Senior Learning Analyst

Purpose

Conduct data-intensive analysis in support of an ambitious research agenda to advance learning analytics for course, instructor, and program improvement.

Essential Functions

The successful candidate will work as a member of the Research, Evaluation, and Learning Analytics (RELA) team in Learning Sciences at The University of Texas at Austin to quantify and evaluate the efficacy of a broad range of educational interventions and programs in an effort to improve student outcomes. Responsibilities fall into two categories: (1) experimental design & data analysis, statistics, including modeling, and (2) data visualization with the pre-requisite computational skills to support both of these activities. The experimental design and modeling involves designing methods by which investigating the variables that impact student learning can be meaningfully identified and represented in useful practical learning models. This includes defining and/or creating appropriate student outcome variables from these data sources at both the student and teacher level. Data analysis includes describing the data and constructing predictive models that use the acquired and extracted features to predict student outcomes. Based on the statistical models and analyses, the outcomes of this work needs to be clearly communicated in sophisticated visual representations along with model testing. Both of these activities require data manipulation skills that start with extracting and processing the data from stored repositories (RDBMS, NoSQL, or others), using data exchange formats such as JSON, XML, RDF, along with common flat log files from different sources including the university Canvas learning management system (LMS), or specific learning tools. The successful candidate will work closely with developers to define new variables to capture from online and in-person learning activities, and support senior Learning Sciences staff as needed in their evaluation of programs, assisting them in development of recommendations based on the emerging data. Finally, to support the data science and learning analytics community, developing and teaching hands-on seminars and workshops designed to help faculty use data to improve instruction are expected.

Required Qualifications:

• Masters degree or equivalent in statistics, engineering, computer science, applied mathematics, or other discipline with a computational and/or data science component and significant experience or equivalent training, education or experience
• Five or more years of experience writing code and/or conducting analyses in one more high-level programming/scripting languages (e.g., Python, Jupyter Project environment) and expertise with JavaScript
• Five or more years of experience working in R, Numpy/SciPy, Matlab, or other statistical and numerical computing environments
• Three or more years of experience applying regression, classification, and clustering methods to large datasets in a predictive modeling context
• Three or more years of experience working with relational/SQL databases, NoSQL databases
• Facility working with data exchange files formatted as RDF, JSON, & XML
• Comfort working with Github & similar web-based version control systems
• Familiarity with IMSGLOBAL interoperability standards (e.g., Caliper, LTI, QTI, etc.)
• Ability to work independently in an academic setting and adapt quickly to new challenges
• Experience analyzing and modeling data from randomized controlled experiments or A/B tests
• Excellent communication skills and ability to distill and present the results of complex analyses to highly interdisciplinary audiences

Preferred Qualifications:

• Ph.D. degree or equivalent in statistics, engineering, computer science, applied mathematics, or other discipline with a computational and/or data science component
• Experience contributing to open source software projects and expertise maintaining code in version control
• History of contributing to open source statistical analysis or machine learning packages
• Familiarity with the literature on predictive modeling and real-time applications of it
• Awareness of stream analytics techniques (e.g., using AWS Kinesis, Microsoft Azure, IBM Bluemix, etc.)
• Understanding of xAPI and contemporary LRS systems
• Ability to express data concepts graphically using common tools such as d3.js, Processing, Gephi and end user oriented packages such as AWS QuickSite, Google Charts, ManyEyes, CartoDB and Weka
• Experience using Big Data analysis platforms (Hadoop, Spark, etc.)
• Familiarity with object oriented languages such as C++, C, Java, Smalltalk (Squeak) Objective C, Swift, or C#, and languages with some object oriented capabilities such as PHP and Ruby.