..continued..
http://www.prep4usmle.com/forum/thread/53861 wrote:In the light of the disscussions that were started here on the forum, I gave some thought on how reliable is the information in the Exam Score Thread... a lot of people find the scores in these topics to be by far greater than most people get. It is fairly common to see in the experience threads of different users advices not to look at this thread because will cause discouragement. What is the purpose of these threads?? Well, people post there their scores that they got in different question banks (Kaplan QBank, USMLE World, NBMEs etc) and their score in the real exam. So what? well, users that are still durring their prep can correlate the scores that they got in the Question banks with those of the previous posters and get a "prediction" of their scores in the real exam and if this is lower thaan their target, they postpne the exam and learn more.. etc.. Great thread, then!... If you think of this issue one can say that the only way this will not be usefull if different posters will lie about their scores.. Personally, I don't think so. The issue that was brought up in the disscussion between Leaders was the fact that many users register just to post in there and there are many posts in there from people who we know nothing about, newbies who we don't know if they can be trusted. There's another issue about this thread that bugs me. No doubt about it that the scores that are posted are very high!!! People say very often that "nowadays on being an IMG and having a shot at a US residency you gotta have scores close to 99s". I don't know about this.. this is debaitable. There's another saying: "Most IMGs have very high scores, 99/99". I cannot believe this as hard as I'd try and I cannot test this, but I can try to make a graphical distribution of the scores that are posted in the exam score thread and compare it to the normal distribution. How can I make the normal distribution? Well, I can't make an exact one, but I can make a comparable one to the distribution from the Exam Score Thread (EST), by using the values that are written on the score report.
How did I do it? I gathered the info from the EST (scores provided by 119 users) and I made a graph. Note that the blue line is drawn by myself. For comparation, I knew the Mean and St Deviation of scores from one of the forum users' Score report ('cause I haven't taken CK yet). I tried to approximate the normal distribution. It is not the real one, but is a distribution that has the same Mean and StDev as the real one.
LEGEND: The distribution with the information gathered from the EST is represented using blue columns. The blue line corresponds to it (EST scores). The red columns are represented at the Mean and at 1 and repectively 2 Standard Deviations from the Mean of the Real Distribution. The red line tries to approximate the distribution of the scores of all the test takers.
The Mean of the distributin from the EST is 240 with a StDev of 19.5
The Mean of the real distribution is 221 with a StDev of 24.
The STATs for the Exam Score Thread for Step 1 are posted here . I expect each of you to draw the coclusions about the validity and accuracy of these scores that are posted in the EST.
|
| hypomanic
Forum Senior
Topics: 11
Posts: 153
|
Thanks for the info Manu
I would draw my own conclusion as you have suggested and as I am currently in my manic phase I will also post my conclusion ;-)
1. Self selection bias- People who score better post their scores more often than others who dont pass-(I really appreciate the guts and honesty of those who post even though they flunked and ask for advice)
2. This forum made the difference- lets admit it - you might be skeptical about this - but then thats your excuse for spending so much time reading posts- to improve your scores!
I believe somebody who would otherwise get just an average score gets a much better score thanks to the prep advice on this forum-thats the reason for the difference- you might think i am naive but like Conrad Fischer says- I live in a happy universe - happy, happy, joy, joy :-D
As for not trusting newbies- dont throw the baby out with the bathwater- look at this post by askdoc http://www.prep4usmle.com/forum/thread/34492/1
He is a "newbie" and you are free not believe his scores but I found his posts really interesting- thats the other thing (apart from the prep advice) makes this forum special: an atmosphere of trust that allows people to post without worry that someone might post something idiotic and hurt their feelings AND a willingness of those who are much more wiser, more experienced to take time to help out others who are struggling- Sometimes thats why I come to this forum- it makes my day just see somebody who's got a 99 (or atleast is capable of it) trying to help out someone who flunked a couple of times- It seems that there is Goodness out there - if not in the real world at least in the virtual world of this forum!!
Edited by hypomanic on 05/14/07 - 01:41 PM
___________________
Life is short, the Art long,opportunity fleeting experience treacherous, Judgement difficult - Hippocrates
|
| ManuNastai
|
very well said, very optimistic, very kind, very encouraging, very.. very.. very.. 
|
| nadiabarati
|
I also believe that most of the time people who get very high scores become very happy and come and write their scores. But people who don't get good scores, don't have enough energy to come and write because somehow maybe they are frustrated and angry and depressed. So what do you expect from someone who don't have energy to do routines even. That's why your chart is different from those that are elicited from hard data and scores without even informing the people, as statistics. So any comparison is not a good idea.
|
| coolmavs
Forum Guru
Topics: 39
Posts: 668
|
Agree with nadia......
___________________
You know my creds........
|
| asha kamat
Forum Senior
Topics: 12
Posts: 101
|
Good work Manu, definately stops you from getting disheartened looking at ginormous scores all the time.
cheers!
|
| NE
Forum Guru
Topics: 53
Posts: 504
|
I guess Manu is right; his statistics is a biased because not all of users have posted their experience. I don't know why because are few people who know the person "behind" the user's name! The role of this forum is not to make us to think "How stupid am I and how smart are the others". I think we shoul help each other because, anyway, the role of the score is a limited one . I mean is important to bring you to interviews but , from there, YOUR PERSONALITY, what you are and what you really want can give you a good position!
|
|
| |
| | | | | | | | 
