JMP Statistical Discovery Academic Ambassador Ross Metusalem is coming to UTK to train students and faculty researchers how to perform statistical analyses in JMP Pro. This workshop will demonstrate how to perform regression and analysis of variance (ANOVA), which are two of the most widely applicable statistical techniques used by researchers. This session begins by presenting simple linear regression and one-way ANOVA before moving on to multiple regression and factorial ANOVA. Additional analyses presented will including how to perform post-hoc tests, residuals diagnostics, and model visualizations using JMP’s model profiling tools. This workshop will end with a quick review of more advanced linear modeling methods available in JMP Pro.

Register for either the face-to-face or online via Zoom session. Workshop occurs in Eastern Standard Time.

This is a researcher focused workshop.

JMP Statistical Discovery Academic Ambassador Ross Metusalem is coming to UTK to train students and faculty researchers how to perform supervised machine learning models in JMP Pro. This workshop will discuss how supervised machine learning (ML) uses training data to build models that can accurately predict outcomes or assign category labels. This session will demonstrate tools for building, validating, and deploying supervised ML models in JMP Pro. We will demonstrate how to fit several model types including neural networks and decision trees, perform automatic model cross-validation, and deploy machine learning models to generate predictions. JMP Pro’s Model Screening tool which simultaneously fits and cross-validates many types of models to determine which model type might perform best on your data will also be presented. Finally, we’ll briefly review the latest supervised ML advance in JMP Pro 18: Image classification models with the popular Torch deep learning package.

Register for either the face-to-face or online via Zoom session. Event occurs in Eastern Standard Time.

This is a researcher focused workshop.

Multilevel regression models are models in which the experimental units are nested within larger groupings such as students within classrooms or workers within factories. Both the individual experimental unit (student) and the higher-level grouping variable (classroom) can contribute unique variance to the variables in the regression and the relationship between them. The workshop will work through a simple two-level model using two popular statistical packages, the SAS 'Mixed' procedure and the Mplus 'Two level Random' analysis. This workshop takes approximately 1 hour and 30 minutes to complete.

This is a researcher focused workshop.

Basic statistical concepts are fundamental to understanding more in-depth analysis that are typically performed when conducting quantitative research. This workshop discusses the contribution of statistics to the scientific method and provides an overview of the statistical methods associated with basic data analysis. Topics include basic statistical concepts, developing a research hypothesis, the importance of statistical power, an explanation of basic statistical tests, and guidelines on how to present research results. An example is provided of a past research project in veterinary medicine. This workshop takes approximately 45 minutes to complete.

This is a researcher focused workshop.

The same data visualization could be good or bad depending on the point you’re trying to make. We will briefly discuss your options of using words and tables, then we will focus a variety of graphical displays, showing when each works best. Several different plot types are discussed including histograms, bar charts, dot plots, pie charts, line charts, box plots, and scatter plots with various types of fit lines. This presentation takes approximately 1 hour to complete. This workshop is optionally followed by R: Data Visualization which shows how to make the graphs discussed in this workshop.

This is a researcher focused workshop.

Students will learn the basics of using the JMP interface to accomplish the following tasks: Manage data effectively; Explore data by using the software's extensive graphical capabilities; Create and manage data summary reports in JMP; Conduct basic data analysis; Find help and learning resources. Hands on tutorials will be conducted throughout the session and prior preliminary knowledge of JMP is not required. This workshop takes approximately 1 hour to complete.

This is a researcher focused workshop.

Design of experiments (DOE) is the design of a process aimed to describe and explain sources of variability to improve product and process development. In simpler terms, experimentation is used to determine causation. This workshop begins with an overview of basic statistical concepts, the scientific method, and the statistical contributions to the scientific method. A simple example is used to walk through the scientific method while demonstrating statistical thinking, methodology and analysis. Basic principles of DOE are then covered leading into specific types of designs. Completely Randomized Designs (CRD) and Randomized Complete Block Designs (RCBD) are introduced. The workshop ends with the development of these designs built in JMP software. This workshop takes approximately 1 hour and 45 minutes to complete.

This is a researcher focused workshop.

A mediated regression is one in which the relationship between the independent and dependent variables is partially or totally due to the action of a third intervening variable. In this workshop we will extend the idea of mediated regression to multilevel models, models made up of variables from more than one level of aggregation, e.g. students within classrooms, etc. A working knowledge of basic mediated regression and multilevel modeling is desirable. This workshop takes approximately 1 hour and 30 minutes to complete.

This is a researcher focused workshop.

Mediator and moderator variables are terms in a regression model which intervene between the independent and dependent variables and alter the relationship between them. This workshop will explain the difference between mediation and moderation effects and show how to model and assess them. Examples will use both traditional statistical software and structural equation modeling software. This workshop takes approximately 1 hour and 30 minutes to complete.

This is a researcher focused workshop.

In a mediated regression model the relationship between the dependent and independent variables is at least partially the result of the intervention of a third variable, the mediator. In moderated mediation this mediation effect is in turn influenced by the value of yet another variable, the moderator. This workshop will cover how to run a moderated mediation model in several popular software packages and how to interpret the results. This will include a discussion of, and SAS and Mplus code for, the Johnson-Neyman technique for determining the range of moderator values for which the mediation is and is not statistically significant. A basic understanding of linear regression as well as the concepts of mediation and moderation will be assumed. This workshop takes approximately 1 hour and 15 minutes to complete.

This is a researcher focused workshop.

A moderated regression is one in which the relationship between a predictor and the dependent variable is altered (moderated) by the value of a third variable, called the moderator. In this 10-minute video we will discuss how to use the multi-group analysis feature of the Amos structural equation modeling software to assess the statistical significance of the moderation effect of a categorical moderator variable.

This is a researcher focused workshop.

NVivo is a qualitative data-analysis package used to organize and analyze resources used in qualitative research. NVivo allows researchers to analyze documents, electronic web surveys, audio and video files, images, database tables, web pages, and social media content in one platform. This workshop will provide an overview of the software. The interface and features are demonstrated, along with examples of research projects utilizing the software. Participants will import and organize data in NVivo, add research notes and annotations, and code qualitative data in a variety of ways. Participants will also understand the potential of NVivo as a tool for organizing, questioning, visualizing, and reporting qualitative research data.

This is a researcher focused workshop.

In this hands-on workshop you will learn how to use the Python programming language to create a python variable-number, string, list, dictionary, and tuple; how to create Python functions, condition statements and iteration; how to install a python package; how to manage your files and workspace; how to read comma-separated data files; how to create plots using Seaborn package. This workshop takes approximately 1.5 hours to complete.

This is a researcher focused workshop.

In this hands-on previously recorded Zoom workshop you will learn to use the R programming language to create vectors, factors, and data frames; Manage your files and workspace; Control R functions; Read comma-separated data files; Select variables and observations; Create plots and perform basic statistical analyses. This workshop takes approximately 2 hours to complete.

This is a researcher focused workshop.

R’s ggplot2 package lets you visualize data and easily make complex graphs. In this hands-on previously recorded Zoom workshop you will learn the grammar of graphics approach and create a wide range of graphs including histograms, bar charts, dot plots, pie charts, line charts, QQ plots, box plots, and scatter plots with various types of fit lines. You will also learn how to easily create graphs divided up by groups and mosaic combinations. This workshop takes approximately 2 hours to complete. A recommended prerequisite for this course is the R: Basics workshop.

This is a researcher focused workshop.

In this hands-on workshop you will learn how to use R to discover useful patterns in your data and to make accurate predictions. Topics will include data preparation, automated variable selection, cross-validation methods, model tuning, and several popular modeling approaches including decision trees, random forests, gradient boosting machines, and neural networks. We will use the popular tidymodels package which integrates the various steps and standardizes them across many modeling methods. This workshop takes approximately 8 hours to complete. A recommended prerequisite for this course is the R: Basics workshop.

This is a researcher focused workshop.

In this hands-on previously recorded Zoom workshop you will learn how to blend your research writing, programming, and output into a single word processing style document using the R Markdown language in RStudio. This style of work is known as Reproducible Research since every phase of the analysis can be recreated by “knitting” a single document. The output document can be HTML with an interactive table of contents, or Word or PDF formats. While we will be using it with the R language, Markdown commands apply to many other languages such as Python and SAS. This workshop takes approximately 2 hours to complete. A recommended prerequisite for this course is the R: R: Basics workshop.

This is a researcher focused workshop.

In this hands-on previously recorded zoom workshop, you will learn to program common statistical analysis including frequency tables, descriptive statistics, chi-square, correlation, linear regression, t-tests, and analysis of variance with posthoc tests. This workshop takes approximately 2 hours to complete. A recommended prerequisite for this course is the R: Basics workshop.

This is a researcher focused workshop.

In this brief previously recorded 30-minute presentation, we will discuss the various software and services that Research Computing Support offers Students, Faculty, and Staff members. We cover how RCS provides assistance with analyzing data via statistical consulting, qualitative analysis, GIS support, content analysis, and data mining. This includes how we can help with: Designing studies, experiments, and determining sample size; Create, deploy, and analyze web survey data; Provide co-authorship, and several other research related topics. In addition, we discuss the research software available on campus and how to get assistance.

This is a researcher focused workshop.

SAS is one of the most popular software packages for data analysis and visualization. The SAS Output Delivery System (ODS) Graphics procedures SGPLOT, SGPANEL, and SGSCATTER are powerful but easy to use tools for creating publication quality graphs and charts from your empirical data. This workshop is geared towards researchers who require publication quality graphics along with a working knowledge of the SAS software package. A recommended prerequisite for this course is the SAS: Programming Basics workshop. This workshop takes approximately 1 hour to complete.

This is a researcher focused workshop.

SPSS is a popular package for data analysis and graphics. This workshop shows you how to run basic statistical procedures and graphics in SPSS, as well as how to interpret the output. Topics include exploring your data to find errors, extreme values and sparse categories; examining the distributions of continuous variables; measuring the association between two continuous variables; comparing two groups on categorial and continuous variables and where to go to learn more. This workshop takes approximately 1 hour and 15 minutes to complete.

This is a researcher focused workshop.