hello everyone hi my name is Taryn climb and I graduated from Dartmouth in 2017 and as an undergrad I co-founded college pulse a survey research company focused on college students and in the past two years we've built the largest and most reliable database of college student opinion in the world every day tens of thousands of verified college students visit our mobile app to ask questions to other students on their campus ranging from which presidential candidate they will support in 2020 to what they plan on doing after graduation to the status of their mental health and
our app quickly aggregates the best and most popular questions from students polls the entire community and provides representative answers to these questions directly to the students at Dartmouth our first University the app spread over 85% of the student body in just three months and the past 10 months we've grown to over a quarter million college students from over 200 universities representing all 50 states making us the largest Student Survey panel and for our students college pulse isn't just a better survey taking tool it is a really sticky habit the average student has answered over 400
questions on the platform and this includes questions about their academic performance or questions about brand preferences spending habits psychographic profiles and demographic information like their university major race gender and more leveraging the largest and most representative database of student opinion with over 80 million responses our team of data scientists and survey experts which includes five Dartmouth alumni answer the most pressing questions facing organizations that care about the college demographic our projects include working with MIT to uncover underlying predictors of student depression helping u.s. senators understand voting preferences working with the knight foundation to produce a
report on the state of freedom of expression on college campuses a report that's gonna be released later this week and we'll be the first ones to admit it we've built an incredibly valuable database that provides unprecedented insights into the college demographic to organizations ranging from universities politicians think tanks corporations but the population that we think can derive the most value from our data are the students themselves because unlike other data collection platforms that extract information from their typically unknowing users to then productize and sell to organizations college pulse shares the data we collect directly with
the students after they answer our questions and we've allow them discuss the results and break it down by any demographic of their choosing in other words students use our app because we provide them with access to information that they can't find anywhere else entertaining data-driven content about their community and today I want to share with you all what went into our decision to share the data directly with students and why it's such a crucial part to our mission and to do this I'd like to begin sharing one persistent phenomenon that we've uncovered in our data
set and that phenomenon is the misperceptions that students have about their communities so in addition to asking students about their own thoughts and report their own behaviors we also regularly ask students to gauge their perceptions of the thoughts and behaviors of their community and by and large they are terrible at it students are terrible at guessing the opinions of their peers and you are probably terrible at it as well for all the non students in the audience so here are a few examples so in the largest study of student mental health we asked eighty thousand
college students from campuses all over the United States to estimate what percentage of students on your campus do you think experience depression the average estimation 30 percent on average students estimate that 30 percent of their campus has experienced depression in reality the actual reported depression levels is 65 percent so over twice as many students experience depression as the student estimation we asked 25 thousand college students how many drinks they think the average student consumes at a week the average estimation 17 drinks a week the actual reported drinking levels nine drinks a week nearly half as
many drinks as students estimate how much do you expect to make in your first job after graduation yeah and for this question we'll limit the results just to Dartmouth students so the median guess of 1,000 Dartmouth students is $80,000 and the harsh reality is that according to pay scale data Dartmouth students with zero to five years experience will earn on average sixty eight thousand five hundred all as well these are just three of the hundreds of misperceptions students have about their community that we have documented we've also found that students overestimate how much fun other
students on their campus are having while they underestimate political diversity on their campus they overestimate how much sex everyone is having and they vastly underestimate how stressed everyone is so why does it matter that students perceptions of community social norms are incorrect or at least unsubstantiated by the data well our team of data scientists have been tackling this question and what we found is that students perceptions of reality have significant effects on their own actions for instance what are the best predictors of binge drinking on college campuses after analyzing 8,000 variables in our database we
identified perception of average alcohol consumption as one of the five strongest predictors even stronger than if your parents have had problems with alcohol abuse specifically the relationship is that the more you think your peers drink the more likely you are to drink here's perhaps a less obvious one the lower your estimate of how many students on your campus are depressed the less likely you are to seek out mental health resources even when controlling for your own depression levels and the availability of mental health resources on a given campus finally the higher your estimate of how
much the average student makes in salary after graduation controlling for major the more stressed you are about finding a job in short these over estimations of problem behavior in our peers cause us to increase our own problem behaviors or say the other way underestimations a prob behaviors in our peers discourage us from engaging in the problematic behavior perceptions influence behaviors we are not the first ones to figure this out there is actually a whole theory and literature built around this in social psychology start in the 80s called social norms theory and it explores the role
peer influence has in individual decision making so what is the solution to this problem well like all great TED Talks I do not have a very well thought-out solution to this problem however I will leave you all with something far more aspirational social norms theory continues to say that if we can correct the underlying misperception of of a given perceived norm then we can change people's behaviors we can effect a decrease in the problem behavior or an increase in the desired behavior and we know this works social norm interventions are currently funded by federal agencies
foundation grants and nonprofit organizations to address a wide range of public health topics including tobacco use driving under the influence prevention seatbelt use and sexual assault prevention and some of you students might even remember being brought into this very room to be shown some college poll statistics during freshman orientation about student drinking these are successful one-off case studies of the theory and practice but a college pulse we believe that the most effective and scalable way to correct these misperceptions is to build a platform that provides a constant flow of accurate information about the topics that
students want to know about and we are trying to create a world where accurate public opinion data is accessible within every community starting with college campuses but expanding to high schools companies towns and even countries and soon we'll be able to show everyone in this room and not just students how bad you are at estimating the opinions and behaviors of your peers thank you
