New Graphs Added - Decision Date Graphs
(0 comments)Posted on October 28, 2008 11:32
When you go to a school's page and click on "Graphs," besides the graph that show you the accepted and rejected users and their numbers, there is a new "Date of Decision Graph" that shows the monthly breakdown of decisions for each school. With these graphs, you'll know better when to expect to hear from your schools.
For example, Georgetown's graphs are here. You can see that their peek months for sending out acceptances are November and December.
Update: The decision date graphs currently only work with Firefox, Safari and Google Chrome. Unfortunately, it doesn't work with Internet Explorer 8 because IE8 is not HTML 5 compliant. I recommend using Firefox. It's open source and a very stable, versatile browser.
Update 2: The graphs now work with Internet Explorer.
Decision Date Graphs now work on Internet Explorer
(0 comments)Posted on November 04, 2008 13:20
Previously, these graphs only worked on Firefox and Safari, but now they also work on Internet Explorer as long as you have flash installed (which you probably do if you've ever accessed YouTube). To see an example, go to Boalt's page.
California Proposition 8 Advertisements
(1 comments)Posted on November 03, 2008 13:58
I apologize. There were a lot of "Vote Yes on Proposition 8" ads this morning. I've put in place a filter to make sure those ads don't appear.
FAQ
(21 comments)Posted on October 21, 2008 23:46
About
What is Chiashu?
Chiashu is a website that helps pre-law and, in the future, law students & graduates navigate their school and career choices.
What if I have questions?
You can email admin at thiswebsite's name dot com or leave a question under this FAQ, and I'll do my best to answer it.
What is this website's email policy
I will NEVER give out a user's email to anyone unless it's required by law. I also can't be responsible for things outside of my control.
Account
How do I register?
Go to this link to create a new account.
How do I edit my account?
Login first, then go to this link to edit your account
What does 'Anonymous' mean?
Anonymous just means that your user details, such as your gender, undergrad, race, etc. are not displayed. It also means that when you update your school status, people won't be able to see the updates on your profile page and they can't link to your profile from the school pages themselves. Remember, your email is NEVER displayed, whether your account is anonymous or not.
Applying to Law Schools
How do I see all the school probabilities?
Go to this news item for a detailed explanation on how to see your probabilities.
How do I see probabilities from previous cycles?
After you have logged in and added schools to your list, either from Law Schools/Edit My Schools or Law Schools/2007 Probability, you can go to Law Schools/View My Schools to see probabilities from 2002 and afterwards.
How are the probabilities determined?
Unlike websites that use user inputs such as Law School Numbers, this website's probabilities use logistic regression based on ALL applicants from a particular year, exactly what LSAC does on its website. What is different is that LSAC has traditionally displayed ranges that are wide, such as "0-25% of applicants with your numbers were accepted to this school." I did not like these wide ranges when I was applying because I wanted to know whether it was closer to 0 or closer to 25, so I analyzed LSAC data to provide exact numbers. I started this project back in 2004, and I've maintained historical data for the past few years.
But it doesn't take into account soft factors.
It is hard to judge how much soft factors matter. People often overestimate how much their activities affect admissions decisions, because they put time and effort into these activities. The most predictive factors are still LSAT, then GPA.
Are there exceptions?
My only observation - race & gender play a part in admissions decisions, but not nearly as big as LSAT & GPA. However, schools and LSAC don't release admissions information based on these factors. I'm currently looking for ways to solve this problem.
How about user-input data?
User inputs suffer from a lot of sampling errors, meaning that certain users are more likely to input their data. For example, a person who is admitted is more likely to want to broadcast it than someone who is rejected. If you want to see an example of logistic regression on user-inputed data, go to "2007 Probability" under "Law Schools." You can then see the difference between data using ALL applicants (2007) and data using applicants who voluntarily contributed their numbers (2007 LSN).
Besides probabilities, what other tools can help me determine where to apply?
Peer schools is one of the tools that can help you choose where to apply. For example, if your dream school is Boalt, but you want some back-ups, go to Boalt's page, and click on "Peer Schools". Set "Status at this school" to "Rejected" and "status at other schools" to "Accepted" or "Attending". You will then see where people are accepted or attending even if they're rejected by Boalt. If you click on the "# Users" link, the users who fit that profile will appear to the right.
Where do people who are rejected by NYU go?
(2 comments)Posted on June 05, 2008 19:52
The answer? Michigan or Fordham.
To get information like this on the website, go to a school's page and click on "Peer schools" under the school's name, you are then led to a page like this one.
You can change the criteria and see where people who apply to NYU (and other schools) also apply to. This is a great way to help you think about where to apply. If NYU is your dream school but you want to have some back-ups, you can see where people who are rejected by NYU are accepted or attending. You can then add those schools to your list.
Update on October 28, 2008: Now when you click on the numbers under "# Users" in Peer Schools, you can see the users who fit the criteria you specified.
How to see all probabilities at the same time
(6 comments)Posted on April 16, 2008 03:02
I have changed the way the website works, so now you can see all the 2007 probabilities at the same time, and also very easily add schools to see all the probabilities over the past years. This is especially helpful for the schools that no longer provide such information. I'm reviewing and learning a statistics software package to predict the numbers for the schools that no longer publish their numbers, but that might take a while.
In any case, to see all the probabilities, you need to:
1. Register
2. Go to Account/Users - Edit My Account and update your LSAT and GPA
3. Go to Law Schools - 2006 Probability to see the probabilities for all the schools. You can add schools to your list from there, and they'll show up elsewhere.
Minor fix to user photo upload
(0 comments)Posted on October 22, 2008 00:16
I have fixed photo upload so when thumbnails are created, they would be square and uniform. I'm not an artist, so it'll be a while (or never) before this website looks "pretty," but I'll continue to make changes here and there.
FAQ Coming Soon
(0 comments)Posted on October 12, 2008 18:21
I'm in the process of writing up an FAQ, expect to see it in a couple weeks. If you have other questions you'd like to ask. You can post them here as comments.
Maybe hacker traffic?
(0 comments)Posted on August 08, 2008 08:20
I apologize for the website being down. I'm on vacation, and I found out that my server was pegged at 100% with random traffic. I've changed some things and hopefully it's okay now. I'll try to make sure it doesn't go down after I get back from vacation.
The Optimism of LSN
(2 comments)Posted on June 09, 2008 11:14
I have put up the official 2007 numbers and an analysis of the data from Law School Numbers. If you login and go to Law Schools - 2007 Probability, you will see the 2007 numbers that LSAC has, which includes all the applicants that year. Next to the 2007 LSAC numbers, there are also 2007 LSN and 2008 LSN numbers, which use the same formula as LSAC but use data from LSN (special thanks to the authors of that website).
It seems to me that there is a slight skew in the data, as the numbers tend to be higher if you use LSN data rather than the entire pool. Do people have clever suggestions on why that is?
Safeties, Targets, Reaches - Graphs!!!
(4 comments)Posted on April 22, 2008 11:09
I've added graphs to each school's page. These graphs are updated every day at 1 am PST. Some things to note about the graphs:1. Green is for "accepted."
2. Red is for "rejected."
3. The green line represents the average probability of users who were accepted. Therefore, all users on that line should have the same probability.
4. What does it mean when a line slopes down more? Well, it means GPA is more important to that school. For example, Harvard cares about LSAT and GPA equally, whereas Stanford really cares about GPA (notice that the lines for Stanford are really steep). On the other hand, NYU thinks LSAT scores are really important (its lines are less steep than Harvard's).
5. I would consider above the green line to be "Safeties," below the red line to be "Reaches," and in between are "Targets," but it really depends on how far from each line you are.
Again, leave a comment in this news item if you have questions.
2007 Numbers!! Extra: 2008
(2 comments)Posted on May 01, 2008 08:32
I have finally assembled the 2007 numbers. It was a much more difficult process because LSAC has not published their data either. I also took a look at Law School Numbers, but I noticed there is a big sampling problem - people tend to put their numbers up more often when they're accepted.
In any case, I used both a statistical model and some computer science model-training techniques to come up with the 2007 numbers. It looks like LSAC might be putting up 2007 data this weekend as well, so I'll compare my numbers with theirs and modify the model.
I've put up 2008 numbers under "View My Schools" as well. These numbers are not final, but they should be close.
iPod Shuffle giveaway!!
(12 comments)Posted on April 16, 2008 14:31
To celebrate chiashu moving to a new platform, I'm giving away a 1 gig iPod shuffle for free. The winner of the contest will be chosen randomly on 5/10/2008 (the day I get back from vacation). To qualify, all you need to do is:1. Register on this website.
2. Have a valid email so I can email you if you win.
3. Leave a comment under this news item (Click on "iPod Shuffle giveaway!!" above).
Other features on the website will increase your chance of winning as well. For example, if you write a review or rate other people's reviews, your chance of winning will double.
Good luck! Oh, you get to pick the color of the Shuffle. :-)
The new chiashu
(2 comments)Posted on April 16, 2008 02:58
Hi everyone. I've been wanting to create a new chiashu for a while. This has become a side project I take pretty seriously, and because I went through the law school experience, I want to add as many tools as I can think of to help people who are either pre-law or law students. Some of you might know I'm working on a stealth mode start-up we hope to launch by October. chiashu kind of takes my mind off of work. It's where I do my "fun" programming.
Anyway, browse around, let me know what you think, you can leave a comment under this news item and I'll definitely read them. Suggestions VERY welcome and I'll consider them seriously as I implement more features.