TO ALL THOSE WHO DID NOT GET THE SCORE THEY PREDICTED... HERE'S SOME INSIGHT

[tiptoe]

I too, did not get an lsat score reflective of my practice exams and was pretty bummed out... It was actually six points lower, then I spoke to someone and they helped put it in perspective.

First, February exams tend to more difficult, for one, alot of law schools and the Lsac are aware that February is the month for a lot of retakers who are not satisified with their October and December scores.  Retaking the LSAT in February gives you an unfair advantage because you had received experience under real testing conditions.  You can't get anymore simulated than that.  Therefore, if you thought taking February's exam is going to dramatically improve your previous score based on simulated practice exams, you are mistaken because the actual exam is harder.  There is a reason that the February tests are undisclosed, so people can not practice on harder exams.

Second, do not focus on your actual numerical score, focus on your overall ranking percentile.  On one exam, i.e. the ranking percentile of a 160 may be equivalent to another exam in which the correlating score is a 165.  That is what the schools look at to keep the actual numerical value of the score in it's proper perspective.

Third, if you look at LSN, you will sometimes see two people with an identical GPA, and LSAT score, one gets in, one doesn't, this is also assuming that the soft factors are equal. What matters is where that GPA came from, what was your ranking at your institution, is your undergrad competitive or did half the school graduate with a 3.5 gpa?  In addition, they look at when you took the LSAT and what your overall ranking percentage was when you took that test.  June tests seem to be not as difficult as February's. 

For example: You may see a person with a 168 who actually ranked in the 95th percentile get in and see another person who got 168 based on another test and ranked in the 90th percentile not get in. 

Once again, this was a brutal test, and I would not focus on the numerical value of the LSAT score but on your overall ranking percentile. So in other words your 160 February, may actually be equivalant to a 165 in June.

I hope this keeps it all in perspective, so do not beat yourself up, and good luck to all in the application process.

[Jogging_Snail]

Certainly a little perspective goes a long way helping us understand the mysterious LSAT better. However I do not agree on some parts of the OP.

First I don't think Februray exams are harder. I think probably Feb exams have a harder conversion table. For example, a 90/100 raw score may get you 173 in October but only 170 in Feb.

Besides, I don't think it is quite likely that two scores that are 5 point apart can have similar percentiles attached. The purpose of the conversion from raw score (how many questions were correctly answered) to the actual score (120-180) is to reduce the discrepancy between different seatings of the exam. I looked at some historical percentile tables and I found the percentiles for the same score have always been consistent.

[cascagrossa]

" For example, a 90/100 raw score may get you 173 in October but only 170 in Feb"

actually, it would have gotten you a 170 in oct.

the op's first 2 points are just wrong.  the feb test isnt any harder than any other test.  ive taken 3 now disclosed feb tests and there was no difference.

a 168 is NEVER 90%.  i dont even think percentiles change from test to test, and if they do it is by a very small amount.

[JDubs]

" For example, a 90/100 raw score may get you 173 in October but only 170 in Feb"

actually, it would have gotten you a 170 in oct.

the op's first 2 points are just wrong.  the feb test isnt any harder than any other test.  ive taken 3 now disclosed feb tests and there was no difference.

a 168 is NEVER 90%.  i dont even think percentiles change from test to test, and if they do it is by a very small amount.

I believe all tests are are adjusted so that the score and percentiles are the same for each test.  the only thing that changes is the # of correct answers needed to get a specific score.  But a 164 is always around the 91st percentile. 

[tiptoe]

168, 90th percentile was just an example, please do not take it literally...I do not know what the percentile ranking is for a 168, in June, October, December and February.  If someone could get this data that would be extremely informative.

My main point is that schools do not just look at your numerical score, they look at your overall ranking percentile.  Do not get caught up because your score is lower than a simulated exam that was not a February exam.  If your ranking percentile jumped dramatically that is what's important, that is why most schools have LSAT RANGES which can vary from five to six points.

Cascagrossa:  You keep taking February tests, the point I am trying to make is between someone who took an actual exam in June or some other month versus February, not a simulated exam but a real test in order to compare your percentile rankings.  How about we start here.  Question: Did you repeatedly score higher in your simulated practice exams, and if so, by how much?

[cascagrossa]

first of all, schools probably do look at your numerical score most because that is what is reported to us news.

percentile rankings are always the same, or very very close. a 160 will NEVER have the same percentile ranking as a 165.

to quote powerscore.com: "The percentile for each score is calculated by using test data from the last three years, not from any individual LSAT (note that the percentile table above is for 2001-2002, yet it only includes LSAT information up to the February 2001 LSAT). Using the three-year pool of information provides Law Services with a stable and accurate percentile for each score. Otherwise percentiles could vary significantly from test to test as different groups of test takers performed better or worse. Historical analysis shows that percentiles do change from year to year, but only by minute amounts. Since percentiles are not calculated on a per test basis, each test taker does not compete against the other students taking the same LSAT. Instead, each test taker competes against the students from the three previous years. In fact, because of question pre-testing through the use of experimental sections, the conversion chart for each LSAT is supposedly set before the test is administered. Only minor adjustments are then made to normalize the test."

and yes, i repeatedly scored higher in my practice exams by 3-5 points on average than i did on the real thing.  i think my real score was fine because of the random factors/stresses of the real thing.  what does that have to do with anything?

[Jogging_Snail]

Question: Did you repeatedly score higher in your simulated practice exams, and if so, by how much?

You can call me lucky but my actual score was almost exactly the same as (0.3 more than) my practice average. And my practice score has been very consistent (within a 3 point range). Please don't take this in the wrong way, personally I think Feb 2005 exam is very easy on an absolute scale. Probably one needs 99/99 to get a 180.

And I always wonder what LSAC will do if hundreds of people somehow get perfect raw score in one seating. Will LSAC give all of them 180 or they draw lots? Or they throw out the whole exam and make everyone retake it?

[A.J]

There is only a VERY small variance in percentile ranking per a given converted score on any exam.  For all intents and purposes the scaled ranking is fixed i.e. 167=95.7% +/- .2 or something like that. 

[alexansergej]

Excerpt from the lsac "IRR additional information document" that I got with the results of my feb 2005 test. "...To aid you in the interpretation of your LSAT score, we have included in this portion of your report a percent-below value for each test score. The values reported in Table 1 are for scores earned since June 1, 1991, and show the percentages of test scores below each score for the 2001–02, 2002–03, and 2003–04 testing years...

Table 1 Scaled for LSAT Score 2001–2004

180 99.9
179 99.9
178 99.9
177 99.8
176 99.7
175 99.6
174 99.5
173 99.3
172 99.0
171 98.5
170 98.1
169 97.5
168 96.7
167 95.7
166 94.6
165 93.2
164 91.4
163 89.7
162 87.3
161 84.9
160 82.2
159 79.1
158 76.5
157 72.6
156 68.7
155 65.7
154 61.5
153 57.3
152 53.2
151 49.1
150 44.9
149 41.0
148 37.0
147 33.4
146 29.6
145 26.4
144 23.3
143 20.2
142 17.7
141 15.2
140 12.9
139 10.9
138 9.2
137 7.8
136 6.5
135 5.3
134 4.4
133 3.5
132 2.9 "

I scored about 5 points below my average as well. How I wish it could be true that my 160 on the Feb 2005 could be equal to a 165 of the Dec 2004, but it "just ain't so." 

[alexansergej]

I would like to offer an alternate explanation for why many of us have scored much lower on the feb 05 lsat than they have scored, on average,  on diags, and I would like some feed back as to whether it seems plausible.

 
 As we know each test has an experimental section.  Presumably, lsac compares one's performance on the questions from the experimental sections to one's general performance on the scored sections.  From this they would rank the difficulty of each experimental question. 
 Since about 2001 there has been a very significant increase in the amount of people applying to law schools.  Along with this there has been an increase in the amount of people taking the lsat.  Presumably this has also brought a significant increase in the amount of people who do very well. 
 This increase in numbers of people taking the test has allowed lsac to chose more difficult questions. 

 To test this, I would like ask the people who scored several points below your average on the feb 05 test the following.  What time period were your diags from?

 If I am correct you'll find that the score you got on the feb 2005 is very close to the scores you got on the tests created after this increase in numbers of test takers.


 If lsac gauges difficulty based percentage of test takers rather than absolute numbers of test takers who get each experimental question right, then the increase in numbers of test takers would be absolutely irrelevent.  In that case, I would be interested in knowing if there was a  significant increase in percentage of test takers who prepared for the test vigourously. 

In any case it seems to me like the recent tests (after 2000) are simply harder then old tests.