out the statistical relationship exists between two or more test variables. One of the two purposes of this analysis is to provide estimates of values of the dependent variable from values of the independent variable. The device used to accomplish this estimation procedure is the 'Regression line'. The other purpose is to obtain a measure of the error involved in using the regression line as a basis for estimation. For this purpose, 'standard error of the estimate' (or its square) and 'error variance around the regression line' are calculated. These are the measures of the scatter or spread of the observed values of 'y' around the corresponding values estimated from regression line.

Regression Effect causes to the tendency for a re-test score to be closer to the mean than the initial test score.

Regression Line is the line that describes the relationship between two variables. It according to some theorem best fits the relationships actually found. It is because of this reason, the line is often referred as the 'line of best fit'. In other words, it is a line which displays mean values of 'y' for the given values of 'x'. the equation of this line provides estimates of the dependent variable when values of the independent variable are inserted into the equation,

y = a + bx

Where, y = a given score to be predicted on the independent variable.
a = the 'y'-intercept, the point where the line crosses the
vertical ordinate.
b = the 'slope' of the line.
x = a given score being the basis of prediction from the
'independent variable'

The constant in a linear regression equation that measures the slope of the regression line is 'Regression Coefficient'. Any smooth curve (including a straight line) fitted to the means of one set of arrays in a double entry table is 'Regression Curve' . [Also see, Scatter Diagram].

Relevance is the extent to which different items of a test contribute to the purpose of the test by making the testee to employ all the required abilities in answering a question.

Reliability refers to the consistency of a test in measuring whatever it measures. In other words, it yields the same results when administered to the same results when administered to the same individuals on different occasions the reliability estimate (value) usually ranges between 0 to 1. Inconsistency or unreliability of a test usually occurs in three levels:

1) Test items,
2) Test administration, and
3) The marker inconsistency

Reliability Coefficient is the estimate of the coefficient of correlation between the scores obtained by a group of students on the same test which is administered twice to them with a considerable amount of time gap. This estimate (index value) varies from 0.00 to 1.00. Several procedures are available for estimating reliability coefficient, of which some are given below :

1) Equivalent forms
2) Internal consistency
3) Kuder Richardson
4) Reader or marker
5) Split-Halves
6) Test-Retest

Replication refers to the building of as many features into a test as possible of the situations being tested for. It can be best assured by a careful pre-specification of all the salient features in order to include them in the test itself.

Reporting plays an important role in the evaluation and monitoring process, and without which continuous evaluation/monitoring may not be possible. The main purpose of reporting is therefore to provide timely, regular and dependent information about the functioning of the programme, to the implementing authorities in order to get the progress in the right direction.

Representative Sample refers to a sample which is representative of a population with respect to the characteristics important for the purposes under investigation. [Also see, Sample].

Responding is one of the affective characteristics. It is the second among the five hierarchical levels of the non-scholastic aspects of the learners' growth (under affective domain) as stated in the TEO. 'Responding' means 'response' whereby learner does more than just note that the stimuli. It may happen in any of the following forms, viz., complying with, following, commanding, approving, volunteering and so on.

Response in the area of evaluation refers to the reply or answer given by an examinee to a question. It may be either in the form of a word phrase, sentence or an essay. Also, it may either be oral or written.

Response Count refers to the counting of responses to the items for an objective test. It indicates the frequency with which one or more of the answer options were chosen by examinees in a particular group. It is a time consuming part of the item/test analysis in a particular group.

Response Cues are the cues that are to be avoided while preparing multiple choice items. Students who have prior exposure to certain types of questions/examinations may have a chance of developing 'test-wiseness', i.e., capable of selecting the correct option independently of any knowledge of the correct field being tested, simply on observing the response cues provided in the option itself. Therefore, care should be taken to avoid providing such a scope to the testees/examinees.

Response Validity is an another type of non-empirical validity. This is intended to describe the extent to which the examinees are responding as per the examiners/test developers. If the examinees respond in a haphazard manner their obtained scores may not represent their actual ability. If both the instructions and the test format are not clear and unfamiliar to the students, their responses may not reflect their true ability and in this way the test may be said to lack response validity.

Restricted Book Examination is a modified form of open-book examination. In this, a restriction is imposed on the books that can be used during the examination. [Also see, Open Book Examination].

Result is a phenomenon that follows an action or examination. Declaration of results after the final examinations/selection tests, etc., is quite common in educational institutions. In general evaluation, it obviously refers to the conclusion (i.e., consequence of the data obtained from investigation) leading to the value judgement.

Retention Test is designed to measure the students' ability to retain the language acquired. Such test is usually designed for L2 learners. Retention test is often considered as diagnostic because the retention ability has not yet been proved directly proportionate to comprehension. It would be more useful if the retention test items are arranged in the order of increasing difficulty.

Review implies examining something, making a judgement and putting it usually in the written form it is somewhat a less formal approach than criticism in referring to literary works. Review is distinctly different from evaluation also. It is not as extensive and exhaustive as the evaluation. As Rastogi, et al [1970, p.44] puts it, review consists of only checking up a few observable points mostly related to physical aspect and giving global impression about the quality and suitability of the academic content. A reviewer usually goes through the book and starts giving his reactions on the points listed in the review sheet.

Right Associates Procedure is a very commonly used procedure in the study of learning and retention. Item (usually verbal) are presented in pairs for learning, then the first of each pair is presented for a brief time and the learner endeavours to reproduce the second. The score is the number of successes or of retained numbers.