7/29/2019 239120.PDF

1/4

Table 1: General Test Interpretive DataUsed on Score Reports

(Based on the performance of all examinees who tested between

October 1, 1995, and September 30, 1998)

Percent of Examinees Scoring Lowerthan Selected Scaled Scores

Scaled VERBAL QUANTITATIVE ANALYTICALScore ABILITY ABILITY ABILITY

800 99 99 99700 97 78 86600 85 57 61500 60 34 34400 27 14 14300 4 2 3

Mean 472 563 547StandardDeviation 113 140 131

Number ofExaminees* 1,115,839 1,115,247 1,110,516PercentWomen 57 57 57PercentMen 42 42 42

Reliability .91 .92 .90

Standard Error of Measurement (SEM):

IndividualScores 33 37 41

ScoreDifferences 47 53 58

* Examinees who earned No Score on a measure are not included in Number ofExaminees for that measure.

INTERPRETING YOUR GRE SCORES 1999-2000

Guidelines for the use of GRE scores, which are included in the 1999-2000 GRE Guide, summarize the considerations for appropriate useof GRE test scores and encourage institutions to use GRE scoresappropriately.

This publication is intended to assist you in interpreting your GraduateRecord Examinations (GRE) test scores. Institutions and fellowshipsponsors receiving your scores are sent the interpretive information youreceive, as well as additional data. These data are revised annually andpublished in the GRE Guide to the Use of Scores.

General Test Score Interpretive Information

The range of scores for each General Test measure is 200 to 800. Ifno answers are given for a measure, an NS (no score) will be reportedfor that measure.

Scores from the different General Test measures should not becompared because each measure is scaled separately.

Research indicates that scores obtained from the computer-basedGeneral Test are comparable to scores from the paper-based GeneralTest. Thus, the mode of testing, which is indicated on examinee scorereports, is not indicated on score reports sent to institutions.

Table 1 shows General Test interpretive data, including the percentileranks, of selected verbal, quantitative, and analytical ability scoresbased on the scores of all examinees who took the General Test betweenOctober 1, 1995, and September 30, 1998. (Definitions of the statisticalterms are presented in the glossary on the back page.) For example, ifyou earned a quantitative score of 600, then 57 percent of the examinees

Table 2: General Test Mean Scores Classified byBroad Intended Graduate Major Field

(Based on the performance of seniors and nonenrolled college graduates*who tested between October 1, 1995, and September 30, 1998)

Broad Intended Graduate Number of Verbal Quantitative Analytical

Major Field Examinees Ability Ability Ability

Life Sciences 118,705 464 552 559

Physical Sciences 38,413 496 668 612

Engineering 48,880 468 695 596

Social Sciences 91,409 481 531 555

Humanities 45,681 537 538 579

Education 36,260 445 507 533Business 6,842 439 538 529

*Limited to those who earned their college degrees up to two years prior to the test date. Notethat this table does not include summary information on the approximately 50,200 examineeswhose response to the department code question was classified as other, the approximately53,600 examinees whose response was invalid (misgrids, blanks, etc.), or the approximately53,600 examinees whose response was Undecided. Most of the standard deviations of thescore distributions represented by the means in this table are between 90 and 125.

earned quantitative scores lower than your score. The percentile ranksfor your General Test scores are printed on your score report. Note thatalthough a given score represents the same level of ability regardless ofwhen the score was earned, its percentile rank may vary, depending on

the scores of the group with which it is compared.

To compare your scores with those of other examinees intending tomajor in similar fields, refer to Table 2. In this table, the number ofexaminees and mean scores are provided for the seven Broad IntendedGraduate Major Fields. Find the group with the intended graduatemajor field most similar to yours, and compare your scores with thescores reported for that group.

For more detailed interpretive data, see theGRE Guide on the GRE Web site at www.gre.org.

Copyright 1999 by Educational Testing Service. All rights reserved.

The Graduate Record Examinations Board and Educational Testing Service are dedicated to the principle of equal opportunity, and their

programs, services, and employment policies are guided by that principle.

EDUCATIONAL TESTING SERVICE, ETS, the ETS logo, GRADUATE RECORD EXAMINATIONS, GRE, TOEFL, TSE, and TWE

are registered trademarks of Educational Testing Service. The modernized ETS logo is a trademark of Educational Testing Service.

This publication can be downloaded from the GRE Web site.

For the paper-based General Test, your score is derived from thenumber of questions you answered correctly. For the computer-basedGeneral Test, your score reflects the number of questions youanswered as well as your performance on those questions. Thefactors that influence which questions you will be presented include(1) the statistical characteristics of those questions already answered(including the difficulty level), (2) the required variety of questiontypes, and (3) appropriate content coverage.

General Test Statistical Tables

7/29/2019 239120.PDF

2/4

Table 3: Subject Test Total Score Interpretive Data Used on Score Reports

(Based on the performance of all examinees who tested between October 1, 1995, and September 30, 1998)

Percent of Examinees Scoring Lower than Selected Scaled Scores

Scaled ScaledScore Score

980 99 83 96 980

900 99 91 98 99 71 89 900

800 99 93 75 85 83 93 55 77 800

700 93 74 57 53 54 78 94 99 96 37 99 58 91 700

600 73 43 32 22 26 51 67 89 76 20 82 35 65 96 600500 41 14 7 4 7 18 21 54 41 6 47 12 29 82 500

400 11 2 1 1 10 10 14 5 46 400

300 1 1 5 300

Number of Number of

Examinees Examinees

Percent Percent

Women 47 58 38 18 27 19 41 28 63 31 53 18 73 37 Women

Percent PercentMen 53 42 62 81 73 81 59 72 37 69 47 82 27 63 Men

Mean 527 620 682 681 678 605 563 492 521 766 501 675 554 417 Mean

Standard StandardDeviation 108 114 139 104 118 117 80 79 99 166 94 154 99 87 Deviation

Reliability .94 .94 .94 .91 .94 .92 .94 .95 .96 .94 .95 .92 .95 .96 Reliability

Standard Error of Measurement (SEM): SEM:

Individual Individual

Scores 24 28 25 28 27 29 20 17 19 42 20 38 20 21 Scores

Score ScoreDifferences 34 40 36 40 38 41 29 24 27 60 28 54 29 30 Differences

*For additional data and interpretive information about subscores for these tests, see Table 4.

Subject Test Score Interpretive Information

The range of scores for each Subject Test is from 200 to 990,although the range for any particular Subject Test is usuallysmaller. The range of subscores is from 20 to 99.

With the exception of the Music Test, your Subject Test score isderived from the number of correct answers minus one-fourth ofthe number of incorrect answers. For the Music Test, your score isderived from answers to multiple-choice questions and free-response questions. The free-response questions are scored by a

panel of expert judges. GRE Subject Test scores should be compared only with other

scores on the same GRE Subject Test because scores are

calculated independently for each test. A score of 680 on theComputer Science Test, for example, is not equivalent to a 680on the Mathematics Test.

Scores on the same Subject Tests generally are directly compa-rable across years. A Chemistry Test score of 650 in 1997, forexample, should be considered equivalent to a Chemistry Testscore of 650 earned in 1999. The exception is for scores earned onthe Music Test before and after October 1, 1990.

The Revised Education and Revised Political Science Tests werediscontinued following the April 1998 administration.

The History and Sociology Tests will be discontinued after theApril 2000 administration.

Subject Test Statistical Tables

Biochem

istry,

Ce

llan

d

Mo

lecu

lar

Bio

logy

*

Biolo

gy

*

Chem

istry

Com

pu

ter

Sc

ience

Econ

om

ics

Engineering

*

Geo

logy

*

Histo

ry*

Litera

ture

inEng

lish

Math

ema

tics

Mus

ic*

Phys

ics

Psyc

ho

logy

*

Socio

logy

Table 3 shows Subject Test interpretive data, including the per-centile ranks of selected Subject Test scores based on the scoresof examinees who took the Subject Tests between October 1, 1995,and September 30, 1998. For example, if you earned an EconomicsTest score of 600, then 26 percent of the examinees earned Econom-ics Test scores lower than your score. The percentile rank for yourSubject Test total score is printed on your score report. Note thatalthough a given score represents the same level of ability regardlessof when the score was earned, its percentile rank may vary, depend-ing on the scores of the group with which it is compared.

Some Subject Tests also report two or three subscores in additionto a total score. Subscores indicate relative strengths and weaknesses

of preparation in subfield areas. Table 4 shows interpretive data forselected Subject Test subscores. The percentile ranks given in Table4 may be used for diagnostic interpretation of the total score. Forexample, an examinee who obtains a score of 500 on the GREHistory Test is likely to have subscores of 50, assuming the examineeis similarly able in the content areas measured by each subscore. Forthat examinee, scores much above or below 50 on a subscore wouldindicate strength or weakness in the content area associated withthat subscore. Note that these strengths or weaknesses may reflectthe amount of training that was targeted toward specific content areas.

10,

618

26,

752

11,

408

11,

165

4,

671

6,

158

2,

270

3,

415

16,

634

7,

563

3,

879

10,

231

44,

186

4,

581

7/29/2019 239120.PDF

3/4

Table 4: Subject Test Interpretive Data for Subscores(Based on the performance of all examinees who tested between October 1, 1995, and September 30, 1998)

Percent of Examinees Scoring Lower than Selected Scaled Scores

Biochemistry, Biology* Engineering Geology* History* Music* Psychology Cell and Molecular

Biology

Scaled ScaledScore Score

90 99 99 99 99 90

80 99 91 93 95 93 93 99 80

70 90 93 97 73 76 75 80 75 95 94 93 99 99 98 91 93 70

60 71 73 76 44 45 43 52 50 62 66 67 88 89 83 83 79 66 65 60

50 41 41 41 14 14 15 19 17 20 21 22 53 55 47 48 47 31 29 50

40 10 10 13 1 1 2 1 1 1 8 11 14 12 20 4 6 40

30 1 1 1 4 30

Number of Number ofExaminees 10,618 26,752 6,158 1,902 3,415 3,879 44,186 Examinees

Mean 53 53 52 62 62 61 60 61 57 57 57 49 49 50 50 49 55 55 Mean

Standard StandardDeviation 11 10 10 12 11 11 12 12 8 8 8 8 8 9 9 11 10 10 Deviation

Reliability .86 .84 .87 .88 .82 .89 .89 .84 .77 .80 .79 .92 .88 .92 .86 .88 .90 .89 Reliability

Standard Error of Measurement (SEM): SEM:

Individual IndividualScores 3.6 3.9 3.5 4.1 5.1 4.1 3.4 4.1 3.9 3.6 3.7 2.2 2.6 2.5 3.5 3.1 2.9 3.1 Scores

Score ScoreDifferences 5.1 5.5 5.0 5.8 7.2 5.7 4.9 5.8 5.5 5.1 5.3 3.0 3.7 3.6 4.9 4.4 4.1 4.3 Differences

Total Score Total ScoreMean 527 620 605 563 492 501 554 Mean

Standard StandardDeviation 108 114 117 80 79 94 99 Deviation

Ce

llularan

dMolecu

lar

Biology

Mo

lecu

lar

Biolog

y

an

dGene

tics

Bioc

hem

istry

Ce

llBiology

Organ

isma

lBiolo

gy

Eco

logyan

dEvo

lution

Eng

ineering

Ma

thema

tics

Usa

ge

Stra

tigrap

hyand

Se

dimen

tology

Struc

tura

lGeology

an

dTec

ton

ics

Minera

logyan

dP

etro

logy

European

History

Un

ite

dStates

His

tory

Historyan

dTheo

ry

Listen

ingan

dLitera

ture

Aura

lSkills

Experimen

talPsyc

ho

logy

Soc

ialPsyc

ho

log

y

* For interpretive information on Biology Test subscores earned prior to October 1, 1991, Geology Test subscores earned prior to October 1, 1996, History Test subscores earned prior toOctober 1, 1992, or Music Test subscores earned prior to October 1, 1990, contact the GRE Program. The interpretive information for the Geology Test subscores is preliminary. It is basedonly on examinees tested between October 1996 and April 1998 and should be used only for examinees tested after October 1, 1996.

Writing Assessment Scoring Information

The single number that constitutes the Writing Assessment score isan average of the ratings given to both writing tasks, which arethemselves averaged scores. Each response is given two indepen-dent ratings. If the two ratings are either identical or contiguouspoints on the rating scale, they are averaged. For instance, if aresponse receives two ratings of 5, the score is obviously 5. If aresponse receives one rating of 5 and a second rating of 6, the scoreis 5.5.

If the two ratings for a response differ by more than one point (forinstance, ratings of 3 and 5), the ratings are called discrepant, andanother evaluation is required. If the discrepant ratings are 3 and 5

and the third rater assigns a rating of 4, the average (4) is used.However, if the third rater assigns a rating of 5, the low score of 3 isdiscarded, and the final score assigned is 5. Because of ongoingrater training and monitoring, discrepant ratings are rare.

Once both essays by a test taker have been scored, the scores areaveraged to provide an overall Writing Assessment score. A testtaker who received a score of 5 on the Issue essay and a score of 4 onthe Argument essay would, for example, have a total score of 4.5.

The same Writing Assessment score can result from various combi-nations of ratings. For example, an overall score of 4.5 (rounded tothe nearest half-point interval) would be assigned to a test taker with

ratings of 5 and 6 on the first topic and 3 and 3 on the second topic,as well as to a test taker with ratings of 5 and 5 on the first topic and4 and 4 on the second topic.

The Writing Assessment is being introduced operationally in October1999. Percentile ranks will be reported when operational, large-scale normative data are available.

Considerations in Score Interpretation

GRE test scores are intended to provide a common measure forcomparing the developed abilities of applicants who come from avariety of colleges and universities with different standards andmethods of measuring success in their programs. The test scores

should be used to supplement the information provided in a personsapplication, undergraduate record, letters of recommendation, orother information provided by the applicant. Officials responsiblefor admission at each institution must determine the significance ofGRE scores for each applicant. Score recipients are advised to payparticular attention to the use of GRE scores earned by individuals

described below.

Examinees Who Test More Than OnceIndividuals are permitted to take GRE tests more than once. Repeat-ers of the General Test and Subject Tests, on average, show scoregains of about 20-30 points, but the possible significance of this

7/29/2019 239120.PDF

4/4

54053-01210 RR99M1100 Printed in U.S.A.

I.N. 239120

finding is tempered by the observation that repeaters are typically aself-selected group who believe that repeating a test will increasetheir scores.

Score recipients are cautioned not to view an increase in scoresnecessarily as a reflection of academic gain, especially over a shorttime period. Differences in Subject Test scores over longer timeperiods, however, may be significant in terms of academic gain dueto intervening learning experiences or of academic loss due toforgetting.

There are several ways in which graduate departments can judgemultiple scores for an individual (e.g., use average of all scores, use

most recent score, use highest score). Whatever approach is adopted,it should be used consistently with all applicants.

Minority ExamineesGRE tests provide measures of certain types of developed abilitiesand achievement, reflecting educational and cultural experiencesover a long period. Special care is required in interpreting the GREscores of students who may have had educational and culturalexperiences somewhat different from those of the traditional major-ity. Research to date indicates that GRE scores are valid predictorsof success in the first year of graduate school for all students.Available samples of minority students, however, have been verysmall. Information on test scores and minority groups is availableupon request.

International Examinees

Various factors complicate the interpretation of GRE scores forinternational students. Degree of fluency in English may affect testperformance. Moreover, since the GRE tests are developed forstudents who have been educated in the United States, cultural andeducational backgrounds of international students must be consid-ered, along with linguistic factors.

ETS offers three tests for testing international students mastery ofEnglish: (1) the paper-based and computer-based Test of English asa Foreign Language (TOEFL), (2) the Test of Written English(TWE), and (3) the Test of Spoken English (TSE). The TOEFLtest is designed to measure English listening and reading compre-hension skills. The productive skill of writing is measured in thecomputer-based TOEFL and by the TWE. Speaking proficiency ismeasured by the TSE. One or more of these tests are often requiredfor admission to graduate as well as undergraduate institutions. An

indicator of the general level of English proficiency of internationalstudents may be obtained by looking at their scores on the TOEFL,TWE, and TSE tests and should provide a basis for determining theextent to which English proficiency may be a factor in the GREscores earned by these students. Information regarding test scoresand international examinees is available upon request.

Examinees with DisabilitiesETS and the GRE Program are committed to serving test takers withdisabilities by providing services and reasonable accommodationsdeemed appropriate given the purpose of the tests. The GRE scoresof such examinees are reported with a footnote stating that the testwas taken with accommodations. No reference is made to the indi-viduals specific disability.

The nature and extent of an examinees disability may affect thedegree to which GRE scores represent his or her educational

achievement. Because there are so few persons taking GRE testswith accommodations and their circumstances vary so widely, ithas not been possible to provide special interpretive data for these

examinees. Therefore, graduate schools are advised to seriouslyconsider waiving GRE requirements for applicants with certaindisabilities.

Score Reporting

Policies pertaining to score reporting and use are periodicallyreviewed and revised by the GRE Board. The policies andprocedures explained in the 1999-2000 GRE Information and Reg-istration Bulletin are effective only for the 1999-2000 testing year

(October 1, 1999, to September 30, 2000) and supersede previouspolicies and procedures in previousBulletins. GRE score reportingis cumulative. Test scores earned during the five-year period priorto the 1999-2000 testing year will be included in your cumulativescore report.

Glossary of Statistical Terms

Mean an average obtained by adding all the scores from a groupof examinees and dividing the sum by the number of examinees inthe group.

Percentile rank the percent of examinees in a group whoobtained scores lower than a specified score.

Reliability a measure of the degree to which individual examin-ees would keep the same relative standing if he or she took the testmore than once. The reliability index ranges from zero to 1.00; areliability index of 1.00 indicates that there is no measurement errorin the test and therefore the test is perfectly reliable.

Scaled score the score that is derived from your performance ona test (e.g., GRE verbal; GRE Subject Test in Biology) and reportedto you and to your designated institutions. The score takes intoaccount the statistical characteristics, including the difficulty level,of the questions that were presented to you and whether youanswered those questions correctly or incorrectly. This proceduremakes it possible to compare the performance of examinees whowere administered different questions on the same test.

Standard deviation a measure of the extent to which examinees

scores on a test generally differ from one another.Standard error of measurement (SEM) an index of the varia-tion in scores to be expected because of imprecise measurement.For a group of examinees, it is an estimate of the average differencebetween observed scores and true scores (i.e., what examineesscores on a test would hypothetically be if there were no error in themeasurement process). Approximately 95 percent of examineeswill have obtained scores that are within a range extending from twostandard errors below to two standard errors above their true scores.

Standard error of measurement of score differences an indexused to determine whether the difference between two scoresis meaningful. Small differences in scores may be due to measure-ment error and not to true differences in the abilities of the examin-ees. This index incorporates the error of measurement in eachexaminees score being compared. Approximately 95 percent of

the observed differences between the obtained scores of examineeswho have the same true score will be less than two times the SEMof score differences.