math/statistics/econometrics (STEM) programing/systems (STEM) data management and visualization/analytics (social science)
people debashis, haydee, iming, jeff, dan, paul, sarah sunil adam
basics mean, median, mode, standard deviation, correlation, distributions, probability packages, interpreter, IDE, lists, data frame, assignment, etc data import/export, recoding, labelling; scatterplots, line plots, bar charts, histograms
intermediate OLS regression ? data merging; APIs; classification trees, k-nearest neighbors; 3D-graphs, fit plots, animated graphs, thematic/choloropleth maps
advanced logistic and multinomial regression, GLM, time series, panel data ? ?
  • social science (FASC)
  • public policy, administration, political science, and urban studies [faculty lead: Adam Okulicz-Kozaryn and Paul Jargowsky]
  • economics/econometrics [faculty lead: Iming Chiu]
  • psychology/psychometrics and childhood studies [faculty lead: Dan Hart and Sarah Allred]
  • sociology, anthropology and criminal justice [faculty lead: Michelle Meloy and Richard Stansfield]
  • STEM fields
  • computer science [faculty lead: Sunil Shende]
  • math and statistics [faculty lead: Debashis Kushary, Haydee Herrera, Will Y.K. Lee]


  • mission/values:
  • open source, reproducibility, collectivism, collaboration
  • automation, data pipeline building
  • readability, simplicity, communication
  • application, utility, usefulness
  • flexibility, freedom, individualism
  • creativity, innovation, independence, critical thinking
  • transparency and ethics
  • mastery of "data story-telling": 1) What data are telling, 2) what I want to say, 3) what audience needs to know


  • Welcome to 12 credit certificate / 30 credit MS in data science!

    30 hour MS curriculum contains 15 hours of core: Python scripting, basic statistics/math, data management, and visualization; 9 hours of electives (extending programming+statistics/math and/or fulfilling "domain" (substantive expertise) part of the Venn diagram), and 3 hours of internship and 3 hours of practicum/capstone/thesis.

    12 hour certificate requires at least 2 (we recommend all 3) core classes + 2 (or 1) electives.

    All core classes are taught in Python. Electives vary. RU-Camden has a dedicated data science lab.

    core classes (5classes*3credits=15credits)

  • (1) Introduction to analytics/data science [adam, sunil, debashis, iming, haydee] AND Statistical learning, reasoning/ critical thinking/problem solving [debashis, haydee]
  • (2) Data structures/storage and manipulation/preparation [adam, sunil]
  • (3) Analysis and data mining [debashis] AND Machine learning [iming]
  • (4) Visualization [adam]
  • (5) Application [co-taught to show various applications: adam, sunil, iming, debashis]


  • add here non-credit (or credit) short workshops/labs on sas, r, gis etc

    here the same electives organized by domain/specialization



    electives (3classes*3credits=9 credits): [depending on your track STEM v CSS pick from your discipline])

    highly recommended electives: closely related data science classes from across the campus

  • 50:220:122 Introduction to Data Science https://economics.camden.rutgers.edu/files/syllabus122_F2019_updated.pdf
  • 50:220:366 Applied Data Mining & Knowledge Discovery iming and jeff (Yuchung Wang) new class on mining, machine learning (maybe part of core OR workshop/lab) [R]
  • 56:198:562 Big Data Algorithms by Sunil [Py]
  • 56:198:501 Data Structures and Algorithmic Problem Solving in Python [Py]
  • (Big Data Algorithms).
    business school
  • 53:623:510: Managing projects and IT
  • 53:716:502 Business Analytics
  • 53:623:517 Data Management & Business Intelligence
  • 53:716:535 Big Data Analytics & Visualization
  • 53:716:540 Social Media Analytics & Sentiment Analysis
  • other classes that fulfill especially 'domain' knowledge (substantive expertise)

    social science
  • 56:163:661 Quantitative Methods
  • 56:202:600 Research Methods in Criminal Justice
  • 56:202:601 Data Analysis in Criminal Justice
  • 56:830:580 Research Methods
  • 56:830:650 Statistics & Research Design
  • 56:824:713 Research Design
  • 56:824:708 Categorical and Limited Dependent Variables
  • 56:824:709 Quantitative Methods II
  • 56:830: Program Evaluation
  • 56:830:638 Survey Research Methods
  • 56:824:718 Data Management
  • 56:163:615 Using Archival Data to Study Children
  • 56:834:608 GIS for the Public Sector
  • 56:645:567 Statistical Models
  • 56:645:565 Time Series & Forecasting
  • digital studies/humanities 56:209:670 Distant Reading with Text Analysis 56:209:530 Creative Coding
    math
  • 50:960:489 Statistical Models, Debashis Kushary [R]
  • 56:645:563 Statistical Reasoning, Debashis Kushary [R]
  • 50:960:491 Time Series and Forecasting, Debashis Kushary [R]
  • 50:640:331 Probability and Stochastic Processes by Siqi Fu [R][more theoretical undergrad]
  • 56:645:549-550 Linear Algebra and Applications
  • 56:645:571 Computational Mathematics I
  • 56:960:481/482 Mathematical statistics, Discrete Mathematics
  • and possibly more from https://math.camden.rutgers.edu/programs/graduate/industrialapplied-mathematics/ (descriptions at https://math.camden.rutgers.edu/programs/graduate/graduate-courses/)

    mandatory internship (3 credits)

    This is important part of curriculum--it builds link between curriculum and application: helps graduates to find a job and then possibly hire more graduates and also helps us to see what are the industry needs and trends.

    In special circumstances if internship is not possible this can be replaced with addition 3 hours of the following (practicum/real world project for practitioners OR capstone/thesis/independent study for researchers)

    also see https://libguides.rutgers.edu/datamanagement

    practicum/real world project for practitioners OR capstone/thesis/independent study for researchers (3 credits)





    what makes us different

    We are trully interdisciplinary and we will fit your profile. Program faculty come from the following fields: math/statistics, computer science, social science, and humanities.

    Our program is not just data science itself but we combine it with substantive fields and prepare people to work specifically in these fields so our distinguishing feature and comparative advantage is small core and plenty of tracks/specializations that involve classes form our dept.

    We focus on flexibility: everyone is at least a little different and we want to accommodate that--only five core classes and electives as opposed to other programs where majority is core. Also value hands on experience: internship/practicum. We focus on Python, and supplement it with R.

    For the sake of added flexibility, we supplement/complement the curriculum with workshops/labs.

    We also offer a 12 hour certificate.

    We focus on skills and application, we want you to be able to do things. We won’t waste time on unnecessary stuff. To that end our core classes are completely new and designed specifically for this degree (not redesigned existing classes or worse yet a collection of existing classes made for other purposes).

    There is an argument to be made that "data science" is mostly about cleaning, shaping data, and moving it from place to place. And we acknowledge it and provide a thorough and dedicated class to teach you the right skills.

    do i need math?

    Mathematics is the universal language of science. It is the foundation of many breakthroughs in science, engineering, and finance. Mathematics enables us to analyze and model physical and social phenomena quantitatively and logically and provides us the tools to solve them. Many jobs in today’s economy require analytic and quantitative skills. The math and statistics courses in this program equips the students with techniques and machineries to succeed in their professional careers.

    Hence, yes, you need some math, and ideally the more the better--math makes your transcript and CV look good--shows you can think logically with numbers--yet much math is not necessary, if you don't think better with math or have a "math block" you'll only have to take most basic math. Yet, also for many math does help and initial math block is often illusory--a person is afraid until she tries it and it works.

    To summarize, only basic math is required, but there's plenty of math offered as electives. We recommend taking as much math as possible/useful/practicable.

    actual program description ends here




    pondering ideas, thinking, logic

    General points/philosophy

    Philosophy is to have as much variety from across FASC in core as possible so it's truly interdisciplinary and students get different perspectives and also get to know different dept and better able to choose electives.
    Note that Python over R is a choice (just for core, not electives), but i have a ton of recent evidence that Python is better, just one here, scroll to table mid page https://mbs.rutgers.edu/program/analytics-discovery-informatics-data-sciences

    STEM v CSS

    However, looks like there are 2 camps: STEM v soc sci/humanities, and so maybe lets have 2 tracks. This is what i think Ray was also kind of suggesting. And this is what others do, eg NB has 3 tracks: https://mbs.rutgers.edu/program/analytics-discovery-informatics-data-sciences but we allow more flexibility with courses. This would allow flexibility and having everyone on board and this could actually be our comparative advantage over competition that we have truly interdisciplinary program. Also, it is easy: most classes already exist, just have to develop 3-4 new classes for the core: I have one almost already ready (data management/Pandas and visualization/Matplotlib) and other ones i can easily develop possibly ideally as co-taught classes, there was enthusiasm on co-teaching: Haydee, I-Ming).

    Initial thinking: everybody needs to take intro and data management/Pandas and visualization/Matplotlib; but the other 2 core classes: possibly 2 tracks: for STEM: CS/math classes that i am narrowing down (see below); and for soc sci/humanities: some lighter more applied equivalents: i am actually happy to develop these from scratch (maybe co-teach with dan, paul, i-ming, jim?)

    Initially was thinking about 2 tracks STEM and CSS for core, but after thinking and conversations (esp with Sunil) decided to go just with one core, and then people can satisfy STEM v CSS and domain area through electives

    Then increased number of core classes to 5 as was thinking about useful stuff, and then cut back to 3: this allows more flexibility and just most important stuff

    possible books to be incorporated in core

  • manual: The Data Science Design Manual Steven S. Skiena, a lot additional stuff at http://www.data-manual.com incl py code http://www.data-manual.com/data
  • thinkstats2: http://greenteapress.com/thinkstats2/html/index.html free, applied, has math, an py code