Biostatistics Information Session

Biostatistics Information Session

- Okay, so, welcome everyone. We're so happy that you're here. So, I think, so we can be very informal, we're a small group. So, what I'd like to do, is just start, and just quickly go around and have people introduce themselves and just say what you're currently doing and I guess, if you're here because you're a prospective applicant. So, we'll skip us, we're the faculty, because we'll get to us soon, but we'll start with all of you.

- [Agero] Hi, my name is Agero or Agi, I'm a second-year biostats student.

- [Rebecca] Anything else you wanna say.

- [Sandy] Hi my name is Sandy, I'm currently working as a coordinator at a health agency in the financial district area. Just interviewing patients and see how their, it's more like quality assurance, basically.

- [Rebecca] Great, thank you, welcome.

- [Uray] Hi, I'm Uray. I'm a first-year grad student here at NYU, and I just switched my concentration from global health to biostats.

- [Rebecca] Great, thank you.

- 'Cause we're awesome.

- [Dennis] My name is Dennis, and I'm also a second-year biostats student.

- [Sally] I'm Sally. I'm a senior majoring in public health, and yup.

- And sociology.

- [Sally] And sociology, and I'm minoring in math.

- Oh cool.

- [Sindu] I'm Sindu, I'm a senior undergrad majoring in public health and biology.

- Okay, so, welcome everyone. So, I am Rebecca Betensky. I am the new chair of biostatistics here at the College of Global Public Health. And, so, we wanted to have this event to sort of try to introduce ourselves to anyone who might be in the area and interested in biostatistics and we wanna tell you everything we can in a short amount of time about our field, what we do, what people do who are bio-statisticians, and what our program is all about. And then, you'll hear from Zach, who's standing back there, who is the head of our admissions office and he can tell you a lot more about details of applying. And then, as you heard, we have current students here who can also talk to you about their experiences. So, we're gonna keep this really informal, so feel free to interrupt. So, the plan is, we're gonna start. The five of us will just introduce ourselves, tell you like a minute about our research and our projects that we work on and our teaching and educational activities. And then Melody will talk about our MPH program and our soon-to-be-approved MS program. And then Zach, as I said, will talk about admissions and financial aid, okay? All right, so, this is me. And my background. I was a math major undergraduate, and then my PhD is in statistics. One thing that came up in discussion is differences or not differences between statistics and biostatistics, so we can talk about that. And, I can tell you, that nowadays there's very little difference, other than the focus of the applications. Back in the day when I was getting my PhD, there were bigger differences. But now, not so much. And I just, as I mentioned, I just started here. I was at Harvard School of Public Health for 24 years. And I just started here in October and I'm really excited to be here. It's a new school. The school is three years old. And, of course the new department. And we're growing and we're increasing our faculty and our students, and building up our course offerings. So, we're in a growth mode in every respect, so it's a really exciting time to be here. So, just a little tiny background on my research. So, my major research area is in statistical methodology in an area called survival analysis. Some of you might be familiar with it. Anybody, you? No. Okay, so, survival analysis is just referring to a certain kind of data structure in which there are incomplete observations. So, we don't get to observe every subject until they have their event. Sometimes they get cut off because the study ends. We still wanna use that information, that's what survival analysis is. So, that's the major area that I work in. I'm actually teaching a new course called Applied Survival Analysis in the spring, so I'm excited about that. Clinical trials is another area that I've worked on, going back to my dissertation. So, I've worked on methods for the design of innovative and efficient clinical trials. I'm always interested in issues of statistical inference. So, we just had a journal club for our students a couple weeks ago. We read the American Statistical Association's statement on p values, which are really important in doing statistical analysis, and what does that all mean? So, that's another area that I like to think about and write about every now and then. And the collaborative work that I do has been in many different areas in medicine, and mainly in Alzheimer's disease recently. So, I've worked in a lot of different contexts in Alzheimer's disease, and a lot of the work that I've done in survival analysis has been motivated by studies in Alzheimer's disease. So, that's the way everything ties together. These are just different projects that I've worked on. So, again, just supporting the survival analysis research, the Alzheimer's research. And, again, they're all tied together. So, this last project is called Treatment of Randomly Censored Covariates in Alzheimer's Disease Studies. So that's combining different complexities in regression analysis that arise due to incomplete observations that I first became aware of in my collaborations in Alzheimer's disease. So, teaching, I told you about the survival analysis class that I'm hoping that I will be able to teach. We just started a journal club for our students. And the students, and Dennis can speak more to this, because he's leading the journal club and Friday, the students seemed very interested in that. So, we're working to engage our students in all kinds of learning, whether it's for credit or not for credit. But learning the field and really being engaged and involved in reading and talking about statistical issues. And, let's see, seminar. So, let me skip to the next one. So, another thing that I do, is I'm the statistical editor for a major clinical journal in neurology that's called Annals of Neurology. And, so, that's another activity that I engage in and it makes me, again, aware of statistical complexities, which, to statisticians who do research, are what is very interesting and what we like to work on. And always, I've always been very, very involved and engaged in outreach to students from diverse and underrepresented backgrounds. That's a really high priority of mine and will be and is of our department here. Okay, so, that's me, and we'll talk more. Let me turn it over to Melody.

我是Rebecca Betensky，环球公共卫生学院生物统计系的新任系主任。在此我们利用这次活动向对生物统计学感兴趣的各位介绍我们。我们想在短时间内介绍我们的学科领域，我们系的教授以及其他生物统计学家都做些什么，还有我们生物统计系以及开设的专业方向。

。。。

我们系的五位教授将每人做自我介绍，用一分钟介绍各自的研究方向，科研课题项目以及教学活动。Melody将介绍我们的公共卫生硕士方向以及即将开设的理学硕士方向（备注：此方向已经开设）。

。。。

好吧，先说说我的教育背景。我本科专业是数学，博士专业是统计。

人们常问到统计学和生物统计学之间的异同。我可以告诉你们，如今，他们的区别非常小，但生物统计学更偏重应用。在我的博士学习阶段，两者有着更大些的差异。而如今， 差异小多了。

我在哈佛大学（工作）24年。在纽约大学的工作是从2018年10月份才刚刚开始。我来到这儿工作很兴奋，环球公共卫生学院是一所新的学院，才成立了三年，生物统计系也是一个新成立不久的系。我们正在成长。我们的人员，包括教师和学生，正在（逐年）增加，我们在逐步开设自己的课程。所以，从各方面来说，我们都处于成长模式，所以这是一个非常激动人心的时刻。

接下来我简单介绍一下我的研究方向。我的科研主要集中于一个叫“生存分析”的统计方法。 生存分析指某种数据结构，它存在不完整的观察数据。（比如，）我们未必总是能明确观察记录到每个研究对象的生存时间，有时因为课题结束了，我们对被试的观测被打断。即便如此，我们希望能使用这些（不完整）的信息，所以我们使用生存分析。这就是我主要研究的领域。我在（2019年）春季学期开设了一门名为“实用生存分析”的新课程。我为此感到很兴奋。早在我的博士论文阶段，我就开始从事临床试验的研究。

一直以来，我致力于设计创新和有效的临床试验的方法。我也一直对统计推断问题感兴趣。所以，几周前我们刚给学生们开设了一个文献精读俱乐部。我们阅读了美国统计协会关于p值的声明，这对于进行统计分析非常重要，又究竟有着怎样的意义呢？这是我喜欢思考这些问题和并时常就此问题撰写论文。我在医学的许多不同领域都有合作项目，最近主要在阿尔茨海默病。我做的不少工作都与阿尔茨海默病有关。 我关于生存分析所做的很多研究都是由阿尔茨海默病的研究推动的。所以，我的各个研究方向就这样联系在一起。我参与过的不同（的合作）项目集中于生存分析研究和阿尔茨海默病研究。他们都是紧密结合的。我最近的一个项目是阿尔茨海默病研究中随机删失协变量的处理。这个研究课题源于我在阿尔茨海默病的合作项目中，发现由不完整观察数据所引起的回归分析中的一些复杂问题。

关于教学方面，我希望我能教刚才提及到的生存分析课（注：此课程已被审批通过，已于2019年春季开设）。我们刚给学生们开设了一个文献精读俱乐部，同学们对此都很感兴趣。因此，我们正致力于让学生参与各种学习形式，无论是有学分的还是无学分的。我们致力于学习这个统计领域，进行阅读并展开谈论。当然还有专题研讨课。

接下来我来说一下我做的另一件事是，我是神经病学一个重要临床期刊， Annals of Neurology，的统计编辑。这项工作很吸引我，它让我再次意识到统计的复杂性，对于做研究的统计学家来说，这是非常有趣的。而且，我一直非常，非常积极地致力于与来自多样，小众背景学生接触。这是我以及我们系的一个非常重要的事情。好的，那就是我的自我介绍，我们以后可以再详谈。

- Hi, I'm Melody Goodman. I'm an associate professor. I did my undergraduate work at Stony Brook University majoring in applied mathematics and statistics in economics. And then I did my graduate work, getting my masters and PhD at Harvard. I wrote--

- Switch? Can you see okay?

- Yeah, I can see.

- Okay.

- I don't have my glasses on, that's the issue. I run the Measurement, Learning and Evaluation Lab, so I'm really interested in conducting mixed-message research, really thinking about how we measure things and creating new measures for things that don't exist. All of my work really focuses on health disparities, primarily racial and ethnic disparities in health across a broad range of disease outcomes, and I do a fair bit of survey research. And, I think I'm really sort of unique in this way as a statistician, but I'm really interested in engaging other stakeholders in research, 'cause I think it's important that we involve those who are gonna be impacted and try to develop solutions for the problems that we're trying to address. So, overall, I think my research really seeks to understand the social risk factors that contribute to health disparities, particularly in urban areas, and really trying to develop solutions. Working collaboratively with others to address those issues that we identify. I have several funded projects. So, I have a project from the Patient-Centered Outcomes Research Institute. And it's to develop a survey tool that asses the level of stakeholder engagement in research studies from the stakeholder perspective. And we're really trying to validate this measure, look at how it performs with other measures, and really look at how to implement it, implementations sort of things. And then I'm a bio-statistician on three new grants that are starting, and you can see they're sort of across a broad range of diseases and outcomes, including informed consent in older populations. So, part of when we do research is we consent people to participate in research. But, then how do you consent people that may have cognitive disabilities and all those types of things? And then, the last one is actually one of the Cancer Moonshot Grants. So, I don't know how much you've read about that or heard about it on the news. But we're gonna really be using electronic medical record data to identify patients who should be targeted for genetic testing for cancers and then figure out how to really implement those in real sort of care settings. So, I used to teach Biostatistics for Public Health. But starting this spring, I'll be teaching two new courses, Introduction to Data Management and Statistical Computing and then Survey Design, Analysis and Reporting. And I wrote a book.

- And I wrote a book!

嗨，我是Melody Goodman，我是一位副教授。本科阶段，我在石溪大学主修数学和经济统计。我在哈佛大学获得了硕士和博士学位。

我主持了测量学习与测评实验室. 我真的很感兴趣的是进行混合方法研究，仔细思考我们如何进行测量，在无现成测量工具时如何创造新的测量标准。我的所有工作都非常关注健康差距，主要集中在各种疾病结果中的存在的种族和民族差异。我做了一些问卷研究。并且，我认为作为一名统计学家，我在这方面确实很独特。我在尽力让其他利益相关者参与研究，因为我认为让那些将会受到影响并试图解决问题的人员参与其中是很重要的。因此，总体而言，我认为我的研究真正旨在了解导致健康差距的社会风险因素，特别是在城市地区，并且试图制定解决方案。我与他人合作，并有好几项课题得到资助。 我有一个来自Patient-Centered Outcomes Research Institute（注：一所非营利科研机构）的项目。此课题意在并且要开发一种问卷调查工具，从利益相关者的角度评估利益相关者参与研究研究的程度。我们真的试图验证这个测评，看看它如何与其他测评一起使用，如何操作等等。然后，我作为生物统计学家涉及了三项新的被资助的课题，您可以看到这些课题涉及广泛的疾病和结果，包括老年人群的知情同意书。比如，当我们进行研究时，我们需要取得被试的参与研究的知情同意书。但是，那么你如何从可能患有认知障碍的被试那里获得同意书呢？我们同时研究与之相关的问题。最后我来谈谈一个课题实际上是关于癌症的Moonshot Grants基金之一。我不知道你是否从新闻里听到或看到过。 但我们将会使用电子病历数据来识别那些应该针对癌症进行基因检测的患者，然后研究如何在真正的护理环境中真正实施。我曾经教过《公共卫生生物统计学》。但是从今年春天开始，我将教授两门新课程：《数据管理和统计计算入门》，然后是《问卷调查设计，分析和报告》。我还写了《公共卫生生物统计学》的教科书。

- Great, thanks. Thanks Melody. Oh there's Hi everyone, my name is Rumi Chunara. So, I'm an assistant professor, here, at NYU. And now that I'm starting my fifth year, so I've been here for a little while. And a little bit uniquely, I'm here in the college, in biostats, and I'm also a professor at the Tandon School of Engineering and Computer Science. So, you'll kinda see from the kind of work I do, that we are bringing together methods, both in computer science and statistics for some different type of data sources and problems. So, kinda my background is a little bit different, also studied engineering and computer science. And then, for a while, I was working in computational epidemiology, so, sort of looking at some of these new data sources, and got more closely related to this line of work. So, to summarize, so some of the things, so I'm broadly interested in these person-generated data sources. So you can think about, nowadays, there's a lot of ways that we can get data beyond surveys and government reports when thinking about health behaviors and risks and outcomes. So, we mine social media, we look at other internet and mobile-connected tools. So, whether it's people reporting symptoms on an app, or the wearables, so tracking people. There's a lot of new statistical problems that arise out of the way that people use these data sources and the human patterns of generating data. And, so, we also address those problems statistically, but then combine them in spacial and temporal models to understand how that contributes to disease in populations. And another interest of mine is from a few of my projects also, a lot of these data, they provide opportunity for looking at things in different places and different contexts. Maybe because there aren't any infrastructure or just, there's other problems. And, so, I have a lot of collaborations in different parts of the world, and we learn from all of those to develop these methods. So, I won't read it You can kinda see, it just brings out what I was telling you about. So, just some example projects. I've been doing a lot of work in, think about things like influenza, which, we all get affected by it every year, but it's something that we don't necessarily go to the doctor for, right? So, it happens at home, and that's why these kind of data sources are useful. Because it's gonna help us capture things that we just don't get data on otherwise. That's really exciting. So, that's something we've been working on, crowdsourcing data, and bringing data on that, comparing it to what happens clinically. Other types of just more statistical issues like, because these data often provide really granular information, like really over time, what happens on a daily or sub-daily basis. Or in space, like block to block. So, kind of like high granularity models. Some of that work's been on, of course, infectious diseases like influenza, but also thinking about our health behaviors on a more granular basis also. So, how people use substances, like alcohol and tobacco. And also, just some of the, again, issues with some new data sources and tracking people's health behaviors in terms of physical activity and their outcomes. So, I've been teach biostats for the undergrads. As well, I teach a course on data science at the School of Engineering. So this is kind of exactly what I was talking about, when we talked about some of these spacial statistics, we also do some introduction to supervised machine learning. And, so, hopefully we're gonna try and build that together with the public health students. So we can not only have students take it, but, actually, it'll be great for all computer science students to meet people who have more of a domain background and expertise, on the health side. Thanks.

哦，大家好，我的名字是Rumi Chunara。我是纽约大学的助理教授, 今年是我在纽大工作的第五年。

我的情况有点独特。我就职于我们环球公共卫生学院生物统计系，同时我也是Tandon工程和计算机科学学院的教授。由此，您可以从我所做的工作中看到，我们将计算机科学和统计数据中的方法集中在一起，用于处理某些不同类型的数据源和问题。

所以，我的背景有点不同，我学习了工程和计算机学。我的专业是计算流行病学，目前的研究着重于新的数据来源，比如，个人数据。除了传统的通过问卷调查和政府报告的数据来源， 我们可以通过很多其他方式获取数据。因此，我们挖掘社交媒体，查看其他互联网和移动连接工具。我们可以使用手机应用程序或可穿戴设备得到数据，追踪被试。由于人们使用这些数据源以及生成数据的人类模式，出现了许多新的统计学问题。

我们还在统计上解决这些问题，然后将它们结合在空间和时间模型中，以了解它如何导致人群中的疾病。我的另一个兴趣来自我的一些项目，这些数据很多提供了研究对象的地点和情境的信息。

也许是因为没有任何基础设施或只是，还有其他问题。

我在世界各地都有很多合作，我们从所有这些方面学习，研发新方法。

在幻灯片中， 你可以看到我的一部分课题项目。我做了大量的工作，比如流感，我们每年都受到影响，但是我们不一定要去找医生，对吧？所以，它发生在家里，这就是为什么这些个人数据源是有用的。因为它会帮助我们捕获那些我们无法从其他途径得到数据。这真的令人兴奋。因此，这是我们一直在努力将个人数据与临床数据进行比较。其他类型的更多统计问题，例如，因为这些数据通常提供非常精细的信息，例如持续一段时间，每天一次或每日发生几次。或者在空间中，就像一个个区间一样。所以，有点像高粒度模型。当然，其中一些工作是关于流感等传染病，但也在更细微的基础上思考我们的健康行为。再比如，人们如何使用酒精和烟草等物质。这些课题都涉及新的数据来源，追踪被试的行为和结果。

我一直在给本科生上生物统计课。同时，我在工程学院教数据科学课程。我们课上涉及空间统计，介绍机器学习里的有监督学习。我希望环球公共卫生学院的学生也有机会可以选这门课。事实上，计算机科学专业的学生和公共卫生专业的同学一同学习，交流，（合作）是非常好的。谢谢。

- Hi, my name is Stephanie. I'm very different from everybody you just heard from. I actually went to Michigan and my undergraduate degree is in women's studies and psychology. And then, when I got to Columbia, I said I really like statistics. So, my MPH is is sociomedical sciences with a focus on quantitative methods in statistics, and so is my DrPH. So, in terms of statistics, I'm really interested in intensive longitudinal designs and analyses. So, these are designs that collect repeated measures on a semi-frequent basis over the course of a week or even a day or a month. And, so, you can really look at temporal ordering and understand the temporal associations between different behaviors, for example. I'm also really interested in, I do a lot of data analysis for longitudinal cole-ber studies with count outcomes. As it stands right now in the field, we're not great at providing really accurate estimates for count outcomes in longitudinal data right now. So, I apply and also do examine some measurement in statistics along this line, to look at these, puss-on. I know, I'm trying to say not puss-on. But these longitudinal models to better specify them. I also run the Attachment and Health Disparities Lab. This is my social science background. So, I'm really interested in how features of close relationships are associated with changes in adaptation in health behaviors among vulnerable populations. So, racial and ethnic minorities, and/or sexual minorities. And lastly, my new work, as a couple of years ago, I'm really interested in biological markers of minority stress. So, now, in my studies, looking at these intensive longitudinal designs. I incorporate biological markers of stress in the form of cortisol or other markers of inflammation. And then, I also do a few cross-sectional studies as well, looking at other biological markers of stress. In terms of my grants right now, I have a lot going on in the lab. Even though I do a lot of secondary data analysis, these are all primary data analysis studies. And, I basically run these studies looking at stress and discrimination and stigma, and how they affect behaviors on a daily basis. And then I treat the questions in different ways to look at resilient mechanisms, for example, and these other features of close relationships, like social support from romantic relationship ideation, things like that. Yes, so that's me in a nutshell. In terms of teaching, so I'm doing appointment in social behavioral sciences and biostatistics. So, these are my biostatistics scores. Between myself and Violet, who you'll hear from in a moment, we teach the Regression one, the Regression two, and the longitudinal class. And I currently teach Intro to Biostats. But, maybe not anymore because we're teaching all these other courses. And then, in social behavioral sciences, I teach an intro course.

嗨，我叫斯蒂芬妮。我和你刚听到的每个人都很不一样。我在密歇根大学获得本科学位，专业方向是女性研究和心理学。然后，我去了哥伦比亚大学，真正喜欢上了统计。所以，我的公共科学硕士和博士学位的专业方向是是社会医学科学，专注于统计学中的定量方法。就统计而言，我对密集纵向设计和分析非常感兴趣。这些是在一天甚至一周或一个月内高频率地反复收集数据的设计。因此，您可以真正查看时间顺序，并了解不同行为之间的时间关联。我也对纵向数据里的计数结果数据很感兴趣。就学科现状来说，我们现在并不擅长为纵向数据中的计数结果提供真正准确的估计。所以，我在使用并检测一些测量（计数结果的）工具以及统计方法。

我指导依恋和健康差距实验室。这是我的社会科学背景。我真正感兴趣的是，在弱势群体中，种族和少数民族中，及/或性少数群体中，亲密关系的特征与健康行为的关联。最后，近年来，我对压力的生物学标记感兴趣。目前在我的研究中，采用了密集纵向设计，并用皮质醇或其他炎症作为压力的生物标志物。我的研究收到不少科研经费的支持，实验室里有很多事情要做。尽管我利用已有数据做了大量的数据分析，我也自己收集科研数据做分析。而且，我基本上都在研究压力，歧视和污名，以及它们如何影响日常行为。我研究心理韧性机制，例如，亲密关系的特征，如恋爱关系里的社会支持等。在教学方面，我就职于社会行为科学系和生物统计系。我和Violet 一起教三门课，回归一，回归二和纵向数据分析。这学期我正在教《生物统计学入门》。然后，在社会行为科学系，我教一门初级课程。

- I'm Shu Xu, you can call me Violet. My background is in psychology. I got my bachelor's degree from China and then after I came to the states, I really missed the best part of my training because I love modeling, I love data, I love computation. So, I went to say-day-vist to our Ph degree in quantitative psychology, so that, as you can see, laid out on my research, and my applied work really reflects the combination of those. My research interests. So, I should start talking about this from psychology study individual difference. Or so, we are so interested in growth, how people change over time. That made me concentrate my work on latent variable methods. So, this concerns some phenomenons that we cannot directly measure with some existing tool. Instead, we collect, usually by using surveys we can collect a lot of items, and then we extract the information based on those items. That's how we say we study our measured variable or our latent variables. And specifically, in terms of statistical modeling, I'm interested in developing and evaluating computational methods like longitudinal data analysis, strategic patient modeling, missing data, because yeah, there are so cannot be observed directly in a psycho metrics. So, in addition to using existing measurement tools, I help to develop and evaluate measures. Another research field I'm concentrated on is applied statistics. So, I like to apply those methods to solve their questions from real study. So, I collaborated and also I'm still collaborating with researchers from different areas, like preventative science, developmental psychology, clinical psychology and also public health. And recently, I have been focusing my applied work in tobacco research, because it'll reflect, I got very interesting things in data, like highly skewed data or no data. And also, I could develop me method in study transition, which is also a very interesting topic in longitudinal study. So, specifically, I'm evaluating patterns of conventional and emerging tobacco product use. For example, like e-Cigarette. It has been dramatically increased in use and it made a lot of parents worried about their kids. So, I think we can use statistical tool to solve this problem in real life and help to improve people's health. Oh teaching. So, yeah, I teach a lot of classes. I'm teaching Biostats for Public Health of undergraduate students and also graduate students. And next semester I will teach Regression and Multivariate Modeling. And this class, actually, has been split into two new courses. So, one is linear regression and another one is categorical regression modeling. And also, I will teach longitudinal data analysis class.

大家好， 我是Violet (徐恕)。我的背景是心理学。我从中国获得了学士学位，然后在我来到美国后，由于我喜欢建模，数据，和计算，所以，我取得了计量心理学博士学位。 一会儿你可以看到，我的研究表明，我的应用工作确实反映了这些元素的组合。我的研究兴趣 要从研究个体差异的心理学开始说起。我们通常对成长，人们如何随时间变化而变化感兴趣。我的工作集中在潜变量的研究方法上。这涉及一些现象，我们无法使用某些现有工具直接测量。我们只好使用量表通过 观测变量提取信息，这就是我们所说的潜变量。就统计模型来说，我有兴趣开发和评估方法，如纵向数据分析，结构线性建模，缺失数据方法等，这些统计学方法都涉及到潜变量。因此，除了使用现有的方法，我还致力于开发和评估统计学方法。我专注于的另一个研究领域是应用统计学。在统计学应用方面，我喜欢运用这些方法来解决实际问题。所以，我与不同领域的研究人员有合作，如预防科学，发展心理学，临床心理学和公共卫生。最近，我把统计学应用的重点放在烟草使用研究上。烟草使用的数据中有非常有趣的问题，比如高度偏态的数据分布，以及缺失数据。烟草使用类别的转变，这也是纵向研究中一个非常有趣的课题。我正在评估传统和新兴烟草产品使用的模式。例如，电子烟。它的使用量急剧增加，这让很多父母为他们的孩子担心。所以，我认为我们可以使用统计工具来解决现实生活中的这个问题，并有改善人们的健康状况。说说教学。我教了很多课。我正在为本科生和研究生教授《生物统计学入门》。下学期我将教一门课，回归和多变量建模。实际上，这门课程已分为两门新课程。因此，一门是线性回归模型（回归I），另一门是广义线性回归建模（回归II）。而且，我将教纵向数据分析课程。

- Thank you. Okay, so now, I'll turn back to Melody who will talk about our programs.

- So, if you study biostatistics, we hope that you will gain a set of skills, we call those competencies, by the time that you complete the program. We have seven major competencies that we hope all of our students obtain. So, I'm not gonna read them to you, but they're listed there, and a lot of them are about being able to work with data, being able to use data in public health, being able to work with software tools and all the kind of things that you would need for a real job. So, our competencies are really based on the things that we would want if we were hiring someone to work for us. So, right now, we have the MPH program with the biostatistics concentration. It's a two-year program. It has a total of 47 credits. You take 18 credits of general public health courses, so you take all of the five core public health discipline courses. You also have additional readings courses, including an informatics course which really tells you how to use digital library services and other information. And then you have 15 credits of biostatistics concentration courses. And these are the courses where we hope you get those seven competencies that I talked about before. And then you get 10 credits of electives, so that's the fun part. Those are the classes that you wanna take, not the classes you have to take. And our electives fall into two groups, Group A is methods, you get three credits of methods and then group B is statistics, and you get seven credits of statistics. And our students take courses all over the university, Steinhardt, Stern, the medical school, so it's really a chance to take classes that you wanna take, both in the College of Public Health, but also across the university. And then, we think of the MPH really as a practice degree, so as part of that, you have to do a practice experience. I think it's really good for students who are coming straight from undergrad, who have never worked before, 'cause it's your chance to get some real work experience. All of our students have to touch data as part of their practice experience, so you will have some real experience by the end of work and I'll talk about that more. Then, the culminating experience is a master's thesis, which is a year-long course which culminates in a thesis report. So, like I mentioned, you take a set of core public health courses, Biostatistics, Epidemiology, Health Care Policy, Public Health Management and Leadership, Global Issues in Social and Behavioral Health, Global Environmental Health, and the Essentials of Public Health Biology. That's really to give you a really good sense of a broad scope of what public health is, which is a really, really broad discipline that covers a lot of sub-fields. And then you take your biostatistics concentration courses. So, you'll take Data Management and Statistical Computing, which will basically give you a survey of a bunch of different software programs that different statisticians use to solve different problems. And then you'll take the regression sequence or Regression one and two. You'll take a class on survey data, just because we encounter a lot of survey data in public health, particularly from the health department, but also from our government and other secondary sources. And then lots of us collect survey data. And then you round it off with longitudinal analysis. So, the applied practice experience is usually completed in the summer between your first and second years. It's 180 hours of work. You get to pick the practice environment of your choosing, so I always tell students, go pick a practicum in the place where you think you want to work when you graduate, 'cause it gives you a good sense of if that's really what you want to do when you graduate. But, students have gone all over. They've gone to the health department, they've gone to pharmaceutical companies, they've stayed in academic research centers. But there's a lot of new opportunities, I think. For you guys, there's a lot of health data startup companies, particularly in New York, which gives you guys a good chance. And then, there's other companies here like Google and Amazon, which will be coming to New York City soon. And they also have health arms. So, I think there's a lot of chances for biostats students to maybe not do things that have traditionally been done in biostat, but also take on some new projects. You must actively work with data. Touch data and do things with data. Massage it and love it and learn it. And gain skills related to those seven competencies that we want you to get. You must be supervised by someone else who also knows how to work with data, so you can't just go work for someone who needs a statistician. You have to go work under supervision of someone who really understands data. And then you submit two products of that, that we evaluate to make sure that you're getting those competencies. So, one of the competencies says use statistical software. So, one of the things a lot of students submit is software code, showing that they know how to write software programs that do specific things. And then you take your masters thesis in your second year, it's a two-semester course. Thesis one and Thesis two, each is two credits for a total of four credits. And really, I think a thesis is like learning how to write a manuscript, right? So you conduct a literature review, you do data analysis, you develop tables, figures, and images and you basically write a report of the results. The nice thing about thesis is you get to select the topic that you get to work on. So, it's your area of interest. And then, you're also paired with a faculty member who will mentor you through that process, and I think a lot of our students have tried and gotten publications out of their thesis 'cause they've done some original work. So, our pending program, which we hope will be approved soon, is the new Masters of Science degree. It's more of a research-focused degree. So I said the MPH is more practice-focused, so we're thinking students who want to go on further for doctoral training or people who really wanna work in a research-intensive environment. You take fewer public health courses and you really focus more on biostatistics and epidemiology courses. So, this is for people who really wanna understand study design and data collection and lots of ways to analyze data. There's no practice experience here, but these students will also have to do a thesis. And then, we're asking that when they take their electives, they pick a thematic area, so that when you think about your electives, you're sort of getting a specialization in something, so that when you go on the market you can say I have an MS in biostatistics, but I'm really good at data visualization. Or GIS or whatever it is that you decide to specialize in. And, I also think this gives you a chance to maybe pick some things different, some of us, we talked about all the different paths that we took to get here and all of our different interests, so some of you may wanna do social justice or some other things that may not be traditionally biostat-focused. So, you take the two core courses, which are just biostat and epi, and then you have lots of concentration courses, which are really a series of epidemiology courses and a slew of biostatistics courses. So you're really well-rounded in terms of both epidemiology and biostat when you're completing this concentration. Those are our two programs. Any questions about that before I turn it over to the admissions folks?

如果您学习生物统计学，我们希望您在完成课程时获得一系列技能，我们称之为能力（competencies）。我们希望所有学生都能获得七项主要能力。（在幻灯片里可以看到），其中很多是关于公共卫生领域中使用数据，能够使用软件工具，以及其他和你真正的工作时所需要的所有能力。这些能力是基于未来工作需要而设计的。目前我们系开设了为期两年的公共卫生硕士生物统计学方向，共计47个学分。您需要完成18个学分的基础公共卫生课程，因此您将参加所有五个核心公共卫生学科课程。您还有其他阅读课程，包括信息学课程，它真正告诉您如何使用数字图书馆服务和其他信息。然后你有15学分的生物统计学专业课程。这些课程我们希望您能够获得我之前谈过的七项能力。然后你还要修10学分的选修课，这些课程会很有趣。因为这些是您想要修的课程，而不是您必须参加的课程。我们的选修课分为两组，A组是方法，3个学分，然后B组是统计学，7个学分。这些课可以从学校的其他学院修得，比如Steinhardt 学院，Stern商学院，或是医学院，所以这是一个真正自己选课的好机会，不仅可以从在全球公共卫生学院内选，还可以在整个大学的其他院系选。然后，我们认为公共卫生硕士学位真的是一个实践型的硕士学位，所以作为培训的一部分，你必须有一个实习经验。我认为对于直接从大学毕业的学生来说，尤其以前从未工作过的学生来说真的很好，因为这是你获得一些真实工作经验的机会。我们所有的学生都必须将使用数据作为他们实习经验的一部分，这样，在工作结束时你会有一些真正的数据工作经验，我会更多地谈论这些。最后是硕士论文，这是一个为期一年的课程，最终完成论文报告。因此，正如我所提到的，您将参加一系列核心公共卫生课程，生物统计学，流行病学，卫生保健政策，公共卫生管理和领导力，社会和行为健康的全球问题，全球环境卫生以及公共卫生生物学导论。这真的是为了让你对公共健康领域有一个很好的广泛认识，这是一个真正的，非常广泛的学科，涵盖了很多子领域。然后你参加生物统计学专业课程。因此，您将学习数据管理和统计计算，您将使用一系列不同软件解决不同的问题。然后你将上的课包括回归系列课程或回归I和II，一门问卷调查数据课程，我们在公共卫生领域使用了大量调查数据，特别是卫生部门，政府部门和其他现有的数据。然后我们很多人收集问卷数据。你也可以选纵向数据分析。因此，应实习通常在您的第一年结束后的夏季完成。 实习的时间是180小时。你可以自选实习环境，所以我总是告诉学生，选择一个你认为毕业后想要工作的地方的实习，因为它能让你很好地体会你毕业后是否真的想做这样的工作。以往的学生有的去了卫生部门，有的去了制药公司，还有的留在了学术研究中心。但我认为，还有很多新的机会。对于你们来说，有很多健康数据创业公司，特别是在纽约，这给你们很好的机会。然后，谷歌和亚马逊等其他公司将很快进入纽约市。而且他们也有健康部门。因此，我认为生物统计的学生有很多机会去做非生物统计传统的工作，也会接受一些新的项目。（实习过程中），您必须积极处理数据。接触数据并处理数据。细细揣摩数据热爱数据，并学习数据。并获得如我们期望的的七项能力相关的技能。您必须在懂得处理数据的人员指导下工作，您不能只为需要统计学家的人工作。（实习结束后），你提交两项成果供我们评估，以确定你获得了这些能力。（比如，）其中一项能力是使用统计软件。因此，许多学生提交软件代码，表明他们知道如何编写执行特定事务的软件程序。

你在第二年开始准备硕士论文，这是一个两学期的课程。论文一和论文二，每个是两个学分，总共四个学分。 实际上，我认为论文就像学习如何撰写稿件一样，对吧？您进行文献综述，进行数据分析，制作表格，做图，并结果报告。关于论文的好处是你可以选择你要学习的主题，这是你感兴趣的领域。你还会有一位指导老师指导你完成论文。 我们的很多学生发表了论文，因为他们已经完成了原创课题。

我们将批准的是新的理学硕士学位方向（注：此方向已被通过）。这是一个研究型的学位。所以我说公共卫生硕士是实践型的。（这个新的理学硕士学位项目的）学生是那些想要继续进行博士学位深造的学生，或者那些真正想在研究密集型环境中工作的人。这个方向涵盖的公共卫生课程较少，更注重生物统计学和流行病学课程。这适用于那些真正想要了解研究设计，数据收集以及分析数据的人。这个方向没有实习要求，但学生也必须完成一篇论文。然后，在选修课程时，学生选择一个专题领域，并发展成自己的专长。这样当你找工作时你就可以说我有生物统计学的理科硕士学位，我擅长数据可视化。或者你可以选择地理信息系统或者其他任何你决定专注的领域。我认为这给你一个机会可能选择一些与他人不同研究领域，有不同的发展方向，发展不同的研究兴趣。所以你们中的一些人可能想做社会保障或其他一些非生物统计传统的事情。 所以，你选择了两个核心课程，它们只是生物统计学和流行病学，然后你有很多选修课程，这些课程是一系列流行病学课程和一系列生物统计学课程。因此，当你结束课程时，你在流行病学和生物统计学方面的训练都非常全面。这是我们的两个方向。