LSAT C2

Ismael Torres-Pizarro RUA 16435, Colegiado 17810

“Don’t know much about LSAT in Spanish, but I know that, if the following is true, the LSAT in English should be”

Don’t know much geography,

don’t know much trigonometry,

don’t know much about algebra,

don’t know what a slide rule is for…

Sam Cooke

I love math. It has been said that God is the Prime Mathematician. My love of math is maybe the main reason I became an engineer. Math is very rational, you see. Now, maybe, because I have this irrational urge to stand before bullies, while in Jr High and HS, I got my fair share of free “face reconstructions.” Like the future Captain America, I did not know when to quit. Bad luck I did not get a shot of the Super Soldier serum; so, I did the next best thing: I became a lawyer.

That being said as an introduction, let me tell you why I picked this one fight. This is taken from the LSAC web regarding the Spanish LSAT[1]

El LSAT en español no es exactamente igual al LSAT en inglés, por lo que no se puede medir con la misma métrica de 120-180. Sin embargo, los resultados del LSAT en español están basados en la misma métrica estandarizada, funciona similarmente y sus resultados se pueden interpretar de la misma manera que el LSAT en inglés.

Los resultados del LSAT en español son reportados en una escala de 320-380. Esta escala tiene el mismo número de puntos que el LSAT en inglés (120-180), pero la relación de los resultados no es una estricta de uno a uno. Por ejemplo, mientras un resultado en el LSAT en español de 335 se puede ver como uno aproximadamente similar en un resultado de 135 en el LSAT en inglés, el resultado de 335 en el LSAT en español no se convierte directamente en un resultado de 135 en el LSAT en inglés, ni garantiza que el mismo estudiante lograría un resultado de 135 en el LSAT en inglés. Cualquier comparación debe hacerse con mucho cuidado y con el entendimiento que son solamente aproximaciones.

The LSAC is wrong. Luckily, they provided a way to correct themselves. Let me explain scales using the most famous (or convenient) ones: thermometers, and then, using that logic, provide you with the most likely correct interpretation of the LSAT in Spanish score. Then, if this is not the way the three Puerto Rican law schools interpret the LSAT in Spanish, why you should not take it but must take the LSAT in English instead.

There are several scales to measure temperature: the Fahrenheit scale (English units) and the Celsius scale (SI unit) based on the water’s freezing and boiling points, and the absolute scales based on thermodynamic principles (the zeroth points where there is not energy), the Rankine scale (English units) and the Kevin scale (SI unit). A drawing illustrates better:

Temperature Scales. From https://blog.beamex.com/temperature-units-and-temperature-unit-conversion.

Now, for instance, that simple figure shows, without any math involved, that the absolute zero in the Celsius scale is -273.15°C but that is equivalent to -459.67°F In the Fahrenheit scale which is equivalent to 0 K[2] in the Kevin scale and to 0° R in the Rankine one. We can select other examples, but I think the idea is crystal clear now. Let me present you with the scales for the LSAT English from LSAT vs candidate percentile[3].

Again, an example suffices to show my point. Let’s says you got an LSAT in Spanish score of 345 that is also reported by the LSAC as in the 94% percentile. The LSAC wrongly states that since their scale for the English version is from 120 to 180 and their scale for the Spanish test is 320 to 380; that is, the 345 Spanish score is roughly translated to a score of 145 in the English version (granted they are given just approximations, but it is not a good one and its effects are prejudicial, as we will see).

There are two possible positions here. The LSAC’s is akin to taking the thermal scales (let’s say the Celsius vs the Kelvin) as equivalent, but for a difference in the starting point. But, there is another position: that, in addition to an initial difference, a proportional difference must be considered (let’s compare it with the Celsius vs the Fahrenheit one).

For example, if we do as they suggest, a 100°C would simply translates to the Kelvin scale by just adding 273.15, reading 373.15 K (as you can see in the chart). Or vice versa, a Kelvin measurement of 373.13 translates to a Celsius scale of 100°C by subtracting 273.15 (in an analogous situation as just subtracting 200 from the LSAT in Spanish to have an approximate number in the LSAT in English).

Now, if we are in the other case, we need to adjust by a proportional constant before doing any subtraction or addition. In the case of Celsius vs Fahrenheit, that constant is 1.8. So, in a reading of 100°C, first, we adjust by multiplying by the 1.8 constant, giving us 180 and adding the starting point of 32, given us the correct reading in that scale of 212°F.

Then, which position is correct? LSAC states it is the first one (just subtract two hundred, the starting point), but their own data contradicts them.

Apart from their own disclosure admission, that by itself casts enough doubts on the simple procedure of just subtracting two hundred, the fact is LSAC also reports another scale: the equivalent percentile ranking of the candidate. And in a Rosetta stone sort of way, it reports it for both the Spanish and the English version.

A LSAT in Spanish candidate reporting a score of 345 is placed in the 94% percentile. That is, 94 people behind her and just 5 ahead of her. This is equivalent to a LSAT in English score of approximately 166.5[4] not just 145!!!! You can see this yourself in the illustrations. A LSAT in English score of 145 places the candidate way behind where she should be: 26.7 percentile. That is, her position is about the 28th and she has 72 people behind her in the line or 27persons ahead of her!!

In another example, a candidate took the LSAT and got 365 and a 98% percentile. Measuring with the simple rule of subtracting two hundred would place her at 165 LSAT in English rough translation. Now her actual position from LSAT in Spanish is 98%. That is, she is the second person in line to get into law school out of 100[5], but the 165-equivalent percentile of 91.4 moves her back to position 92. That is 7 persons are moved ahead of her, just by taking the English version.

Talk about a penalty for speaking Spanish!!!

The lack of linearity in these examples is clearly shown in the next figure, where it can be blatantly seen that the use of just a linear subtraction does not reflect the curvature of the distribution, which is more correctly defined as positive skewed (more candidates to the left side than to the right side). Thus, superior candidates’ (those with scores in the upper 98% or so) penalties are not as bad as those with scores more closely to the mean (those with scores of 160, 150 etc.). But, in any case, a penalty for taking the test in Spanish is plainly wrong.

Distribution of LSAT Scores. From https://www.kaptest.com/study/lsat/lsat-raw-score-conversion/

If any of the three law schools in Puerto Rico just do the simple procedure of subtracting 200 to the Spanish LSAT score without any proper adjustment to compare against those who take it in English, they are, in fact, discriminating against those that take the Spanish version, in favor of those who take it in English.

My advice would be not to take the Spanish version, unless the schools disclose the proper adjustment in scales. Instead of equalizing the admission chances, it decreases it greatly, when there are other candidates that take it in English. As Sam Cooke would say:

But, I know that one and one is two

and if this one could be with you

what a wonderful this would be

---

[1] https://www.lsac.org/lsat/register-lsat/lsat-en-espanol

The Spanish LSAT is not exactly the same as the English LSAT, so it cannot be measured with the same metric of 120-180. However, the results of the Spanish LSAT are based on the same standardized metric, work similarly, and its results can be interpreted in the same way as the English LSAT.

Results of the Spanish LSAT are reported on a scale of 320-380. This scale has the same number of points as the LSAT in English (120-180), but the relationship of the results is not a strict one to one. For example, while a Spanish LSAT score of 335 can be viewed as roughly similar to an English LSAT score of 135, a Spanish LSAT score of 335 does not directly convert to a LSAT score of 135 on the English LSAT. the LSAT in English, nor does it guarantee that the same student would achieve a score of 135 on the LSAT in English. Any comparisons must be made with great care and with the understanding that they are only approximations.

[2] When you write a temperature using the Celsius, Fahrenheit, or Rankine scales, you include a degree symbol. There is no degree in Kelvin because of the definition of a degree. A degree is a change in temperature that is measured against a scale. The size of a degree is arbitrary. Kelvin is different because it is an absolute scale. 0K is absolute zero — the point at which gas molecules have no thermal energy. There is no negative temperature on the Kelvin temperature scale. Why is Rankine temperature is reported in degrees? It has to do with the size of the unit. The Rankine degree (temperature interval) is defined to equal a Fahrenheit degree. A temperature of 459.67 °R exactly equals 0 °F. Fahrenheit, like Celsius, has degrees defined around the freezing and boiling points of water. The National Institute of Standards and Technology (NIST) advises against using the degree when citing Rankine temperatures. From https://sciencenotes.org/why-there-is-no-degree-in-kelvin-temperature/.

[3] You can see the percentile as the candidate place in a row of 100 persons. That is, a candidate in the 99% percentile is the first one on that row (there being ninety-nine persons behind her and there is no one  ahead of her), a candidate with a 94% percentile is the sixth person in the row as there are five persons ahead of him and ninety-four behind.

[4] Using the 2009 to 2012 scale. The other scales are similar.

[5] For the sake of simplicity, we are not considering GPA’s here nor any other possible element, as we just want to illustrate the most likely correct reading of the LSAT in Spanish.