Research Services offers tutorials and workshops on a variety of topics. Each semester, we present a series of tutorials. If you have suggestions, please contact us (researchservices@bc.edu). We are also available for consulting.
Tutorial Descriptions - Fall 2023
Introduction to Machine Learning
Machine learning is a data analysis method of getting computers to act without being explicitly programmed. It is based on the algorithms that use statistics to build models and find patterns in massive amounts of data. Machine Learning is extensively used in a wide variety of applications and changing our day-to-day life.
This tutorial is for beginners to learn and will cover:
- Introduction/Definition
- Where and Why Machine Learning is used
- Types of Learning
- Supervised Learning
- Unsupervised Learning
Presented by Yixin Pan.
Friday, September 22, 2023 from 11am–12pm (Zoom)
Introduction to BC’s Linux Cluster
This tutorial is intended to be an introduction to the Linux cluster at Boston College. Currently the user can access the Andromeda cluster. An overview, the primary components, and examples of how to use BC’s Linux cluster.
This hands-on tutorial will cover:
- Overview of the Andromeda Linux cluster system at Boston College
- The hardware architecture
- Management of Linux Cluster
- How to remote access the cluster
- Common Unix/Linux commands
- How to use software modules and SLURM queuing system
- How to submit jobs to Andromeda
Presented by Wei Qiu.
Monday, September 18, 2023 from 12-1:30pm (Zoom)
Intro to Stata 1: Getting Started, Descriptive Stats & Do Files
Stata is a powerful, yet easy to use statistical package. This hands-on tutorial is designed as an introduction for beginning users who are just getting started using Stata. The emphasis in this tutorial is on exploring the data, cleaning the data for research purposes, and generating descriptive Statistics.
Accessing Stata on BC Apps: You will receive instructions on how to do this prior to the tutorial. How to access apps on BC Apps.
- Loading data
- Data manipulation
- Descriptive statistics
- Do-files and log files
Presented by Deyin Jia.
Monday, September 18, 2023 from 3-4:30pm (Zoom)
Introduction to Regression
As the most common methodology in statistical analysis regression is an important tool for any modern researcher. This course is intended as an introduction to standard or linear regression. We will focus on estimation methods, identifying and validating model assumptions. We will also focus on hypothesis testing for regression estimates and statistical model building. We will use R software but the goal of the course is to learn concepts and is not intended as a tutorial for any specific software.
Note: The mixed modeling course is a natural sequel to Introduction to Regression.
Presented by Matt Gregas.
Thursday, September 21, 2023 from 10–11:30am (Zoom)
Linear Mixed Effects Modeling
This tutorial is a brief introduction to linear mixed effects (LME) modeling, also known as multilevel modeling or hierarchical linear modeling. LME models are essential for researchers handling either longitudinal (repeated measures) data or data that is 'clustered' (e.g. students nested within classrooms, and classrooms nested within schools). Many familiar methods such as ANOVA or regression assume that all observations are recorded independently; Clustered data and data with repeated measures violate this assumption. LME modeling is an extension of regression that accounts for the correlated data structure inherent in repeated-measures and clustered designs. In this tutorial, we introduce the model, discuss when and why this method should be used, and how to interpret results in common statistical programs. This tutorial is appropriate for anyone with a background in linear regression. Those wanting a refresher may consider attending the Research Services tutorial on regression.
Presented by Melissa McTernan.
Thursday, September 21, 2023 from 12-1:30pm (Zoom)
An Introduction to Latent Growth Curve Modeling
This tutorial is a brief introduction to latent growth curve modeling (LGCM). LGCM is a flexible approach for modeling longitudinal or repeated measures data. These models are fit within a structural equation modeling (SEM) framework where latent variables are used to capture linear or nonlinear "growth trajectories", or change across time. A researcher may be interested in modeling within-person growth patterns (e.g. how are student math scores changing across time?) or interested in explaining between-person differences in within-person growth patterns (e.g. what factors explain the differences between the students who are improving across time and those who are not?). LGCM will allow a researcher to study these kinds of questions- and many more- within a single model. LGCM can be implemented in R, Mplus, Stata, AMOS, SAS, or JMP. In this tutorial, we introduce the model, discuss when and why this method should be used, and briefly demonstrate the approach and how to interpret the estimated parameters. This tutorial is appropriate for anyone with a background in linear regression and some familiarity with SEM.
Presented by Melissa McTernan.
Tuesday, September 26, 2023 from 12-1:30pm (Zoom)
Introduction to REDCap (Research Electronic Data Capture)
This tutorial is geared towards Boston College Principal Investigators, researchers and research project team managers. REDCap stands for Research Electronic Data Capture. REDCap is a web-based, data collection, database management system that was originally developed at Vanderbilt University, initially for medical research. REDCap is now overseen by a consortium of academic research partners in the United States and throughout the world. Boston College is part of the REDCap consortium.
In this introduction to REDCap we will discuss:
- How to request a REDCap project at Boston College
- How to make sure that your REDCap project complies with the mandates of your project's IRB approval
- How to create basic data collection forms
- An introduction to best practices for setting up your REDCap project
- We will discuss additional REDCap functionality including offline survey capabilities, text to voice capability, potential for using twilio.com SMS services (for an additional fee), improved field calculations, repeating forms and more
- How to enter data into REDCap
- How to control REDCap user access rights
- How to export your data
Research Services staff are available to meet with members of the Boston College community to discuss individual REDCap projects. Individual consultations or customized class consultations are available by emailing researchservices@bc.edu or dalgin@bc.edu.
If possible, prior to the tutorial, please fill out the BC REDCap Terms of Use survey described on this google doc and indicate that you will be attending the REDCap Tutorial.
Presented by Rani Dalgin.
Thursday, September 28, 2023 from 3-4:30pm (Zoom)
Creating Web-Based Surveys with Qualtrics
Qualtrics offers a fairly intuitive graphical user interface to create complex surveys without complicated programming or coding. Qualtrics offers extensive documentation, free online tutorials, an extensive library of surveys and options for encryption and anonymity, and excellent customer support. Qualtrics also offers built in social media sharing functions and an accessibility checker. Working within pre-defined templates, you can use many different types of questions, including text, multiple checkboxes, sliders, single-answer radio buttons, and Likert scales. Qualtrics offers extensive branching functionality.
Once the survey is completed, data can be downloaded into a format that can be used with a variety of quantitative and qualitative analysis programs. Qualtrics also offers foreign language functionality.
This tutorial will demonstrate how to create a survey in Qualtrics and also include a section on research protections and informed consent with respect to online survey development, distribution, and analysis. Boston College faculty, students, researchers, and administrative staff may create their own Qualtrics accounts in advance of the tutorial by logging on at bostoncollege.qualtrics.com (with your BC credentials).
If possible, please complete this short BC Qualtrics Terms of Use Survey before attending the tutorial.
Research Services staff are available to meet with members of the Boston College community to discuss individual Qualtrics projects.
Individual consultations or customized class consultations are available by emailing researchservices@bc.edu or dalgin@bc.edu.
Presented by Rani Dalgin.
Tuesday, October 2, 2023 from 3-4:30pm (Zoom)
Intro to Stata 2: Graphing, Dataset Combining, Linear Regression, Stat/Transfer
Stata is a powerful, yet easy to use statistical package. This hands-on tutorial is designed as an introduction for beginning users who are just getting started using Stata. The emphasis in this tutorial is on basic graphing, merging data, and linear regression.
- Basic graphing and graph editor
- Combining multiple datasets
- Linear Regression in Stata
Presented by Eric Williamson.
Tuesday, October 24, 2023 from 10:30am-12pm (Zoom)