planeblue

SDN Gold Donor
2+ Year Member
Feb 28, 2016
28
153
Status (Visible)
  1. Medical Student
All:

Over the past several months, I have slowly made my way through each of the school-specific threads for the 2015-2016 application cycle and logged each reported interview invitation (II) by month. The objective of this project is to provide applicants with a general idea of how many II's are left after each month of the cycle and to encourage people to apply early (48.7% of II's have been offered by the end of September). Of course, keep in mind the limitation that these trends are based on SDN self-reported data only.

The tracker.

Format:

The first sheet contains II's by month for each school. The tables provide numerical data and are color coded to match the graphs.

The second sheet contains the last reported II for each school sorted chronologically. Hopefully this is helpful to those of you who wonder in February and March if your schools are still interviewing (though this date of final II may change from year to year).

The third sheet contains a list of the schools with the earliest and latest interview cycles.

Scope:

Approximately 4229, or 5.4%, of the approximately 78,000 interview invitations (source: MSAR) that were given out last year by US allopathic schools were reported on SDN over the 132 included schools for an average of 32 II's per school.

Data:

The numbers:
upload_2017-3-5_17-10-36.png

Total II's across all schools:
upload_2017-3-5_17-9-28.png

Percent of II's remaining:
upload_2017-3-5_17-9-38.png

Limitations/notes
:

-Based on SDN self-reported II data for one cycle (2015-2016)
-Schools included are US MD schools that use AMCAS
-Does not contain EDP or MSTP/MDPhD II's
-This data should not be used to gauge how many II's are left at a given time for a certain medical school, since that data is likely too flexible from year to year to have any real meaning here. Instead, use it to determine overall trends and how early/late schools tend to interview.

Future improvements:

Feel free to PM me if you notice any errors in my data worth correcting, especially with regards to the last reported II's. I'll double check and update the google doc.

There are likely several analyses that can be performed with this data, such as the distribution of II's for "top 10" schools vs. state schools vs. "low-yield" schools, etc. Please post your analyses in this thread if you complete any. I'm sure they would be interesting for all of us to see.

Lastly, thanks to my good friend @efle for guidance.

Cheers!
 
Last edited:
  • Like
Reactions: 84 users

efle

not an elf
7+ Year Member
Apr 6, 2014
13,504
21,099
Status (Visible)
  1. Medical Student
AMAZING

Just want to say I know this dude IRL and he's been working on this forever to answer our age old question. THANKS PLANEBLUE
 
  • Like
Reactions: 24 users
About the Ads

efle

not an elf
7+ Year Member
Apr 6, 2014
13,504
21,099
Status (Visible)
  1. Medical Student
This image in particular I think is going to help many people fight their urge to procrastinate in the upcoming cycles. Waiting to hit complete until late Sept/early Oct cuts the seats remaining in half!!
 
  • Like
Reactions: 4 users

Hospitalized

5+ Year Member
Sep 9, 2014
1,920
2,651
Status (Visible)
  1. Medical Student (Accepted)
Only 5% of all allopathic interviews received were reported on SDN. That's such a small amount...
 
  • Like
Reactions: 2 users

efle

not an elf
7+ Year Member
Apr 6, 2014
13,504
21,099
Status (Visible)
  1. Medical Student
Only 5% of all allopathic interviews received were reported on SDN. That's such a small amount...
I would guess only a minority of SDN readers have active accounts and post often, let alone posting their interview invites. Luckily the population of applicants is so large every year that even a few percent gives you thousands of datapoints
 
  • Like
Reactions: 1 users

Lucca

Will Walk Rope for Sandwich
Staff member
Administrator
Volunteer Staff
7+ Year Member
Oct 22, 2013
8,477
19,321
City of the Future
Status (Visible)
  1. MD/PhD Student
wow this must have been an extremely time consuming effort, thanks for the data!

I've always wondered if the trend captured in the WashU MSTP interview data would line up with the trend in the general MD pool.



(from the Wash U MSTP website)

It looks like the timeline for interview slot depletion is slightly faster in the general MD cycle (from your data). The 50% mark is hit in about early Oct for MSTP and hits by the end of September.

Great stuff! @Lawper add it to the sticky
 
Last edited:

libertyyne

5+ Year Member
Mar 5, 2015
10,957
22,430
Status (Visible)
  1. Medical Student
Excellent work.
How did you mine the data?
 
  • Like
Reactions: 1 user

Gurby

5+ Year Member
Jul 28, 2014
1,977
5,594
Status (Visible)
  1. Medical Student
Excellent work.
How did you mine the data?

Very interested in this too. Did you literally scroll through every single thread and control-F for "II" and manually update a spreadsheet, or did you write a clever script somehow?
 

Huggy

2+ Year Member
Apr 24, 2016
949
2,522
Status (Visible)
  1. Medical Student
This really is amazing.

Another potential source of error: Every school specific thread that I've been involved in gets significantly more and more quiet as the year progresses which means less postings in general (including IIs received). I feel that many members feel more inclined to report that they received an II when the last post was from 20 mins ago rather than when it was from 10 days ago... nonetheless, probably not a significant factor
 
  • Like
Reactions: 1 users

planeblue

SDN Gold Donor
2+ Year Member
Feb 28, 2016
28
153
Status (Visible)
  1. Medical Student
Very interested in this too. Did you literally scroll through every single thread and control-F for "II" and manually update a spreadsheet, or did you write a clever script somehow?

Not sure if a script would have worked here, especially given that there's great variation in how people announce their II's, but I have very little programming experience. I saw a HUGE range of things from "II!" to "interview invite" to "I have the great privilege to interview here next month," and even "i i" (with a space). I tried ctrl+F for the first couple threads for "II" and "interview," but a lot of people have signatures that include those, so it was much too confusing for me to follow. Basically, I just scrolled through the threads and kept an eye out. After a while, it became much easier since I knew what I was looking for, and the first few and last few pages of most threads didn't have any IIs. Wouldn't wish the 74 page VCU thread on anyone, though.

@Huggy totally agree. Definitely a valid source of error.
 
  • Like
Reactions: 13 users

libertyyne

5+ Year Member
Mar 5, 2015
10,957
22,430
Status (Visible)
  1. Medical Student
Not sure if a script would have worked here, especially given that there's great variation in how people announce their II's, but I have very little programming experience. I saw a HUGE range of things from "II!" to "interview invite" to "I have the great privilege to interview here next month," and even "i i" (with a space). I tried ctrl+F for the first couple threads for "II" and "interview," but a lot of people have signatures that include those, so it was much too confusing for me to follow. Basically, I just scrolled through the threads and kept an eye out. After a while, it became much easier since I knew what I was looking for, and the first few and last few pages of most threads didn't have any IIs. Wouldn't wish the 74 page VCU thread on anyone, though.

@Huggy totally agree. Definitely a valid source of error.
Hats off to you for slogging through that.
 
  • Like
Reactions: 1 user
About the Ads

libertyyne

5+ Year Member
Mar 5, 2015
10,957
22,430
Status (Visible)
  1. Medical Student
This really is amazing.

Another potential source of error: Every school specific thread that I've been involved in gets significantly more and more quiet as the year progresses which means less postings in general (including IIs received). I feel that many members feel more inclined to report that they received an II when the last post was from 20 mins ago rather than when it was from 10 days ago... nonetheless, probably not a significant factor
There is also a liklihood that applicants that are more inclined to post stuff on SDN have submitted their app early in the season and receive II earlier.
Another source might be decrease in general participation as the season goes on. I know I was quick to share II information early in the season, once I had received an acceptance i got lazy. There are 4 IIs that I failed to post on the school specific boards.
 
Last edited:
  • Like
Reactions: 5 users

efle

not an elf
7+ Year Member
Apr 6, 2014
13,504
21,099
Status (Visible)
  1. Medical Student
One last thing to think about is that the data are likely sets, with each SDN user that reports likely reporting several across the various school threads.
 
  • Like
Reactions: 1 user

JustAPhD

Not a hummingbird expert
5+ Year Member
Jan 5, 2016
3,472
8,831
Ice Coast
This really is amazing.

Another potential source of error: Every school specific thread that I've been involved in gets significantly more and more quiet as the year progresses which means less postings in general (including IIs received). I feel that many members feel more inclined to report that they received an II when the last post was from 20 mins ago rather than when it was from 10 days ago... nonetheless, probably not a significant factor

No that's actually a very possible source of error, and perhaps a fairly significant one. As @libertyyne said, I think it's very common for people to post their early IIs (when they're likely most excited about receiving them) and then get "lazy" as the cycle progresses. Or to post IIs at schools near the top of their list but not post the IIs for lower-ranked schools, where they might not be as excited for.

Is the data perfect? No, but OP worked with what is available. My hat is off to you, @planeblue, great work.
 
  • Like
Reactions: 1 user

efle

not an elf
7+ Year Member
Apr 6, 2014
13,504
21,099
Status (Visible)
  1. Medical Student
I think a lot of people get their exciting II's later, too. Helps balance out. Like if you've had silence since August and then get some interviews in Jan/Feb you are def gonna be excited enough
 
  • Like
Reactions: 1 user

ready2go2

7+ Year Member
Mar 29, 2013
390
701
Status (Visible)
  1. Medical Student
The objective of this project is to provide applicants with a general idea of how many II's are left after each month of the cycle and to encourage people to apply early (48.7% of II's have been offered by the end of September).

First I would like to say nice work! I am sure it took you a long time to bring it all together and I appreciate your effort. I have to say that the quoted objective was not reached though. I don't believe for one second that this dataset and your analysis will be helpful for the average applicant. SDN is full of extremely high quality applicants who a) are for the most part above the median for applicants (dare I say matriculants too?) for both GPA and MCAT and b) who are much more likely to apply early in the cycle. Taken together, that means people who report on SDN are more likely to get their interview invites first and early (before the end of September). Moreover it is an extremely small dataset (5%). I don't think anyone would for a second argue this dataset is representative of the reality.
 

efle

not an elf
7+ Year Member
Apr 6, 2014
13,504
21,099
Status (Visible)
  1. Medical Student
First I would like to say nice work! I am sure it took you a long time to bring it all together and I appreciate your effort. I have to say that the quoted objective was not reached though. I don't believe for one second that this dataset and your analysis will be helpful for the average applicant. SDN is full of extremely high quality applicants who a) are for the most part above the median for applicants (dare I say matriculants too?) for both GPA and MCAT and b) who are much more likely to apply early in the cycle. Taken together, that means people who report on SDN are more likely to get their interview invites first and early (before the end of September). Moreover it is an extremely small dataset (5%). I don't think anyone would for a second argue this dataset is representative of the reality.
I'd argue the 5% part is not a real problem (think about what percent of the voting population gets called for the typical political survey).

And I'd argue the SDN bias is cancelled out by the data being presented on SDN. That is, the crowd that reads this should be similar to the crowd that the data was taken from, though neither may be similar to the national premed population.
 
  • Like
Reactions: 1 user

lpp06

7+ Year Member
Sep 17, 2012
1,104
2,049
Status (Visible)
  1. Medical Student (Accepted)
I'd argue the 5% part is not a real problem (think about what percent of the voting population gets called for the typical political survey).

And I'd argue the SDN bias is cancelled out by the data being presented on SDN. That is, the crowd that reads this should be similar to the crowd that the data was taken from, though neither may be similar to the national premed population.


Has anyone tried to quantify the SDN "bias"?
 
  • Like
Reactions: 1 user

walloobi

5+ Year Member
Oct 30, 2014
2,046
3,358
The capital of Maine, which is Montpelier Vermont
Status (Visible)
  1. Medical Student
I'd argue the 5% part is not a real problem (think about what percent of the voting population gets called for the typical political survey).

And I'd argue the SDN bias is cancelled out by the data being presented on SDN. That is, the crowd that reads this should be similar to the crowd that the data was taken from, though neither may be similar to the national premed population.
That didn't turn out too well with this last presidential election haha
 
  • Like
Reactions: 2 users
About the Ads

libertyyne

5+ Year Member
Mar 5, 2015
10,957
22,430
Status (Visible)
  1. Medical Student
I don't think so, and I can't even begin to imagine a good approach to do that.
Maybe write a script to go through the wamc forum and collect gpa /MCAT info and compare to the national applicant pool.
 
  • Like
Reactions: 1 user

lpp06

7+ Year Member
Sep 17, 2012
1,104
2,049
Status (Visible)
  1. Medical Student (Accepted)
I don't think so, and I can't even begin to imagine a good approach to do that.

Could you run a survey collecting GPA/MCAT for simple comparison to national mean/medians?
 

JustAPhD

Not a hummingbird expert
5+ Year Member
Jan 5, 2016
3,472
8,831
Ice Coast
I'd argue the 5% part is not a real problem (think about what percent of the voting population gets called for the typical political survey).

And I'd argue the SDN bias is cancelled out by the data being presented on SDN. That is, the crowd that reads this should be similar to the crowd that the data was taken from, though neither may be similar to the national premed population.

Right, but a political survey utilizes a random sampling to reflect the broader population. It can be argued that SDN is not a great random sampling of the general medical applicant pool. Your second part I'd generally agree with.
 
  • Like
Reactions: 1 user

efle

not an elf
7+ Year Member
Apr 6, 2014
13,504
21,099
Status (Visible)
  1. Medical Student
That didn't turn out too well with this last presidential election haha
Didn't they decide the numbers were all off because so many people were planning to vote Trump in the privacy of the booth but wouldn't admit it in-person
 

efle

not an elf
7+ Year Member
Apr 6, 2014
13,504
21,099
Status (Visible)
  1. Medical Student
Right, but a political survey utilizes a random sampling to reflect the broader population. It can be argued that SDN is not a great random sampling of the general medical applicant pool. Your second part I'd generally agree with.
Oh I think the two points go together. It's a good random sampling of SDN premeds and a great tool for other SDN premeds.
 
  • Like
Reactions: 2 users

efle

not an elf
7+ Year Member
Apr 6, 2014
13,504
21,099
Status (Visible)
  1. Medical Student
Maybe write a script to go through the wamc forum and collect gpa /MCAT info and compare to the national applicant pool.
Could you run a survey collecting GPA/MCAT for simple comparison to national mean/medians?
I think WAMC is not very representative of SDN at large though, with WAMC having a lot more borderline people worried about their odds and trying to make the best possible school list.
 

libertyyne

5+ Year Member
Mar 5, 2015
10,957
22,430
Status (Visible)
  1. Medical Student
I think WAMC is not very representative of SDN at large though, with WAMC having a lot more borderline people worried about their odds and trying to make the best possible school list.
I feel like that would make it so that people who read sdn are represented better. It seems like the posters who post the most also have MCATs in the top few percentiles.

I can think of at least 20 posters with top 3 percentile MCATs.
 
  • Like
Reactions: 1 user

JustAPhD

Not a hummingbird expert
5+ Year Member
Jan 5, 2016
3,472
8,831
Ice Coast
Oh I think the two points go together. It's a good random sampling of SDN premeds and a great tool for other SDN premeds.

I think because of how the data was gathered the only "correct" way to interpret the results is the bolded above. But, I will say that that doesn't mean the take-home message isn't useful for the entirety of applicants: apply early!
 
  • Like
Reactions: 2 users
About the Ads

walloobi

5+ Year Member
Oct 30, 2014
2,046
3,358
The capital of Maine, which is Montpelier Vermont
Status (Visible)
  1. Medical Student
All:

Over the past several months, I have slowly made my way through each of the school-specific threads for the 2015-2016 application cycle and logged each reported interview invitation (II) by month. The objective of this project is to provide applicants with a general idea of how many II's are left after each month of the cycle and to encourage people to apply early (48.7% of II's have been offered by the end of September). Of course, keep in mind the limitation that these trends are based on SDN self-reported data only.

The tracker.

Format:

The first sheet contains II's by month for each school. The tables provide numerical data and are color coded to match the graphs.

The second sheet contains the last reported II for each school sorted chronologically. Hopefully this is helpful to those of you who wonder in February and March if your schools are still interviewing (though this date of final II may change from year to year).

The third sheet contains a list of the schools with the earliest and latest interview cycles.

Scope:

Approximately 4229, or 5.4%, of the approximately 78,000 interview invitations (source: MSAR) that were given out last year by US allopathic schools were reported on SDN over the 132 included schools for an average of 32 II's per school.

Data:

The numbers:
View attachment 215786

Total II's across all schools:
View attachment 215784

Percent of II's remaining:
View attachment 215785

Limitations/notes
:

-Based on SDN self-reported II data for one cycle (2015-2016)
-Schools included are US MD schools that use AMCAS
-Does not contain EDP or MSTP/MDPhD II's
-This data should not be used to gauge how many II's are left at a given time for a certain medical school, since that data is likely too flexible from year to year to have any real meaning here. Instead, use it to determine overall trends and how early/late schools tend to interview.

Future improvements:

Feel free to PM me if you notice any errors in my data worth correcting, especially with regards to the last reported II's. I'll double check and update the google doc.

There are likely several analyses that can be performed with this data, such as the distribution of II's for "top 10" schools vs. state schools vs. "low-yield" schools, etc. Please post your analyses in this thread if you complete any. I'm sure they would be interesting for all of us to see.

Lastly, thanks to my good friend @efle for guidance.

Cheers!
This is great work, but I have one major problem with it: without data showing what percent of applicants have applied by any given month, this data doesn't actually suggest that it's important to apply early. In order to conclude that the chances of receiving an II decrease over time, you'd have to show that the rate of decrease in remaining II's is greater than the rate of decrease in applications being submitted/reviewed. If the number of remaining II's decreases proportionally to the number of remaining applications that have yet to be submitted, then we can't conclude that application date is of any relevance.

Again, this is great work. We should just be careful about how we interpret the results, sampling bias aside.
 

efle

not an elf
7+ Year Member
Apr 6, 2014
13,504
21,099
Status (Visible)
  1. Medical Student
This is great work, but I have one major problem with it: without data showing what percent of applicants have applied by any given month, this data doesn't actually suggest that it's important to apply early. In order to conclude that the chances of receiving an II decrease over time, you'd have to show that the rate of decrease in remaining II's is greater than the rate of decrease in applications being submitted/reviewed. If the number of remaining II's decreases proportionally to the number of remaining applications that have yet to be submitted, then we can't conclude that application date is of any relevance.

Again, this is great work. We should just be careful about how we interpret the results, sampling bias aside.
I think the verification timing graph is pretty good evidence that the vast majority of submissions are transmitted by end of sept or oct.

Plus there are deadlines for a lot of the schools in the middle of the cycle, after which no more are being taken, yet there is no disruption in the trend at all.

I'd think it is a non issue myself
 

walloobi

5+ Year Member
Oct 30, 2014
2,046
3,358
The capital of Maine, which is Montpelier Vermont
Status (Visible)
  1. Medical Student
I think the verification timing graph is pretty good evidence that the vast majority of submissions are transmitted by end of sept or oct.

Plus there are deadlines for a lot of the schools in the middle of the cycle, after which no more are being taken, yet there is no disruption in the trend at all.

I'd think it is a non issue myself
That graph shows how many days it takes to get verified, right? I'm not so sure we can infer very accurately how that relates to overall % of apps submitted. Even if it is an accurate representation, there are relatively consistent numbers of apps being submitted throughout Sept, Oct, and Nov, with a slight decrease in Dec, so it doesn't look like application deadlines have much of an impact on slowing down the number of apps being submitted until Dec. That's not to mention that there are a ton of complicated variables affecting the amount of time between verification and II, which really makes it impossible to compare the percent of II's remaining to the amount of time it takes to get verified. It all starts to get far too mucky.

Throughout the cycle, we will always see a decrease in remaining apps to be submitted, and a decrease in II's remaining. I'm just saying that without better data with which to contextualize the (awesome) graphs in the OP, we can't even come close to quantifying the importance of applying early, and we can barely even make qualitative conclusions about it either.
 
  • Like
Reactions: 1 user

efle

not an elf
7+ Year Member
Apr 6, 2014
13,504
21,099
Status (Visible)
  1. Medical Student
That graph shows how many days it takes to get verified, right? I'm not so sure we can infer very accurately how that relates to overall % of apps submitted. Even if it is an accurate representation, there are relatively consistent numbers of apps being submitted throughout Sept, Oct, and Nov, with a slight decrease in Dec, so it doesn't look like application deadlines have much of an impact on slowing down the number of apps being submitted until Dec. That's not to mention that there are a ton of complicated variables affecting the amount of time between verification and II, which really makes it impossible to compare the percent of II's remaining to the amount of time it takes to get verified. It all starts to get far too mucky.

Throughout the cycle, we will always see a decrease in remaining apps to be submitted, and a decrease in II's remaining. I'm just saying that without better data with which to contextualize the (awesome) graphs in the OP, we can't even come close to quantifying the importance of applying early, and we can barely even make qualitative conclusions about it either.
Yes it's the time to verify, and when you have a minimum of 3d and a 400-500% spike for a couple months before going back down to 3-5d you can say at least qualitatively that the vast, vast majority of apps have been transmitted by end of Sept.

These kinds of things might make sense to think about if we needed to scrounge up a reason why applying late isn't a disadvantage. But given the qualitative sense that early is advantageous (from both applicants and admissions perspectives on these boards) and then we see an uninterrupted trend of higher fractions earlier, it is good evidence imo.

This is also not considering that the bottleneck is often on the admissions side. They are going to go through review at a pretty steady rate regardless of the transmission curve, where things back up a bunch early in the season and then they play catch-up in the later months. If a school reviews in order of LizzyM then yeah you'd expect interviews to go out faster at first. But if they review in order of receipt, then the interpretation I'd have is that they review a similar number of apps in Aug/Sept and Dec/Jan but the former cohort gets invited at a higher rate.
 
  • Like
Reactions: 1 user

lpp06

7+ Year Member
Sep 17, 2012
1,104
2,049
Status (Visible)
  1. Medical Student (Accepted)
I agree with the above - there's no advantage to not applying early.

Also, this isn't a measurable - but I'd imagine that reviewers get progressively more exhausted and desensitized as the cycle gets older. I doubt that the 1,000th app that gets read is given as much attention as the 1st and I'd imagine that after recommending 100 3.9/35's for interview, the comparative standard gets raised*.


*I'm aware that reviewers are supposed to consider each application as a distinct entity, but there must be some subconscious comparisons that are made.
 
  • Like
Reactions: 1 user

walloobi

5+ Year Member
Oct 30, 2014
2,046
3,358
The capital of Maine, which is Montpelier Vermont
Status (Visible)
  1. Medical Student
Yes it's the time to verify, and when you have a minimum of 3d and a 400-500% spike for a couple months before going back down to 3-5d you can say at least qualitatively that the vast, vast majority of apps have been transmitted by end of Sept.

These kinds of things might make sense to think about if we needed to scrounge up a reason why applying late isn't a disadvantage. But given the qualitative sense that early is advantageous (from both applicants and admissions perspectives on these boards) and then we see an uninterrupted trend of higher fractions earlier, it is good evidence imo.

This is also not considering that the bottleneck is often on the admissions side. They are going to go through review at a pretty steady rate regardless of the transmission curve, where things back up a bunch early in the season and then they play catch-up in the later months. If a school reviews in order of LizzyM then yeah you'd expect interviews to go out faster at first. But if they review in order of receipt, then the interpretation I'd have is that they review a similar number of apps in Aug/Sept and Dec/Jan but the former cohort gets invited at a higher rate.
I don't think we can say that the "vast, vast majority of apps have been transmitted by the end of Sept," it looks like about 70% of apps have been transmitted by that point (we can roughly double the # of days to verification in order to approximate % of apps submitted any given month, if we want to trust that relationship on that graph). So by the end of Sept, only about 30% of apps are still remaining to be sent, and there are still 50% of II's left to be sent out. Doesn't sound too bad to me. In fact, the more apps that have been sent by the end of Sept, the better it looks for late applicants; if only 5% of people haven't yet sent their apps by the end of Sept, and there are still 50% of II's remaining, then it doesn't look like those 5% have anything to worry about. I could argue that later submissions have the advantage because the number of remaining II's is relatively larger than the number of remaining apps with which they're being compared; obviously I don't actually think that's the case, and late submitters aren't only being compared to other late submitters (unless admissions departments review apps very quickly and don't have to play catch-up), but I'm just playing devil's advocate to point out that we need to be careful about over-interpreting any of this data.

We see a higher fraction of remaining II's earlier by definition; we can only say the earlier cohorts are invited to interviews at a higher rate if we see a relatively higher fraction of II's earlier in the cycle.

I don't know enough about admissions dept bottlenecks or app review rates to try to make any inferences there, you very well may be right that they review similar numbers of apps in any given month.
 

efle

not an elf
7+ Year Member
Apr 6, 2014
13,504
21,099
Status (Visible)
  1. Medical Student
I don't think we can say that the "vast, vast majority of apps have been transmitted by the end of Sept," it looks like about 70% of apps have been transmitted by that point (we can roughly double the # of days to verification in order to approximate % of apps submitted any given month, if we want to trust that relationship on that graph). So by the end of Sept, only about 30% of apps are still remaining to be sent, and there are still 50% of II's left to be sent out. Doesn't sound too bad to me. In fact, the more apps that have been sent by the end of Sept, the better it looks for late applicants; if only 5% of people haven't yet sent their apps by the end of Sept, and there are still 50% of II's remaining, then it doesn't look like those 5% have anything to worry about. I could argue that later submissions have the advantage because the number of remaining II's is relatively larger than the number of remaining apps with which they're being compared; obviously I don't actually think that's the case, and late submitters aren't only being compared to other late submitters (unless admissions departments review apps very quickly and don't have to play catch-up), but I'm just playing devil's advocate to point out that we need to be careful about over-interpreting any of this data.

We see a higher fraction of remaining II's earlier by definition; we can only say the earlier cohorts are invited to interviews at a higher rate if we see a relatively higher fraction of II's earlier in the cycle.

I don't know enough about admissions dept bottlenecks or app review rates to try to make any inferences there, you very well may be right that they review similar numbers of apps in any given month.
Maybe 30% of apps have yet to be submitted for the last 50% of spots (and I think that is the wrong way to read the graph to begin with), but I bet far more than 50% of apps are still awaiting review when the halfway mark is passed. I was hella late on some of mine (like, end of Oct) and rather than a nice quick turnaround like your first paragraph would predict, I didn't get the interview invite until several months later. Plus yeah, it's not a single pass system, the pool of "maybe" apps is going to grow as the cycle goes on.
 

Gurkhali

2+ Year Member
Jun 15, 2016
680
1,460
Status (Visible)
  1. Medical Student (Accepted)
I don't think we can say that the "vast, vast majority of apps have been transmitted by the end of Sept," it looks like about 70% of apps have been transmitted by that point (we can roughly double the # of days to verification in order to approximate % of apps submitted any given month, if we want to trust that relationship on that graph). So by the end of Sept, only about 30% of apps are still remaining to be sent, and there are still 50% of II's left to be sent out. Doesn't sound too bad to me. In fact, the more apps that have been sent by the end of Sept, the better it looks for late applicants; if only 5% of people haven't yet sent their apps by the end of Sept, and there are still 50% of II's remaining, then it doesn't look like those 5% have anything to worry about. I could argue that later submissions have the advantage because the number of remaining II's is relatively larger than the number of remaining apps with which they're being compared; obviously I don't actually think that's the case, and late submitters aren't only being compared to other late submitters (unless admissions departments review apps very quickly and don't have to play catch-up), but I'm just playing devil's advocate to point out that we need to be careful about over-interpreting any of this data.

We see a higher fraction of remaining II's earlier by definition; we can only say the earlier cohorts are invited to interviews at a higher rate if we see a relatively higher fraction of II's earlier in the cycle.

I don't know enough about admissions dept bottlenecks or app review rates to try to make any inferences there, you very well may be right that they review similar numbers of apps in any given month.
Considering that most admissions offices/committees seem to be running on a constrained time schedule and resources (mostly staff), coordinating the pre-II review process with the actual interviews might be done on a batch-by-batch basis. It doesn't really seem like the most efficient process and they clearly take quite a bit of time spreading out interviews and reviews.

As far as the benefit of applying early, I think it can only help you. Applying later can only hurt you or have no effect. After the first month of August, according to the imperfect but beautiful data @planeblue has gathered, 27% of IIs are gone. Dunno how many applicants are reviewed (i.e. invited, rejected or temporary hold) at that point but I think the benefit of being reviewed earlier is real. I really doubt 27% of applicants have been 'reviewed' by the end of August and so the remaining % IIs is much smaller than the % of applicants remaining+left to apply.
 
  • Like
Reactions: 1 user

walloobi

5+ Year Member
Oct 30, 2014
2,046
3,358
The capital of Maine, which is Montpelier Vermont
Status (Visible)
  1. Medical Student
Maybe 30% of apps have yet to be submitted for the last 50% of spots (and I think that is the wrong way to read the graph to begin with), but I bet far more than 50% of apps are still awaiting review when the halfway mark is passed. I was hella late on some of mine (like, end of Oct) and rather than a nice quick turnaround like your first paragraph would predict, I didn't get the interview invite until several months later. Plus yeah, it's not a single pass system, the pool of "maybe" apps is going to grow as the cycle goes on.
Yeah I bet more than 50% of apps are still awaiting review by that point too, and I very well may be reading the graph wrong, but that's my whole point - we don't have enough info to do anything more than "bet" or wildly extrapolate; it's a very weak graph (the verification time one, not the ones in the OP), with 7 data points, grossly inadequate axes, and we don't have enough behind-the-scenes info about how admissions work with respect to bottlenecks or second/third passes over apps or reviewing strong applicants early vs chronological review, etc etc. I'm not saying the data in the OP isn't valuable, I'm just saying that the actual effect of applying early vs late can't be readily ascertained by looking at the graphs of remaining II's. Some people have been making unfounded inferences based on the OP, so I'm just saying we should be cautious.
 

walloobi

5+ Year Member
Oct 30, 2014
2,046
3,358
The capital of Maine, which is Montpelier Vermont
Status (Visible)
  1. Medical Student
Considering that most admissions offices/committees seem to be running on a constrained time schedule and resources (mostly staff), coordinating the pre-II review process with the actual interviews might be done on a batch-by-batch basis. It doesn't really seem like the most efficient process and they clearly take quite a bit of time spreading out interviews and reviews.

As far as the benefit of applying early, I think it can only help you. Applying later can only hurt you or have no effect. After the first month of August, according to the imperfect but beautiful data @planeblue has gathered, 27% of IIs are gone. Dunno how many applicants are reviewed (i.e. invited, rejected or temporary hold) at that point but I think the benefit of being reviewed earlier is real. I really doubt 27% of applicants have been 'reviewed' by the end of August and so the remaining % IIs is much smaller than the % of applicants remaining+left to apply.
But this distinction (hurtful vs. no effect) is incredibly important. A lot of people push their application back a whole year because they need another month to study for the MCAT and don't want to apply in August or September because they think it'll significantly decrease their chances of acceptance. Based on the data in this thread, we can't conclude that pushing their app back a whole year is a good idea for that reason. We can suggest that based on advice from adcoms on SDN, but not from the graphs in the OP.
 
  • Like
Reactions: 1 users

Gurkhali

2+ Year Member
Jun 15, 2016
680
1,460
Status (Visible)
  1. Medical Student (Accepted)
But this distinction (hurtful vs. no effect) is incredibly important. A lot of people push their application back a whole year because they need another month to study for the MCAT and don't want to apply in August or September because they think it'll significantly decrease their chances of acceptance. Based on the data in this thread, we can't conclude that pushing their app back a whole year is a good idea for that reason. We can suggest that based on advice from adcoms on SDN, but not from the graphs in the OP.
Yes, I agree. Strong candidates will get interviews well into the cycle and the large amount of interviews given out early in the cycle is probably an indication of that. I don't think they are less jaded at that point but just there are a large number of strong applicants applying early on.
 
  • Like
Reactions: 1 user

planeblue

SDN Gold Donor
2+ Year Member
Feb 28, 2016
28
153
Status (Visible)
  1. Medical Student
I agree with the vast majority of critiques mentioned above. I think it's fair to say that this can be used however (and if) it benefits you. No matter who you are, I think at least seeing the date of the last II will be useful for future applicants at a bare minimum, regardless of whether or not you take stock in the theoretical percents of II's remaining. Like others have said, I just worked with what was available and in the off chance that a future applicant sees this and says, "I should really get my app in asap," I'll be pretty happy.
 
  • Like
Reactions: 5 users

efle

not an elf
7+ Year Member
Apr 6, 2014
13,504
21,099
Status (Visible)
  1. Medical Student
Well planeblue I disagree with most of the critiques and think you captured a very real, if exaggerated in this data, phenomenon.

It's staying in my damn sig
 
  • Like
Reactions: 1 user

planeblue

SDN Gold Donor
2+ Year Member
Feb 28, 2016
28
153
Status (Visible)
  1. Medical Student
Well planeblue I disagree with most of the critiques and think you captured a very real, if exaggerated in this data, phenomenon.

It's staying in my damn sig

Maybe I should have said "I recognize most critiques and understand where most of the posters are coming from."
 
  • Like
Reactions: 1 user

efle

not an elf
7+ Year Member
Apr 6, 2014
13,504
21,099
Status (Visible)
  1. Medical Student
I'm bumping this thread in the hopes that it scares a few procrastinators out there into finishing their submission ASAP.
 
  • Like
Reactions: 3 users
About the Ads
This thread is more than 3 years old.

Your message may be considered spam for the following reasons:

  1. Your new thread title is very short, and likely is unhelpful.
  2. Your reply is very short and likely does not add anything to the thread.
  3. Your reply is very long and likely does not add anything to the thread.
  4. It is very likely that it does not need any further discussion and thus bumping it serves no purpose.
  5. Your message is mostly quotes or spoilers.
  6. Your reply has occurred very quickly after a previous reply and likely does not add anything to the thread.
  7. This thread is locked.