Average Step 1 score for SGU??

[RAMAIR57]

We always hear about pass rates, but what is the average score of an SGU student on step 1? Just want to know how I did in comarison

[basupran]

~215 would be my guess. Not a bad # to be honest.

[argazul]

275.....since we are playing the guessing game anyway. ;)

[BrendaB_MD]

This number is never published but you can estimate it using basic statistics.

If you assume the US scores and SGU scores are approximately normal, we know the average US score is about 216 and the sd is about 16. The cutoff is 185 and the overall USMG pass rate is something like 96%. As I recall, the SGU pass rate is roughly 91%. So, summarizing

avg_US = 216
sd_US = sd_SGU = 16
pass rate SGU = 0.91
pass rate US = 0.96

First, calculate the US pass rate to check whether a normal distribution gives reasonable answers:

Let z_US = (185-215)/16= -1.93

Let F(z) = the cumulative normal distribution [in Excel use NormSdist(z) ]

So, the calculated US pass rate is:
1 - F(z_US) = 0.974 ~ 0.96

This result is pretty close to the actual value (0.96) so the assumption of a normal distribution is not too bad. So we are justified in using the normal distribution to estimate the SGU average.

Now, work backwards using the SGU data to calculate the average score required to get a pass rate of 91% :

1-F(z_SGU) = 0.91 => z_SGU = -1.37 => avg_SGU = 207

So, the SGU average is roughly 10 pts lower than the US average which.

The numbers I used (pass rates, averages, etc) may not be exactly correct (I was too lazy to look them up) but if you have the correct numbers handy you can easily improve on my calculations; however, I suspect my estimate is roughly correct.




We always hear about pass rates, but what is the average score of an SGU student on step 1? Just want to know how I did in comarison

[jaywalk81]

always impressed by the numbers and stats brenda can come up with

[CANeh]

Dido... I was just about to answer the same thing. With the exact formulas and calculation. Good thing BrendaB was faster than me. ;)

This number is never published but you can estimate it using basic statistics.

If you assume the US scores and SGU scores are approximately normal, we know the average US score is about 216 and the sd is about 16. The cutoff is 185 and the overall USMG pass rate is something like 96%. As I recall, the SGU pass rate is roughly 91%. So, summarizing

avg_US = 216
sd_US = sd_SGU = 16
pass rate SGU = 0.91
pass rate US = 0.96

First, calculate the US pass rate to check whether a normal distribution gives reasonable answers:

Let z_US = (185-215)/16= -1.93

Let F(z) = the cumulative normal distribution [in Excel use NormSdist(z) ]

So, the calculated US pass rate is:
1 - F(z_US) = 0.974 ~ 0.96

This result is pretty close to the actual value (0.96) so the assumption of a normal distribution is not too bad. So we are justified in using the normal distribution to estimate the SGU average.

Now, work backwards using the SGU data to calculate the average score required to get a pass rate of 91% :

1-F(z_SGU) = 0.91 => z_SGU = -1.37 => avg_SGU = 207

So, the SGU average is roughly 10 pts lower than the US average which.

The numbers I used (pass rates, averages, etc) may not be exactly correct (I was too lazy to look them up) but if you have the correct numbers handy you can easily improve on my calculations; however, I suspect my estimate is roughly correct.

[AngryBaby]

This number is never published but you can estimate it using basic statistics.

If you assume the US scores and SGU scores are approximately normal, we know the average US score is about 216 and the sd is about 16. The cutoff is 185 and the overall USMG pass rate is something like 96%. As I recall, the SGU pass rate is roughly 91%. So, summarizing

avg_US = 216
sd_US = sd_SGU = 16
pass rate SGU = 0.91
pass rate US = 0.96

First, calculate the US pass rate to check whether a normal distribution gives reasonable answers:

Let z_US = (185-215)/16= -1.93

Let F(z) = the cumulative normal distribution [in Excel use NormSdist(z) ]

So, the calculated US pass rate is:
1 - F(z_US) = 0.974 ~ 0.96

This result is pretty close to the actual value (0.96) so the assumption of a normal distribution is not too bad. So we are justified in using the normal distribution to estimate the SGU average.

Now, work backwards using the SGU data to calculate the average score required to get a pass rate of 91% :

1-F(z_SGU) = 0.91 => z_SGU = -1.37 => avg_SGU = 207

So, the SGU average is roughly 10 pts lower than the US average which.

The numbers I used (pass rates, averages, etc) may not be exactly correct (I was too lazy to look them up) but if you have the correct numbers handy you can easily improve on my calculations; however, I suspect my estimate is roughly correct.
It is wayyy too early in the morning of my first day in GND to read this post.

EDIT: OK that wasn't too bad, actually. Nice work.

[cooolguy]

i go to sgu and i get the impression that ross has slightly better step 1 scores

[TraperJohnMD]

i go to sgu and i get the impression that ross has slightly better step 1 scores

Yeah, I kind of had this feeling too. But I can't pinpoint the source. Is it real or just an Urban Legend?

Ross has less stringent addmissions, but they produce better step 1 scores?

This seems so counter intuitive.

Any more insight on this?

I know it depends on the student, but as a whole SGU scores should be higher than ROSS, right?

[cooolguy]

even though it is easier to get accepted into ross, it is also much easier to get kicked out/fail out. In addition, their curriculum is set up so they have a lot of classes at once, allowing for better integration. They also have one integrated test with the classes they are taken whereas at SGU we have seperate exams for each class. Overall ross has a tougher basic sciences curriculum.

That is not to say that sgu doesnt...i worked my behind off for my grades my first term, i just feel that they are more usmle geared in terms of their exam prep.

[CANeh]

I will wait for confirmation from BrendaB on USMLE average at ROSS before we start spreading rumors that they are better.

[TraperJohnMD]

even though it is easier to get accepted into ross, it is also much easier to get kicked out/fail out. In addition, their curriculum is set up so they have a lot of classes at once, allowing for better integration. They also have one integrated test with the classes they are taken whereas at SGU we have seperate exams for each class. Overall ross has a tougher basic sciences curriculum.

That is not to say that sgu doesnt...i worked my behind off for my grades my first term, i just feel that they are more usmle geared in terms of their exam prep.


Thanks for the information. I didn't know that. Did you go to both Ross and SGU?

I guess the best school would be Ross first two years and then the SGU clinical connections for years 3 and 4. Lots of Ross and AUC are complaining about how difficult it is to schedule in all the necessary clinicals. Best of luck and thanks.

[BrendaB_MD]

It is very hard to make meaningful comparisons of USMLE scores between carib schools. For one, the data are usually manipulated and presented in so many different ways that it is hard to compare apples to apples. Second, admissions selectivity confounds the educational effect so it is hard to infer that differences in scores reflect differences in educational effectiveness.

As someone pointed out above, Ross has a high attrition rate. Thus, the group who take the USMLE are not representative of those who started the program. At the extreme, you could create enough filters so that only superior students actually get to take the exam.

For the Ross website:
"In 2007, Ross students USMLE Step 1 first-time pass rate was over 92%. For Ross students with a GPA greater than 2.5, the USMLE first-time pass rate was 96.9%"

This pass rate is similar to SGU but you can't infer anything about the relative effectiveness of the programs from this data. To do so, you would need to create a statistical model that controlled for selecitivity (those with higher MCATs and gpas score higher on the USMLE). Also, you would need to create a way of handling dropouts (a similar problem occurs in clinical trials).

USMLE = b0 + b1(MCAT) + b2(gpa) + b3(school)

This model would control for selectivity and enable you to separate the effect of the school vs the effect of selective admissions. You will never get the data to make this comparison because you would have to collect the MCAT, gpa and USMLE score for a large number of students from both Ross and SGU -- which will never happen. There is no evidence to suggest that a given student would perform better on the USMLE if they went to one school vs the other.

[argazul]

i took a semester of statistics and i feel like i didn't learn anything. all this seems foreign to me.

That's what happens when the practice exam is the same as the actual exam with no.s changed.

[cooolguy]

note: this thread was about step 1 scores and i commented on step 1 scores, not on what happens in the clinical years. Clinical years are irrelevant to this thread and to step 1 scores

[shadyhtown]

Maybe there is a correlation between ease of getting clinicals and Step 1 scores. If, as people say, Ross students have to wait longer to start clinicals, they probably utilize that time to study for Step 1, so they have more time to study, and hence higher pass rates/better scores. Also, if during clinicals, they have scheduling problems and unavoidable gaps in between rotations, they might utilize that time to prepare for Step 2, and may possibly have higher pass rates/better scores there as well.

Just a hypothesis, but one worth considering.

It would also be interesting to compare the USMLE scores of the SGU Jan class vs Aug class - we might find that Jan class, which has more time to study, does better for the same reasons.

[kananaskis_girl]

hmmm, that is an interesting hypothesis, I know at least personally that having the extra time probably didn't do much for increasing my score. Most likely would have gotten around the same if I had been in the fall class. Spent most of my first 2 months of "studying" watching CSI, lol I think once you get past a certain period of time, the extra studying is not beneficial, that why the majority of the courses are short and tell their students to take the exam relatively fast after finishing the course. I think the better correlation would be to do a survey, calculate the amount of hours people put into studying and compare the scores based on that. Of course, like it's been said before, you'd also need GPA, MCAT, undergrad GPA to make an accurate correlation. You'll always have outliers...people who barely studied in the January class who scored at the top end, and people who studied for 16 hours a day who barely scraped through. It's always so arbitrary when it comes to standardized testing.

Average step I score?? no idea, I would defer to BrendaB on this one, she's got her stats in order ;)

[cooolguy]

Do not put your faith in what statistics say until you have carefully considered what they do not say. ~William W. Watt

[DrFraud]

This number is never published but you can estimate it using basic statistics.

If you assume the US scores and SGU scores are approximately normal, we know the average US score is about 216 and the sd is about 16. The cutoff is 185 and the overall USMG pass rate is something like 96%. As I recall, the SGU pass rate is roughly 91%. So, summarizing

avg_US = 216
sd_US = sd_SGU = 16
pass rate SGU = 0.91
pass rate US = 0.96

First, calculate the US pass rate to check whether a normal distribution gives reasonable answers:

Let z_US = (185-215)/16= -1.93

Let F(z) = the cumulative normal distribution [in Excel use NormSdist(z) ]

So, the calculated US pass rate is:
1 - F(z_US) = 0.974 ~ 0.96

This result is pretty close to the actual value (0.96) so the assumption of a normal distribution is not too bad. So we are justified in using the normal distribution to estimate the SGU average.

Now, work backwards using the SGU data to calculate the average score required to get a pass rate of 91% :

1-F(z_SGU) = 0.91 => z_SGU = -1.37 => avg_SGU = 207

So, the SGU average is roughly 10 pts lower than the US average which.

The numbers I used (pass rates, averages, etc) may not be exactly correct (I was too lazy to look them up) but if you have the correct numbers handy you can easily improve on my calculations; however, I suspect my estimate is roughly correct.

Ok, your use of #s and calculations is impressive, and its easy to get lulled into accepting your final numbers, but I find it interesting that ultimately, your results rest on the assumption that a SGU student population is normally distributed like a US student population.

I know you have some good students, but I would guess that an SGU population would be positively skewed towards a 185 like most caribbean student populations. Just my 2 cents.

[DrFraud]

It is very hard to make meaningful comparisons of USMLE scores between carib schools. For one, the data are usually manipulated and presented in so many different ways that it is hard to compare apples to apples. Second, admissions selectivity confounds the educational effect so it is hard to infer that differences in scores reflect differences in educational effectiveness.


Yes, but there is one major difference that stands out when I hear comparisons between Ross, AUC, and SGU. Both Ross and AUC have classes continuously through the basic sciences without any extended breaks. The extra two months not only provides additional prep but it also is time when students are not forgetting everything learned during year 1 since all the pharm and path I learned re-enforced my year 1 material.

Given the extra lecture time and shorter breaks at Ross and AUC, it stands to reason that there is an advantage for the top students at Ross and AUC to be better prepared to get higher step 1 scores.

[RussianJoo]

This number is never published but you can estimate it using basic statistics.

If you assume the US scores and SGU scores are approximately normal, we know the average US score is about 216 and the sd is about 16. The cutoff is 185 and the overall USMG pass rate is something like 96%. As I recall, the SGU pass rate is roughly 91%. So, summarizing

avg_US = 216
sd_US = sd_SGU = 16
pass rate SGU = 0.91
pass rate US = 0.96

First, calculate the US pass rate to check whether a normal distribution gives reasonable answers:

Let z_US = (185-215)/16= -1.93

Let F(z) = the cumulative normal distribution [in Excel use NormSdist(z) ]

So, the calculated US pass rate is:
1 - F(z_US) = 0.974 ~ 0.96

This result is pretty close to the actual value (0.96) so the assumption of a normal distribution is not too bad. So we are justified in using the normal distribution to estimate the SGU average.

Now, work backwards using the SGU data to calculate the average score required to get a pass rate of 91% :

1-F(z_SGU) = 0.91 => z_SGU = -1.37 => avg_SGU = 207

So, the SGU average is roughly 10 pts lower than the US average which.

The numbers I used (pass rates, averages, etc) may not be exactly correct (I was too lazy to look them up) but if you have the correct numbers handy you can easily improve on my calculations; however, I suspect my estimate is roughly correct.


awesome work on the calculations.. but i think you used your numbers which are older than the average and sd. for current test takers. I took my test on April 17th and they said the ave. is 218 and SD is 23.

[BrendaB_MD]

That's possible. If so, my estimate would be an upper bound on the average.

While the US average may be different from the SGU average, I don't see any reason why one would expect the shape of the SGU distribution to be much different than the US distribution. I would guess that both distributions are probably "mound shaped" (i.e., approximately normal) but they have different means. Also, if you take large samples, averages are aways normally distributed (via central limit theorem).

Ok, your use of #s and calculations is impressive, and its easy to get lulled into accepting your final numbers, but I find it interesting that ultimately, your results rest on the assumption that a SGU student population is normally distributed like a US student population.

I know you have some good students, but I would guess that an SGU population would be positively skewed towards a 185 like most caribbean student populations. Just my 2 cents.

[BrendaB_MD]

Interesting theory but there is no evidence to suggest that there is a significant difference between schools. Further, there are lots of reasons why they would be similar. Both schools have very similar curricula and approaches to education. The differences you point menttion could have an effect but I suspect they are relatively minor compared to the overwhelming similarities. In any case, we will never know because the data to test any hypothesis comparing schools will never be made available. So, any theories will have to remain just that .....


Yes, but there is one major difference that stands out when I hear comparisons between Ross, AUC, and SGU. Both Ross and AUC have classes continuously through the basic sciences without any extended breaks. The extra two months not only provides additional prep but it also is time when students are not forgetting everything learned during year 1 since all the pharm and path I learned re-enforced my year 1 material.

Given the extra lecture time and shorter breaks at Ross and AUC, it stands to reason that there is an advantage for the top students at Ross and AUC to be better prepared to get higher step 1 scores.

[DrFraud]

That's possible. If so, my estimate would be an upper bound on the average.

While the US average may be different from the SGU average, I don't see any reason why one would expect the shape of the SGU distribution to be much different than the US distribution. I would guess that both distributions are probably "mound shaped" (i.e., approximately normal) but they have different means. Also, if you take large samples, averages are aways normally distributed (via central limit theorem).

Its too bad that more info isn't released, otherwise, we wouldn't have to speculate, but I just wanted to point out that your results rested on a premise that might not be correct if a larger preportion of SGU students scored between a 185-205 range rather than a 210-230 range. You might not have any reason to believe that what I am saying is possible, but to play the devils advocate, my estimate can not be proven wrong either.

[DrFraud]

Interesting theory but there is no evidence to suggest that there is a significant difference between schools. Further, there are lots of reasons why they would be similar. Both schools have very similar curricula and approaches to education. The differences you point menttion could have an effect but I suspect they are relatively minor compared to the overwhelming similarities. In any case, we will never know because the data to test any hypothesis comparing schools will never be made available. So, any theories will have to remain just that .....

Actually, my thoughts about this come from a lack of evidence. When I went to the SGU info seminar in New York a few years ago, the representive was very forthcoming with infomation about statistics that were favorable to SGU, and the pass rates were spoken about quite a bit. When the topic of USMLE averages came up, the rep was less than forthcoming, and did not give a direct answer. I was surprised that they had such accurate data about the pass rates, but not the averages.

I have heard of impressive scores from students at all three, SGU, Ross, and AUC.

For students that I have known at all three schools, SGU, Ross, and AUC, excluding those who did 8 week Falcon or Kaplan review courses, the highest scores I have heard, and these are entirely anecdotal stories from reputable sources, have come from students who attended Ross and AUC.

[spreebee]

Actually, my thoughts about this come from a lack of evidence. When I went to the SGU info seminar in New York a few years ago, the representive was very forthcoming with infomation about statistics that were favorable to SGU, and the pass rates were spoken about quite a bit. When the topic of USMLE averages came up, the rep was less than forthcoming, and did not give a direct answer. I was surprised that they had such accurate data about the pass rates, but not the averages.

I have heard of impressive scores from students at all three, SGU, Ross, and AUC.

For students that I have known at all three schools, SGU, Ross, and AUC, excluding those who did 8 week Falcon or Kaplan review courses, the highest scores I have heard, and these are entirely anecdotal stories from reputable sources, have come from students who attended Ross and AUC.

Yeah, the school sent an email to AUC stating that all 19 people or so that took STEP 1 in april passed and 4 or 5 were 99% and 10 were 90% or something close to that... It sounded too good to be true but hey, it's fine with me....

[BrendaB_MD]

There is virtually no information content in outliers except, perhaps, to indicate what is possible. By definition, outliers do not reflect the underlying process performance so they tell you very little about the educational effectiveness of a school. For example, a Ross student achieved the highest score on the USMLE a few years ago (I believe it was a 273). What can we infer from this? it only shows that it is possible for a student at Ross to score 273. It says nothing about the relative effectiveness of the school. The same holds true for an anecdote about 10 students with a 2 digit score of 99 (which DOES NOT equal 99th percentile).

[BrendaB_MD]

The marketing dept should receive an award for selective sampling. Well done!

Still, their example can be improved:
All 10 students who took the exams on
Oct 7 2001, Apr 7 2002, .....Mar 3, 2008 received scores over 250!

Such a statistic might be true but hardly representative.

Yeah, the school sent an email to AUC stating that all 19 people or so that took STEP 1 in april passed and 4 or 5 were 99% and 10 were 90% or something close to that... It sounded too good to be true but hey, it's fine with me....

[DrFraud]

There is virtually no information content in outliers except, perhaps, to indicate what is possible. By definition, outliers do not reflect the underlying process performance so they tell you very little about the educational effectiveness of a school. For example, a Ross student achieved the highest score on the USMLE a few years ago (I believe it was a 273). What can we infer from this? it only shows that it is possible for a student at Ross to score 273. It says nothing about the relative effectiveness of the school. The same holds true for an anecdote about 10 students with a 2 digit score of 99 (which DOES NOT equal 99th percentile).
Yes, I have to agree that this is the most misleading USMLE score.

[DrFraud]

The marketing dept should receive an award for selective sampling. Well done!

Still, their example can be improved:
All 10 students who took the exams on
Oct 7 2001, Apr 7 2002, .....Mar 3, 2008 received scores over 250!

Such a statistic might be true but hardly representative.

i think its ironic when a medical school releases stats that would be a form of statistical error based on questions from the actual exam, step 1, that the stats are being used to describe. :confused:

[mmod]

even though it is easier to get accepted into ross, it is also much easier to get kicked out/fail out. In addition, their curriculum is set up so they have a lot of classes at once, allowing for better integration. They also have one integrated test with the classes they are taken whereas at SGU we have seperate exams for each class. Overall ross has a tougher basic sciences curriculum.

That is not to say that sgu doesnt...i worked my behind off for my grades my first term, i just feel that they are more usmle geared in terms of their exam prep.

i go to Ross currently. i looked at SGU's cirriculum, its no joke (see below on why i read the SGU boards in the first place). bottom line medical school is tough. but i do agree its pretty easy to get kicked out of Ross.... and unfortunately its getting even easier to fail out/tougher to pass. i don't really want to go into specifics (its too late... my brain doesn't want to think in such detail and i don't want to force it do so either.)

if the SGU avg step 1 is 215 then my personal opinion is that is definately higher than Ross's. i don't know what the avg Ross Step 1 score is but i'm almost certain its not above 215. also at Ross before students can even take step 1 they have to take and pass an exam called the COMP....its pretty much a like a shelf exam for Step 1 (I'm pretty sure the NBME adminsters.... if all goes well, i will find out in a couple of months when i take the COMP myself).

SGU probably has a similar type mechanism before students can take Step 1.... for what i understand most Carib schools do.

(reason why i sometimes browse the SGU board is because a guy who I sort of knew....but didn't know well.... currently goes to SGU.... so I just was curious about what SGU was all about.)

[BrendaB_MD]

MCAT scores and gpa are predictors of USMLE scores. Because SGU is more selective, you would expect the average USMLE score at SGU to be higher than the USMLE score at Ross. While we will never be able to put the proposition to the test, I seriously doubt that there is a difference in educational effectiveness. The curricula and educational approach are very similar at both schools and, for that reason, it would be surprising to find a difference in educational effectiveness.

Although Ross starts at a disadvantage with respect to the USMLE (due to less selective admissions) I suspect Ross' high attrition rate evens things out. Both schools have attrition; however, by all accounts, Ross' attrition rate is much higher than SGUs. Thus, Ross lops off the bottom end of the curve and the survivors are on par with SGU by the end of basic sciences.

If neither school had significant attrition and the USMLE scores were similar, you could make an argument that educational effectiveness is higher at Ross. However, given the preclinical bloodbath at Ross such an argument is not valid.

At the end of the day you are where you are and it is up to you to do the best you can on the USMLE. Thus, for students at either school, it makes little sense to worry about differences in USMLE scores. For students choosing a school, there is no evidence to suggest that one school is more effective than the other with respect to USMLE scores.

[DrFraud]

MCAT scores and gpa are predictors of USMLE scores. Because SGU is more selective, you would expect the average USMLE score at SGU to be higher than the USMLE score at Ross. While we will never be able to put the proposition to the test, I seriously doubt that there is a difference in educational effectiveness. The curricula and educational approach are very similar at both schools and, for that reason, it would be surprising to find a difference in educational effectiveness.

Although Ross starts at a disadvantage with respect to the USMLE (due to less selective admissions) I suspect Ross' high attrition rate evens things out. Both schools have attrition; however, by all accounts, Ross' attrition rate is much higher than SGUs. Thus, Ross lops off the bottom end of the curve and the survivors are on par with SGU by the end of basic sciences.

If neither school had significant attrition and the USMLE scores were similar, you could make an argument that educational effectiveness is higher at Ross. However, given the preclinical bloodbath at Ross such an argument is not valid.

At the end of the day you are where you are and it is up to you to do the best you can on the USMLE. Thus, for students at either school, it makes little sense to worry about differences in USMLE scores. For students choosing a school, there is no evidence to suggest that one school is more effective than the other with respect to USMLE scores.

I guess we can go back and forth on this forever, but minus students who go to an eight week review course like falcon/kaplan, there are some indications that 1) top scores at Ross and AUC are higher than SGU due to the differences in curriculum and increased period of lecture time at AUC and Ross and 2) SGU obfuscates this point by withholding data on USMLE averages.

[BrendaB_MD]

How can you tell whether this is due random variaton rather than a systematic difference between programs? I am sorry, but you can't attribute the high scores at Ross to any systematic difference between schools. It is much more likely to be due to random variation. Unlike averages, outliers (such as the Ross score) tell you very little. Also, we don't have the SGU data. For all we know, the SGU scores are higher. You can't assume they are lower just because SGU doesn't make them public. Further, even if Ross and AUC did have the highest scores, it would tell you very little about the relative effectiveness of the programs.

For the record, I have no connection to SGU, Ross or any carib school. If you look at my past posts, you will see that I am hardly a SGU cheerleader.




I guess we can go back and forth on this forever, but minus students who go to an eight week review course like falcon/kaplan, there are some indications that 1) top scores at Ross and AUC are higher than SGU due to the differences in curriculum and increased period of lecture time at AUC and Ross and 2) SGU obfuscates this point by withholding data on USMLE averages.

[DrFraud]

How can you tell whether this is due random variaton rather than a systematic difference between programs? I am sorry, but you can't attribute the high scores at Ross to any systematic difference between schools. It is much more likely to be due to random variation. Unlike averages, outliers (such as the Ross score) tell you very little. Also, we don't have the SGU data. For all we know, the SGU scores are higher. You can't assume they are lower just because SGU doesn't make them public. Further, even if Ross and AUC did have the highest scores, it would tell you very little about the relative effectiveness of the programs

I'm not trying to make any assumptions. I was commenting on your point earlier that you can not make inferences if there is no evidence, and one key point I was trying to make is that SGU, which went to great lengths to use stats to persuade me that it has a program superior to other Carib schools using step passing rates, would be inclined to release the data about usmle averages if it supported their arguments.....and therefore, their not releasing this data could suggest that the data is less favorable to other Carib. schools and less favorable than what you indicated earlier in this thread.

[rokshana]

funny how 273 is the number always touted.

Dr. M will tell you about the student at sgu who scored a 273 (and verified with the USMLE that it tied for the highet score that year). He is an EM resident at Duke (though by now he may be finished, since he graduated in 04 i believe)- his sis is an ob resident in DC (such a smart family!!:D).

there are a number of high scores (99 - NOT %tile people!!! is not uncommon, and well 90 is in the 220 range-strong score, but not unattainable)coming from sgu- maybe the reason its not screamed from the rooftops is that, well, its not that unusual...

but realize there are people who fail or just barely pass as well....how YOU do is really going to be up to you- the info is there, some tests are given in a manner that help with thinking USMLE style, and review courses abound...pay attention, use the basics years as the time to really, really learn and understand the material, do umpteen questions to get into the mindset of the USMLEs and you will perform well...

FYI an average WAS posted a few years ago...sgu's avg was ~209 then (whether that was a 1st time pass average, well)....

[BrendaB_MD]

I love it when this happens....

...sgu's avg was ~209

[DrFraud]

Vindication!
I love it when this happens....

And how exactly is this a vindication? Interesting..........if this came from another school, you would call it anactodal selective sampling?

Im sorry, but if the 'real' step 1 average was anything close to a 207 or 209 at SGU, the reps who spoke to my group 2-3 years ago would have been throwing this # around like the pass rate numbers, instead of giving vague 'we don't keep track of this' or 'I dont have this info available' type of answers, and you can be sure this number would be promiantly displayed somewhere on the upper right hand corner of the Web site in with flashing font bigger than 20.

[rockstar1999]

From Ross Site:

For 2007, Ross students had an overall first-time pass rate for the United States Medical Licensing Examination (USMLE) Step 1 of 91%, with an average exam score of 211. Students who earned a basic science GPA of 2.5 or better at Ross had a first-time pass rate of 97%, with an average exam score of 217.

My 2c. - Since you have to pass the comp at ross before taking step 1 - many who fail the comp 3 times will be tossed and never take step 1 ... therefore only those kids who PASS the comp take step 1 ... and if you pass the comp its pretty damn likely that you will pass step 1

[cooolguy]

dont we have to pass the bsce 2 to take step 1? How different is that from the comp you guys are talking about?

[kananaskis_girl]

no, you don't have to pass the bsce 2 to take step I...you need to pass it to start clinicals in the UK. If you fail bsce 2, you can still write step I, they just "recommend" that you don't until you pass bsce 2

[rokshana]

and i gotta feeling that ross's comp is way more difficult than the bsce II....its like a real test unlike the bsce II...

[argazul]

so this is the comp that Ross uses?

NBME® : Services for Medical Schools : Subject Examination Services (http://www.nbme.org/programs-services/medical-schools/subject-examinations/comprehensive-exams.html)

[cooolguy]

can sgu students choose to take these to see where they stand or do you have ot take them through your medical school? Is it a good idea to take these?

[Kongakut]

Usually students take the NBMEs when they are specifically studying for the step to gauge how they might perform. Supposedly, the 2nd NBME is very much like the step and the scores are not too far off, anecdotally of course. The NBMEs do not give answers or explanations so they are not good for reviewing specific topics but they can identify areas that you may be weaker in.

[DrFraud]

so this is the comp that Ross uses?

NBME® : Services for Medical Schools : Subject Examination Services (http://www.nbme.org/programs-services/medical-schools/subject-examinations/comprehensive-exams.html)

Usually students take the NBMEs when they are specifically studying for the step to gauge how they might perform. Supposedly, the 2nd NBME is very much like the step and the scores are not too far off, anecdotally of course. The NBMEs do not give answers or explanations so they are not good for reviewing specific topics but they can identify areas that you may be weaker in.


Ok, I think were are confusing the two types of NBMEs. For the benefit of argazul.....the first group, from the first line on the link that you provided, are usally called 'comps.' and are 4 hour tests that students from AUC and Ross have to get a 62 or above on. (you get three attempts). These are administered by medical schools, and can be used for determining eligibility for the step by a medical school.

The second type on nbme referenced in the post from Kongakut are for students, not medical schools. Students log onto the NBME site, pay $45, and take a 200 question internet test, and get a number grade that is scaled to a three digit usmle grade. Medical schools do not administer these exams, and they have no bearing on medical school grades or eligibility for the step.

In both cases, the questions are 'retired' step questions, but it is possible to see similar questions and pictures from these mock exams on the actual step.

[argazul]

ok that clears it up. I figured the comp was different from the NBMEs 1-6. that's why looked undet "services for med school". Also a friend of mine at a US med school had to take all subject shelves at the end of their courses and the comp at the end of the 2nd year.

[shadyhtown]

Do the schools or residency programs find out our scores on the student NBME? Or is it just like a practice USMLE for our benefit?

[DrFraud]

Do the schools or residency programs find out our scores on the student NBME? Or is it just like a practice USMLE for our benefit?

that's exactly what it is for, assuming of course you are not referring to the paying $45 to the NBME per exam as something that is for 'our benefit'....:evil:

[endymion]

So based on what I'm reading in this topic, Ross may be a little easier to get into, but has both a higher attrition rate and a higher "filter" to get by to even be allowed to sit for the STEP 1.

I note neither school seems to be promoting their graduation RATE, which is something even college football teams do!

Isn't that something that would be of even more interest to a potential student perusing valueMD ?

It also seems strange to me that many international schools don't require passage of any STEPs to receive an MD, but I read where many Caribbean Med Students are being forced to transfer because of low STEP 1 scores. I can understand it being a requirement (STEP1) to enter clinical rotations in the US, but why take students money to repeat basic science semesters but not let them take more time to pass the STEP 1 ?

[jaywalk81]

keep in mind that the amount of time USMLE can be taken is set by the FSMB board, not by the schools, tho some schools do have certain guidelines.

[BrendaB_MD]

Some people are negative about Ross because of the high attrition rate; however, I don't think there is anything wrong with it so long as prospective students know in advance what they are getting themselves into. Medical schools in other countries (I believe France and Argentina are examples) work on the Ross model: admission is easy but there is a very high rate of attrition. Medical school is hard and you would expect a higher rate of attrition at a school that admits less academically qualified students.

You can take a couple of perspectives on this. On the one hand one might say that Ross treats applicants like adults and recognizes that people may have less than stellar records for various reasons (having to work, family problems, illness) and leaves it to the student to decide whether they can succeed in medical school. On the other hand, some might say that Ross is just preying on student's desperate hopes to make money. In my view, there is nothing wrong with the Ross approach so long as students understand the process. Various estimates of attrition rates have been posted over the years; however, as you say, the schools don't make this information public.








So based on what I'm reading in this topic, Ross may be a little easier to get into, but has both a higher attrition rate and a higher "filter" to get by to even be allowed to sit for the STEP 1.

I note neither school seems to be promoting their graduation RATE, which is something even college football teams do!

Isn't that something that would be of even more interest to a potential student perusing valueMD ?

[DrFraud]

It also seems strange to me that many international schools don't require passage of any STEPs to receive an MD, but I read where many Caribbean Med Students are being forced to transfer because of low STEP 1 scores. I can understand it being a requirement (STEP1) to enter clinical rotations in the US, but why take students money to repeat basic science semesters but not let them take more time to pass the STEP 1 ?

I'll try and answer your last question first, and then address what you wrote in the first paragraph. Basically, the more time a school gives its students to take the step, the less likely some students will pass. This might sound paradoxical, but basically, if you gave some students one year instead of three months to write the exam, some students might actually elect to study for this year when they are ready earlier. Doing this is often not condusive to learning medicine or passing step 1. During the extra time, students often forget more then they learn, and the only thing the extra time creates is an administrative headache for a school trying to keep track of the many students who are 'officially' enrolled but who are stuck in medical school purgatory between the basic sciences and clinical rotations.

[DrFraud]

So based on what I'm reading in this topic, Ross may be a little easier to get into, but has both a higher attrition rate and a higher "filter" to get by to even be allowed to sit for the STEP 1.


There is actually something more to this that isn't often discussed by students but is by faculty when comparisons are made between teaching at SGU vs. Ross or SGU vs. AUC. Basically, it's how the examinations affect learning. In general, professors have some flexibility in how agressive or how easy they make a test, but in addition, an administration will often have discussions with a prof as to how aggressive or easy the tests should be. This will affect both how many students pass, but it will also affect how rigorous and challinging a curiculum is for students who are in the upper third of a class.

At the end of the day, its difficult to design tests to both optimally challenge the upper third of a class and at the same time pass a large majority of a caribbean medical school students, and so a balance must be found.

At SGU, and this is communicated to me by profs who have taught there so I am not making this up, the balance is found in an area that insures that most of the students pass. They can get away with this because they have higher admissions standards and therefore, less concern about students failing step 1. At Ross and AUC, the testing tends to be more aggressive, which can lead to higher attrition, but it also is more demanding on students in the upper third of a class.

Whats the end result....well, more students at a school that has easier tests will be that there will be higher passing rates at the school but lower averages on standardized tests (normalize the results so that comparisons are made between students with similar entrance stats).

On the other hand, a school that has aggressive testing practices will be more advantagous for students in the upper tier, assuming they are able to pass, where they would benefit from a more challanging curriculum.

And so, while many students who make comparisons to Step 1 averages between SGU, Ross, and AUC attribute the differences in Step 1 averages to weeding out, there are some differences in the curriculum that also can affect step 1 performance, and this is one of them that may provide an advantage to upper tier students who elect to go to AUC or Ross instead of SGU in addition to the other differeces I mentioned in prior posts, however, it must be noted that this is minor in comparison to how hard a student works and applies themself.

[AngryBaby]

There is actually something more to this that isn't often discussed by students but is by faculty when comparisons are made between teaching at SGU vs. Ross or SGU vs. AUC. Basically, it's how the examinations affect learning. In general, professors have some flexibility in how agressive or how easy they make a test, but in addition, an administration will often have discussions with a prof as to how aggressive or easy the tests should be. This will affect both how many students pass, but it will also affect how rigorous and challinging a curiculum is for students who are in the upper third of a class.

At the end of the day, its difficult to design tests to both optimally challenge the upper third of a class and at the same time pass a large majority of a caribbean medical school students, and so a balance must be found.

At SGU, and this is communicated to me by profs who have taught there so I am not making this up, the balance is found in an area that insures that most of the students pass. They can get away with this because they have higher admissions standards and therefore, less concern about students failing step 1. At Ross and AUC, the testing tends to be more aggressive, which can lead to higher attrition, but it also is more demanding on students in the upper third of a class.

Whats the end result....well, more students at a school that has easier tests will be that there will be higher passing rates at the school but lower averages on standardized tests (normalize the results so that comparisons are made between students with similar entrance stats).

On the other hand, a school that has aggressive testing practices will be more advantagous for students in the upper tier, assuming they are able to pass, where they would benefit from a more challanging curriculum.

And so, while many students who make comparisons to Step 1 averages between SGU, Ross, and AUC attribute the differences in Step 1 averages to weeding out, there are some differences in the curriculum that also can affect step 1 performance, and this is one of them that may provide an advantage to upper tier students who elect to go to AUC or Ross instead of SGU in addition to the other differeces I mentioned in prior posts, however, it must be noted that this is minor in comparison to how hard a student works and applies themself.
While I'm following your logic and it certainly seems plausible, its absolutely speculative nature can't be ignored.

You may or may not be on to something and no offense, but it's so speculative and there's such a dearth of info and ways to adequately compare that I don't see the point in debating. Your point may or may not be accurate, we'll never know and it really doesn't matter.

[argazul]

there is only one true and tested theory, spamming = high step 1 score.:twisted::lol::twisted:

[rokshana]

I note neither school seems to be promoting their graduation RATE, which is something even college football teams do!

basically the schools state their attrition rate- what ever left over is basically the graduation rate (not necessarily the % that get into a residency...)

It also seems strange to me that many international schools don't require passage of any STEPs to receive an MD,

so why should an international school require the USmles as a requirement for an MD?- that would be like the US schools requiring the PLAB as a requirement for graduation...there are many people who go to a school like sgu that will not go on to practice (or train) in the US...the steps are not needed for these students...

[endymion]

I don't find anything inherently negative about a high attrition rate, on this I tend to agree with you.

I do recall a former faculty member of RUSM feeling the same way. He/she advocated an even more open admission policy but an even higher bar to progress to second year i.e. third semester. Their concern was enough faculty and faculty hours to work with the students who would succeed.

I suppose one could argue taking two semesters of tuition is only half as bad as taking four.

It doesn't appear Ross has updated attrition or STEP 1 passage rates since 2003 figures. I'm not sure about the others in the big three or four.

What doesn't strike me as fair is moving the bar to students in mid stream, and that seems to be happening with greater frequency as the class size increases and students are making best guesses with old data.

Some people are negative about Ross because of the high attrition rate; however, I don't think there is anything wrong with it so long as prospective students know in advance what they are getting themselves into. Medical schools in other countries (I believe France and Argentina are examples) work on the Ross model: admission is easy but there is a very high rate of attrition. Medical school is hard and you would expect a higher rate of attrition at a school that admits less academically qualified students.

You can take a couple of perspectives on this. On the one hand one might say that Ross treats applicants like adults and recognizes that people may have less than stellar records for various reasons (having to work, family problems, illness) and leaves it to the student to decide whether they can succeed in medical school. On the other hand, some might say that Ross is just preying on student's desperate hopes to make money. In my view, there is nothing wrong with the Ross approach so long as students understand the process. Various estimates of attrition rates have been posted over the years; however, as you say, the schools don't make this information public.

[endymion]

I can see some sense in the "stale bread" methodology you outline. I certainly agree students SHOULD be pushed to take the STEP 1 as soon as possible. It would move them along faster in their career path, and I'd bet a fair amount of delays are due to nerves and worry, which for many, thankfully turns out to be needless.

But doesn't the fact that international medical school graduates seeking US licensure take the STEPs over much greater time intervals and tend to do very well in the USS medical workforce indicate a potential hole in that theory ?


I'll try and answer your last question first, and then address what you wrote in the first paragraph. Basically, the more time a school gives its students to take the step, the less likely some students will pass. This might sound paradoxical, but basically, if you gave some students one year instead of three months to write the exam, some students might actually elect to study for this year when they are ready earlier. Doing this is often not condusive to learning medicine or passing step 1. During the extra time, students often forget more then they learn, and the only thing the extra time creates is an administrative headache for a school trying to keep track of the many students who are 'officially' enrolled but who are stuck in medical school purgatory between the basic sciences and clinical rotations.

My original post you quoted:

Originally Posted by endymion http://www.valuemd.com/images/buttons/viewpost.gif (http://www.valuemd.com/st-georges-university-school-medicine/159269-average-step-1-score-sgu-post851055.html#post851055)
It also seems strange to me that many international schools don't require passage of any STEPs to receive an MD, but I read where many Caribbean Med Students are being forced to transfer because of low STEP 1 scores. I can understand it being a requirement (STEP1) to enter clinical rotations in the US, but why take students money to repeat basic science semesters but not let them take more time to pass the STEP 1 ?

[endymion]

keep in mind that the amount of time USMLE can be taken is set by the FSMB board, not by the schools, tho some schools do have certain guidelines.

It looks like the FSMB is a federation with different State Boards setting their own standards as well as often granting some reciprocity to other states. Few have caps on STEP 1 though they seem to have more caps as the STEP number rises.

The USMLE has a limitations on how often one can take an exam and a cap per year. I don't think they have any "lifetime" caps.

[endymion]

so why should an international school require the USmles as a requirement for an MD?- that would be like the US schools requiring the PLAB as a requirement for graduation...there are many people who go to a school like sgu that will not go on to practice (or train) in the US...the steps are not needed for these students...

I did phrase that badly didn't I ? I'm certainly NOT seeking to impose any US pax medicana on the world. I do note that the PLAB is required in the UK NOT for UK medical grads, but for non EC doctors seeking entry into the UK medical system.

I don't think international schools should be required to, nor require STEPs for an MD degree. My point was supposed to be, it appears international medical education may well be faculty driven, whereas the US/Caribe medical system is administratively driven. Why throw away a pool of potentially good MDs from one system when they might thrive and do well under the other ?

[endymion]

the only thing the extra time creates is an administrative headache for a school trying to keep track of the many students who are 'officially' enrolled but who are stuck in medical school purgatory between the basic sciences and clinical rotations.

How much can it cost to buy an extra file cabinet and a box of file folders ? By all mean charge a small administrative fee each semester to cover those costs. I'd imagine those students who want to continue to pursue their MD would gladly pay it.

Certainly being "off track" shouldn't be allowed to penalize those coming up behind them who stayed "on track". A detour not a dead end.

[cooolguy]

What happens if you attend SGU, pass the steps and are not accepted into a residency in your first attempt? Im sure this happens to students, does anyone have a rough percentage of students that have to undergo this unfortunate event? I see the stats son their website but Im not sure how to read it.

[Wet_Coma]

Coolguy hit the nail on the head. And dont forget SGU has DES, while Ross give you the boot.

even though it is easier to get accepted into ross, it is also much easier to get kicked out/fail out. In addition, their curriculum is set up so they have a lot of classes at once, allowing for better integration. They also have one integrated test with the classes they are taken whereas at SGU we have seperate exams for each class. Overall ross has a tougher basic sciences curriculum.

That is not to say that sgu doesnt...i worked my behind off for my grades my first term, i just feel that they are more usmle geared in terms of their exam prep.

[Wet_Coma]

If you are eligible (i.e. pass the USLMLE steps) then you shouldnt have a problem as long as you shoot within your means and also have backup choices. If that fails, you still have the scramble as a safety net. If you dont have USMLE in the 90's you shouldnt be applying to certain residencies.

[rokshana]

If you are eligible (i.e. pass the USLMLE steps) then you shouldnt have a problem as long as you shoot within your means and also have backup choices. If that fails, you still have the scramble as a safety net. If you dont have USMLE in the 90's you shouldnt be applying to certain residencies.

well, its not always that rosy- just because you pass all your steps doesn't guarantee a residency- not even a crappy one. just from my feel- i would say ~5-7% of those that actually try for a spot don't get one- for many reasons known and unknown, but interestingly enough unexpected openings do happen and its possible to get a spot later in the year.

and there ARE people who have 90s and still don't match (there are usually other issues though...).

[argazul]

One thing i don't know about Ross when they say that they have an integrated exam. Does that mean the questions are integrated, or just the exam is integrated with certain % of questions from Biochem, Anatomy etc.?

[Kongakut]

As you know, I am a Ross expert (sarcasm here) and from their forums, it appears that they have a certain percentage of questions in each discipline.

[argazul]

So its not truly integrated.....

[Kongakut]

It appears so...here (http://www.doctorsofthecaribbean.com/) is a blog from a student who failed and she lamented about that very topic.

[argazul]

Right. She did mention the % of questions on one subject distributed among exams. I think that it might help with test taking abilities when during a test you constantly have to switch between Biochem, Anatomy, Histo, Physio, instead of thinking about one subject at a time. But it appears that its not a true practice of USMLE type questions.

[Kongakut]

Well, Rossies do take an integrative comp exam at the end of 2nd year and from other bloggers out there like J (http://www.whatsjessupto.blogspot.com/), it seems it is very much like Step 1.

[MDSkillz]

i just finished my 2nd year down on the island a few months ago, studied like there was no tomorrow all summer long, and got my score today... 261.... boo yaaaaaaaaaaaaaaa

[NasW]

WOW!!!!!!!!!!! That's really amazing, CONGRATULATION!!!!!!! would you please tell us a bit about how you studied for step 1 and what books and materials you used to prepare for it?

really appreciate it, and congrats again!!!!

[CANeh]

Very impressive. Congrats!
Do you mind sharing your GPA from first 2 years?

[shadyhtown]

Nice - congrats! So you studied all summer - does that mean you have to start rotations later than usual?

[DrFraud]

i just finished my 2nd year down on the island a few months ago, studied like there was no tomorrow all summer long, and got my score today... 261.... boo yaaaaaaaaaaaaaaa

Congrats!!! That's a great score!!! :-happy:rock::-rainbow:dancing::banana::dance::yeah:

[dunsoon]

Wow! That's amazing!! Good job!

[jaywalk81]

great job. good luck and have fun in clinicals