**Provided Courtesy of Frank X. J. Homer, University of Scranton and NAPLA, The Northeastern Association of Pre-Law Advisors

The LSAT is a standardized test required of all applicants to all ABA approved schools. The test is offered at numerous test sites throughout the Unites States and abroad four times each year in June, early fall (either late September or early October), December, and February. The June test is given on a Monday afternoon, while the others are on Saturday mornings - with special Monday administrations offered for Saturday Sabbath observers. Since its introduction in 1948, the LSAT has undergone several revisions in length, question types, and scoring scale. The current version was introduced June 1991 and has not been modified since. Inasmuch as the test development staff at Newtown are presently working on a long-term project to create a computerized version of the LSAT, further revisions in the current pencil-and-paper test are not likely in the immediate future.

The present LSAT is a five multiple-choice section, standardized aptitude test, followed by a thirty minute writing sample. Each of the five multiple-choice sections, containing a total of about 120-130 questions, is separately timed at thirty-five minutes apiece, with a brief fifteen minute break between the third and fourth sections and a shorter one between completion of the multiple-choice test and distribution of the writing sample. Taking the test requires three hours and twenty- five minutes, not including the breaks or the time needed for the distribution and collection of test materials, as well as other administrative test center procedures. When these are factored in, the time between when test takers arrive at a test center and when they leave is at least four and one-half or five hours.

There are three different question types:

Reading Comprehension:

This type consists of four sets, each containing a reading passage approximately 450 words in length followed by between five and eight questions for a total about twenty-six to twenty-eight questions in an entire section. Individual questions deal with the main idea, primary purpose, or tone of passages as a whole; the meaning or function of individual words and phrases; information or ideas either explicitly stated in or inferred from a passage; or the application of information in a passage to a new context.

Analytical Reasoning:

Often called Logic Games, this type consists of four sets, each containing a problem involving the grouping, assignment, or ordering of certain items, or possibly their spatial relationship. Following the explanation of each problem, including the conditions governing the particular arrangement, are between five and seven questions for a total of approximately twenty-four in the entire section. Individual questions are either based solely upon the general problem and specific conditions given initially which apply to all questions in the set, or involve adding to these a new condition that is only relevant to the answering of that one question.

Logical Thinking:

This type consists of approximately twenty-four to twenty-six questions that are not grouped into sets, except for a few where two questions are based upon a common stimulus. Each question begins with an argument contained in either a few sentences, a very brief paragraph, or a short piece of dialogue. Individual questions deal with identifying the point of an argument, assumptions or premises upon which it is based, inferences that follow from the premises or evidence given, the overall structure of the reasoning used in an argument, errors or fallacies contained in an argument, the application of the principle in an argument a new context, or whether an additional piece of evidence strengthens or weakens an argument.

Note well that 60% of the questions on the test are of the analytic reasoning and logical thinking type. To prepare for these, take Critical Thinking and Logic courses.

There is no fixed sequence in which sections of the different question types are presented; the order in which they appear in test booklets varies from one edition of the test to another. Likewise, while individual questions do vary in terms of difficulty level, the order in which they appear in the test booklet is not necessarily determined by this. Test takers should not assume that the next question, or set of questions, is going to be more difficult than the one on which they are working at any given point in the test.

In accordance with the legal requirements established by certain states, most editions of the LSAT are disclosed following their administration. Those taking these tests receive with their score a copy of all scored test questions and a comparison of their responses to the correct or "credited" answers. Presently, all tests administered in February, as well as the special Monday administrations for Saturday Sabbath observers in the early fall and December, are not disclosed. Those taking these tests receive nothing other than their scores. All disclosed tests, from June 1991 on, are available for purchase by prospective test takers and represent the single most useful preparation material. Disclosed questions will not appear on any subsequent LSAT.

Only four of the five multiple-choice sections are scored. They consist of one section of Reading Comprehension, one section of Analytical Reasoning, and two sections of Logical Reasoning questions. There is one unscored section which can be of anyone of the three question types. The placement of the unscored section is completely random, varying from one edition of the test to another.

Since there is no way a test taker can identify the non-scored section while taking the test, every section must be approached with the assumption that it will be scored. When tests are disclosed, only the scored sections are made public. The unscored sections are used as a means of pretesting individual questions and preequating new test forms.

An LSAT score is a three-digit number ranging from 120 to 180, determined by the total number of correct answers to the approximately 99-101 questions contained in the four scored sections. The total number of correct answers on all of the four scored sections represents a test taker's raw score. This raw score is then plugged into a conversion chart to determine what score on the 120 to 180 corresponds to that specific number of correct answers on that specific edition of the test.

Each edition of the test has its own unique conversion chart that is the product of intricate calculations on the part of the Test Development & Research staff in Newtown and the external test writing service which produces the LSAT under contract for Law Services. Using data compiled from the pretesting of each of the scored questions, the conversion chart is designed to ensure that scores on the 120-180 scale are absolutely equatable from one test administration to another.

While every conversion chart is different, and is constructed to take into account variations in the overall difficulty level between one test edition and another, they all have a somewhat similar appearance. For every edition of the test there is a threshold number of correct answers, usually somewhere between fifteen and seventeen, that is required before a test taker's score begins to rise above 120. Generally, once test takers get beyond this initial threshold, every two additional correct answers will add roughly three points to their score. Reproduced below is a sample conversion chart used for the October 1993 LSAT to convert raw scores (the number of correct answers) to a score on the 120-180 scale. On all conversion charts, there are several points at which a difference of just one correct answer does not change the score. On the October 1993 test, for example, those with either seventy-five or seventy-six correct answers out of the total of 101 scored questions received the same score of 163. Likewise, there are usually one or two points where there is no raw score corresponding to a particular scaled score; in October 1993, no test taker could receive a 122.

Differences between conversion charts are not all that great, but they are designed to take into account variations in overall difficulty level among editions of the test. Thus, the fact that October 1993 test takers with seventy-five correct answers received a score of 163, while the same number of correct answers in the next test administered in February 1994 produced a score of 160 indicates that the October test had a somewhat higher degree of difficulty than the February test, where a score of 163 required either eighty or eighty.-one correct answers. The respective conversion charts ensure that test takers whose raw scores are somewhat lower (because of a higher difficulty level in the test they took) are not penalized and that their scaled score between 120-180 is absolutely equatable with the same score received by others taking the test at different administrations.

Conversion Chart for the October 1993 LSAT

Raw Score	LSAT Score	Raw Score	LSAT Score	Raw Score	LSAT Score
98-101	180	68-69	159	35	138
97	179	66-67	158	33-34	137
96	178	64-65	157	32	136
95	177	63	156	31	135
94	176	61-62	155	29-30	134
93	175	59-60	154	28	133
92	174	58	153	27	132
90-91	173	56-57	152	26	131
89	172	54-55	151	25	130
88	171	53	150	24	129
86-87	170	51-52	149	23	128
85	169	49-50	148	22	127
83-84	168	48	147	21	126
82	167	46-47	146	20	125
80-81	166	45	145	19	124
78-79	165	43-44	144	18	123
77	164	42	143	-- *	122
75-76	163	40-41	142	17	121
73-74	162	39	141	0-16	120
71-72	161	37-38	140	* No raw score will produce a 122 on this test.
70	160	36	139
Copyright LSAC Reproduced with Permission.

Throughout the test, there is a definite variation in difficulty level among individual question types, but this has no relevance to the test taker's raw score which is based solely and entirely on the total of correct answers. For test takers, all that matters is increasing the number of correct answers. Each right answer has the same weight in the raw score, regardless of how more or less difficult the question may be, and increasing the raw score is only way test takers can in any way influence their LSAT score.

The scoring of the LSAT, unlike that of some other standardized tests, does not make any distinction between an answer that is incorrect on the answer sheet and one where an answer is simply not given. The raw score is nothing other than the total of correct answers. There is therefore absolutely no penalty for guessing: don't leave anything blank!

Whenever an LSAT score is reported, it is also accompanied by a percentile ranking. These are not based on how that score compares to the others achieved on that particular administration of the test, but rather on the distribution of scores among all test takers on all administrations of the test over the most recent three full testing years. The percentile ranking indicates what percentage of all of those test takers had scores below the reported score.

Shown below is the distribution of scores for all test takers between June 1996 and February 1999 upon which the percentiles for scores during the 1999-2000 year are based. Since the present test format was introduced in 1991, there has been hardly any difference in the distribution of scores across the 120-180 scale from one testing year to another; a 156, for instance, has consistently been a score that approximately 70% of the test takers fell below. Thus, percentile rankings are not likely to vary from one year to the next under the current test format.

LSAT Score Percentiles for 1996-1999

Scaled Score	Percent Scoring below (1996-99)		Scaled Score	Percent Scoring below (1996-99)		Scaled Score	Percent Scoring below (1996-99)
180	99.9		159	80.7		139	13.0
179	99.9		158	77.7		138	11.0
178	99.9		157	74.3		137	9.4
177	99.8		156	71.2		136	7.8
176	99.7		155	67.6		135	6.6
175	99.6		154	63.8		134	5.5
174	99.4		153	60.2		133	4.5
173	99.2		152	55.8		132	3.7
172	99.0		151	51.9		131	3.0
171	98.6		150	48.0		130	2.4
170	98.1		149	44.1		129	2.0
169	97.5		148	40.0		128	1.6
168	96.7		147	36.8		127	1.3
167	68.9		146	32.9		126	1.0
166	94.7		145	29.5		125	0.8
165	93.4		144	26.3		124	0.7
164	91.8		143	23.0		123	0.6
163	90.0		142	20.6		122	0.5
162	88.3		141	17.7		121	0.4
161	85.9		140	15.1		120	0.0
160	83.4
Copyright LSAC Reproduced with Permission

The distributions of scores on the current LSAT show that roughly 70% of all scores consistently fall somewhere between 142, below which are the bottom 20%, and 163, above which are the top 10%. Thus, within that twenty-one point range, what may seem to be relatively modest score differences do actually represent rather significant increases in percentile rankings. For example, a test taker with a 152 is only among the top 46% of all test takers, while someone with a 156 would rise to the top 30%.

The writing sample is administered separately, following completion of the five-section test. The sample consists of a stimulus describing a situation in which a person or group is deciding between two alternative courses of action. The criteria upon which the decision is to be made and the two alternatives are explained, and the test taker is asked to write in support of one of the two alternatives. Since in every stimulus both of the alternative choices are quite reasonable options, the test taker's choice of which option to favor is of little concern; what matters is how well the test taker can write on behalf of whichever option is being supported. The reading of the stimulus and writing of the sample must all be done within a thirty-minute time limit. Unlike questions on disclosed editions of the test, previously used writing sample stimuli may reappear.

The test taker is restricted to a relatively compact space in the writing sample booklet within which the sample must be placed, and the thirty minutes provide ample time for test takers to organize their thoughts before actually writing their sample. Scratch paper and a black ink pen are provided. The booklet contains a sheet of carbonless-paper which produces a copy of the sample for the test taker.

The writing sample is unscored; it is simply duplicated and a copy sent to law schools along with an applicant's LSAT score. The way in which the sample is used by law schools varies considerably; however, for the most part, its role in the admissions process is minimal compared to the substantial weight given to the LSAT score. Given the fact that the sample is done immediately following completion of a physically and intellectually demanding test of several hours' duration, law schools hardly expect scintillating prose. Test takers are well advised to concentrate on avoiding glaring grammatical or spelling errors and making sure that what they write is at least clear and to the point.

Most directly, the LSAT is designed to measure a number of intellectual abilities found to be critical in legal study, in particular:

Verbal Facility:

While most obviously the focus of the Reading Comprehension type of questions, test takers are being tested throughout the entire test on their ability to understand the meaning of words and phrases, both in terms of explicit meaning and inferences that follow from their connotation; to recognize grammatical structure, to be sensitive to how tone and attitude are reflected in language, and to apply information presented in one context to another.

Analytical Skills:

The LSAT requires test takers to be able to break down and to organize information in a systematic fashion. This is particularly true in the case of the Analytical Reasoning question type where test takers must first get a sense of the overall shape of a problem, and then see how individual pieces of information fit into it, which often involves construction of a rudimentary chart or diagram to aid in answering the questions that follow. Analytical skills, however, relate to other parts of the LSAT, as well; in the reading comprehension passages, for instance, understanding a passage's organizational structure is frequently a key to answering questions. Throughout the test, test takers need to be able to sift through information presented in a variety of formats, and to sort out what is and is not relevant to the answering of particular questions.

In addition to the two scored sections of logical Reasoning questions, which deal explicitly with the logic of given arguments, the entire LSAT involves an even more fundamental logic implicit in the fact that, among the five answer options for every question in every section (A through E), there is always to be found the one correct response and four others that are incorrect. Thus, whenever one particular answer option can be identified as necessarily correct, the others must necessarily be incorrect. Conversely, whenever four of the five are found to be unacceptable, the remaining choice must necessarily be the correct answer. For most of the test, in the Reading Comprehension and logical Reasoning sections, identification of the correct answer cannot be done without taking all of the five answers into consideration. Most of the questions in those sections deal with what might be called relative correctness, where, for example, a question on a reading passage might ask "which of the following best represents the passage's main point?" In such cases, what essentially makes the correct answer correct is that it is better than all of the other choices on the menu. Questions in the Analytical Reasoning sections, however, involve absolute correctness in that each answer option, independently of the others, is either absolutely correct or absolutely incorrect. In such cases, once a test taker identifies one correct response, all the others, whether looked at or not, can be discarded. Analytical Reasoning questions, in particular, also require test takers to understand the distinction between logical possibility and logical necessity, or, to put it another way, the distinction between "can" and "must."

Beyond these intellectual abilities, the LSAT is also designed to test several other capabilities:

Stability Under Pressure:

As much a part of the LSAT as the questions is the time factor built into the test, which in two ways creates a pressurized environment for the test taker. First of all, the very length of the test requires test takers to work continuously for almost three hours with only one brief break; thus, a certain amount of physical stamina comes into play. Secondly, and a source of even more pressure, is the thirty-five minute time limit for every section of the test. Most test takers at some point during the test find themselves running out of time and not able to finish completely one or more sections, in which case they need to suppress the temptation to panic and be able to concentrate on getting as many correct answers as possible.

Tolerance of Ambiguity:

As noted above, in both the Reading Comprehension and Logical Reasoning question types, the correctness of the right answer is relative to the other options in the test booklet, where the test taker is asked to identify an answer that is somehow superior to the other options available (e.g., "which of the following best summarizes the author's viewpoint?" or "which of the following most strengthens/weakens the argument above?") often, in such cases, test takers can conceive of answers that are more clearly correct than any of the five options presented in the test booklet. Test takers need to understand that if there is a discernable flaw in four of the options presented, the remaining choice is the correct answer, even if it is not a perfect response.

Ability to Deal with the Unfamiliar:

Not only is the overall format of the LSAT a rather artificial test-taking situation in which test takers must be able to function, but also many of the questions will almost certainly present all test takers, regardless of their academic background, with unfamiliar material. This is particularly true of the reading passages in the Reading Comprehension section which involve topics from virtually any field in the natural sciences, social sciences, or humanities. At least one of the four passages is almost always technical or scientific in nature, a fact which not infrequently causes humanities majors some consternation. Test takers need to understand that all the information needed to answer questions is given to them in the test. The test is carefully designed not to give any advantage to test takers who majored in any one particular undergraduate concentration. In fact, in cases where a reading passage happens to deal with a topic a test taker has studied, care must be taken not to choose answers on the basis of knowledge that goes beyond information presented in the test booklet.

While the LSAT tests a variety of skills and aptitudes, it is used to predict only one thing, first-year grades in law school. Almost all law schools participate in validity studies that are regularly conducted by the Law School Admission Council in which students' first-year law school grades are compared with their LSAT scores. While there is not perfect correlation between the way students rank by LSAT score and their ranking by first-year law grades, there is enough correlation to show that the LSAT does help in predicting first-year performance. Correlation between the two is measured in terms of a coefficient between 1.00 (an exact correlation) and zero (no correlation at all other than pure coincidence). While the level of correlation at individual law schools varies greatly, the median coefficient based on the most recent validity studies among 171 different schools was .42, and, when LSAT scores were combined with undergraduate grades, the correlation between them and first-year grades rose to .50. Thus while not a perfect predictor, there is documented correlation between it and first-year law school performance. The LSAC issues guidelines on use of the LSAT to law schools which point out the limitations of the LSAT as a predictor. Law schools are told not to rely solely upon the test as a criterion for admission, and are encouraged to assess the predictive utility of the LSAT on the basis of validity studies involving the actual performance of students at their own institution.

The reliability and precision of the LSAT is also measured by two other statistics, a reliability coefficient and a standard error of measurement. The relability coeffiecient indicates how consistent a test taker's performance would be if the same test were to be taken repeatedly. The reliability coefficient for every LSAT administered over the past several years has been above .90, indicating a high degree of reliability. The standard error of measurement (SEM) is a more useful statistic for both the interpretation on individual scores and for comparing the scores among different test takers. In terms of individual scores, the SEM for all recent LSATs has ranged from 2.54 to 2.73, approximately two and one-half points on the scoring scale. A test taker's actual score will be within one SEM of his or her "true score" approximately 68% of the time, and within two SEMs approximately 95 % of the time. In comparing the scores of two different test takers, the SEM is about 1.4 times greater than it is for the interpretation of individual scores, or just under four points on the scoring scale. Small score differences among different test takers, therefore, may not represent real differences in abilities, as much as measurement error, and the LSAC cautionary guidelines advise law schools not to put undue weight on small score differences among applicants.

The skills which the LSAT tests are ultimately the product of years of both formal and informal education. Prospective test takers cannot expect to remedy substantive deficiencies in their educational background going back many years in just a few months prior to taking the LSAT. Preparation for the LSAT essentially involves familiarization with the format of the test so that the test takers can use those skills which they already possess to their fullest possible extent. Repeated use of authentic practice tests under simulated test conditions is at the heart of effective test preparation.

There are several ways in which test takers can prepare for the test, including entirely on their own. Every disclosed test since 1991 is available for purchase from Law Services, either as an individual PrepTest or as part of TriplePrep books containing three authentic past LSATs. One is also provided absolutely free of charge in the LSAT/LSDAS Information and Registration Book. These past tests represent the single best preparation material available any where at any price. Since preparation on one's own requires a certain amount of self-discipline, many prospective test takers seek assistance and are often drawn to commercial test preparation services which advertise extensively at campuses throughout the country. There are a good number of such commercial organizations operating either nationally or regionally. All seek to persuade students that using their services will give them an advantage over other test takers and somehow guarantee them higher scores; all for a fee that can be as high as $850.

Conscientious pre-law advisors have an obligation to inform their pre-law candidates that, however helpful commercial test preparers might be, their services do not, in fact, offer anything that prospective test takers cannot, given time and motivation, do on their own. Advisors should seek to insure that the misleading advertising put out by commercial groups does not lead their advisees to believe that commercial test preparers possess some secret, inside information to which they and they alone have access. A growing number of advisors at undergraduate institutions provide an alternative to commercials by means of low-cost, on-campus workshops. Advisors interested in organizing such alternatives can make contact with those with experience in this area through the regional pre-law advisor associations, just one of many benefits to be derived from membership in one's regional APLA.

Whatever method of preparation one chooses, the following elements are of importance:

• Repeated timed practice using authentic past LSATs. If at all possible, four scored sections should be done in one sitting, or at least one or more full thirty-five minute sections. After each practice test, areas of weakness (and strength) should be reviewed. Doing questions at random, flipping back and forth to the answers in the back of the test booklet, is not only of no help, but can be positively harmful in that it does not properly simulate the actual test environment.
  
  

• Understanding the structure of the test and how it is scored in order to make the most productive use of the time available, i.e., to get the highest number of correct responses possible within the thirty-five minutes allotted to each section. Test takers who, after repeated practice, discover that they consistently have difficulty finishing sections of a particular question type, might realize that they actually get more right answers working on just three reading passages or analytical reasoning problems and simply guessing on the remaining passage or problem, rather than rushing to complete the entire section. (Answering seventy-five percent of all scored questions correctly will always produce a fairly competitive score.) Such a defensive strategy is not, of course, appropriate to all test takers; however, it is just one example of the importance of time management in taking the LSAT.
  
  

• Managing the answer sheet. Like all standardized tests, the LSAT uses a machine-scored answer sheet to record test taker responses. Part of effective time management is being able to transfer answers quickly and accurately to the answer sheet, making sure not to lose one's place on the sheet, or to forget to record guesses on any questions left unanswered as the thirty five minutes are about to run out.
  
  

• Recognizing at what point, after repeated use of practice tests, one's score reaches a point where it does not seem to rise from one test to another. There is a point at which test preparation can become obsessive; test takers need to understand that, once they are able to achieve a score that matches the upper range of their actual abilities, they have done all that any test preparation method can provide.
  
  

THE ABOVE TEXT IS TAKEN FROM THE FIRST EDITION OF THE NAPLA HANDBOOK FOR PRE-LA W ADVISORS (1996) AND IS COPYRIGHTED BY NAPLA. IT CANNOT BE REPRODUCED WITHOUT THE PERMISSION OF NAPLA. A SECOND EDITION OF THE HANDBOOK WILL BE AVAILABLE IN THE FALL OF 1999.