  |
Sensometrics tutorials
The Sensometric Society in collaboration with the scientific committee of the 17th Sensometrics Conference will organize five sensometrics tutorials on Monday 3 of June, 2024. One day previous to the conference. The cost of the tutorial is of 50€ and need to be paid in advance using the conference platform. See bellow in each tutorial description.
The five Sensometrics tutorials will have a duration of 3 hours, and each tutorial have a capacity of 20 persons. Space on each tutorial will be limited and early booking is advised. Please note that the tutorials will be split in three tutorials during the morning, and two tutorials in the afternoon of the 3rd of June, 2024. Tutorials will take place at Pasteur Institute.
Click in each tutorial section to register, as they are split in morning or afternoon session.
Tutorials:
- Publication-quality data visualizations using the R tidyverse
- Create automatic and reproducible sensory reports with Quarto
- Sparse Factor Analysis of Sensory Data
- Introduction to R’s Shiny Applications Through Practical Examples
- A comprehensive guide to efficiently analyze Sensory data with XLSTAT
Tutorial 1: Publication-quality data visualizations using the R tidyverse
Instructors: Jacob Lahne & Leah Hamilton (Virginia State University, USA)
Monday June 3, at 9:00 - 12:00 (registration at 8:30)
Even medium-sized sensory studies often yield large amounts of noisy data with many variables. Good visualizations can condense these data so patterns are visible at a glance, but it’s all too easy to make plots that mislead or confuse. The free and flexible R programming language allows researchers to quickly generate and iterate on highly-customizable visualizations suitable for presentations and publications, but the number of possibilities can quickly become overwhelming. In this tutorial, we aim to teach participants how to make useful visualizations of highly multivariate data in R. The R skills covered, including data import, transformation, and visualization, are necessary for most multivariate statistical analyses and plots.
Participants will get hands-on experience coding in R, gaining familiarity with flexible techniques and packages. This tutorial will demonstrate the process and best practices for creating side-by-side plots for viewing many variables at once as well as “multivariate maps” from dimensionality-reducing analyses like Principal Components Analysis or Correspondence Analysis. It is not intended as a statistics course and will not dive deeply into the technical details of specific analyses, but by the end participants will visualize Check-All-That-Apply data using multiple simultaneously-generated bar plots and Correspondence Analysis biplots. These methods allow participants to visually represent ordinal or categorical data about stimuli, and so could also be used on data from open comment questions, existing texts, or many other kinds of sensory surveys.
In this tutorial, we will introduce the audience to ggplot2 and the rest of the tidyverse R packages with the aim of developing sufficient basic skills to visualize multivariate sensory and consumer data. We will provide a learning dataset for the analysis—a set of free response comments and overall liking scores from a central location test on berries. We will teach participants how to import, manipulate, and plot data using user-friendly, “tidy” R programming. All resources used in the tutorial are open-source and will remain available to attendees, including an R script covering the full workflow.
At the end of the tutorial, attendees will be able to prepare raw sensory data for common multivariate visual representations in R.
Duration 3 hours
Audience Sensory and consumer scientists who are interested in making better data visualizations using freely available/open-source software (R/RStudio).
Background Basic familiarity with data types, variables, functions, and installing/using packages in R/RStudio. Basic understanding of statistics is helpful but not required. We will email registered participants before the workshop with some basic setup requirements (R/RStudio software installation) and suggested “catch-up” reading for participants worried about their level of existing R knowledge.
Laptop This is a coding workshop, and so we ask all participants to bring a laptop with access to RStudio. We will ask for minimal pre-work (installation of R/RStudio).
Tutorial 2: Create automatic and reproducible sensory reports with Quarto
Instructor: Margot BRARD (Consultant & trainer at ThinkR, France)
Monday June 3, at 9:00 - 12:00 (registration at 8:30)
The R programming language offers sensory and consumer scientists a range of methods to analyze their data. Indeed, R seems to be the preferred language to make available new methods of analysis of sensory and consumer data, as evidenced by the large number of R packages dedicated to the analysis of sensory data. Data wrangling and data visualization play a crucial role in the data analysis process. But the communication of results and the way in which they are shared are also important. One of the workflows sometimes used to create sensory reports is the following one:
It presents some weaknesses, with a lot of pain points for sensory and consumer scientists in their daily practices:
- Time-consuming
- Copying and pasting errors
- Lack of standardization
- Lack of customizability of the analysis
- Lack of customizability of the design
In this tutorial, we will see how Quarto can help overcome these weaknesses and improve the daily lives of sensory and consumer scientists. We will see how it allows them to switch to this type of workflow for the creation of sensory reports:
A sensory report built with Quarto has the advantages of being reproducible with different types of datasets, and shareable as is with users. An update of the input data (as long as they have the same structure) and a recompilation of the Quarto document will automatically update the content of the report (tables, plots, etc.).
Tutorial 3: Sparse Factor Analysis of Sensory Data
Instructor: Vincent Guillemot (Institut Pasteur, France)
Monday June 3, at 9:00 - 12:00 (registration at 8:30)
This tutorial is designed for sensory and consumer scientists who want to improve their data analysis skills using component-based methods in R. The focus of this training session is to teach participants how to navigate through complex and noisy multivariate sensory datasets to extract relevant observations, assessors, properties, and variables. The main objective is to provide attendees with the expertise to carry out a range of sparse multivariate analyses in R, including sparse PCA, CA, and MCA.
During the course, attendees will work with specific packages created for sparse multivariate analysis (such as SPAFAC). They will be provided with datasets containing both quantitative and qualitative variables to apply sparse MCA, CA, and PCA methods. This hands-on approach will help them determine the optimal degree of sparsity and accurately interpret their findings. All the course materials, including an extensive R script, are open-source and will be available for use after the completion of the tutorial.
This three-hour session is particularly beneficial for sensory and consumer scientists looking to enhance their data visualization skills using R/Rstudio. Participants should have a basic understanding of R/RStudio, including data types, variables, functions, and package management. A fundamental grasp of statistical concepts will also be helpful. Participants must bring a laptop with R and Rstudio pre-installed. Pre-tutorial communication will provide detailed software installation guidelines and a list of suggested readings.
Tutorial 4: Introduction to R’s Shiny Applications Through Practical Examples
Instructor: Thierry Worch (FrieslandCampina, Netherlands)
Monday June 3, at 13:30 - 16:30
This tutorial provides an exploration into the integration of Shiny applications within the realm of sensory and consumer research using the R programming language.
Shiny is an interactive web application framework for R that offers a dynamic and user-friendly environment for visualizing and analyzing data. The tutorial aims to empower researchers to leverage Shiny's capabilities for creating engaging and interactive tools to their needs (e.g. data processing, data analysis, data visualization, reporting, etc.).
The tutorial will cover the key aspects of Shiny applications’ development (data processing and visualization, automated reporting) through a real-life example. Participants will gain hands-on experience, enabling them to create custom applications tailored to their needs.
Topics covered will include:
- Introduction to Shiny: An overview of Shiny's architecture and capabilities, with a focus on its relevance to sensory and consumer research.
- Building Interactive Dashboards: Step-by-step guidance on designing and developing interactive dashboards to visualize sensory data, consumer preferences, and trends.
- Data Integration and Processing: Techniques for integrating diverse datasets into Shiny apps, handling data (from processing to analysis/visualization and reporting).
- User Experience: Best practices for creating intuitive and user-friendly Shiny applications.
By the end of the tutorial, participants will possess the skills to harness the potential of Shiny applications for advancing their sensory and consumer research endeavors, ultimately facilitating more impactful data-driven decision-making in this dynamic field.
Duration 3 hours
Audience Sensory and consumer scientists who are interested in building their own analysis dashboard using a freely available/open-source software (R/RStudio).
Background Basic knowledge in R and the tidyverse framework, as well as RStudio is preferred. Basic understanding of statistics is helpful but not required. We will email registered participants before the workshop with some basic setup requirements (R/RStudio software installation).
Laptop This is a coding workshop, and so we ask all participants to bring a laptop with access to R, RStudio, and some of the relevant packages. We will ask for minimal pre-work (installation of R/RStudio).
Tutorial 5: A comprehensive guide to efficiently analyze Sensory data with XLSTAT
Instructor: Fabien Llobell & Efthalia Anagnostou (Lumivero, XLSTAT, France)
Monday June 3, at 13:30 - 16:30 (registration before 12:00)
Today, sensory analysts have little time to analyze their data. But this phase is crucial to making the right decisions. Whether it's choosing the right methods for the right type of data, or calculating the methods themselves, a lot of time is needed.
It's in this context that XLSTAT enables you to be guided in your choice of algorithms and to analyze your data easily without having to use code. A lot of time is therefore saved for the most crucial part of your data analysis: the conclusions. XLSTAT is a widely used software program that is particularly popular in the sensory world. Its main qualities are that it can be used with simple clicks, has more than 40 functions designed for sensory analysis and is integrated with Excel.
This tutorial will help you learn how to use XLSTAT for all your main data types: Panel, CATA, Just About Right, TDS, triangle tests, liking, etc. What's more, we'll be giving you a host of tips to help you get the most out of your work quickly and easily. You'll also learn how to interpret the results to make effective decisions. Particular attention will be paid to helping you choose the right method for your needs. Finally, you'll be able to go even further in your visualizations for ever-better quality reports with the Dataviz tool.
At the end of the tutorial, attendees will be able to analyze classical sensory task data with XLSTAT and know how to go further and improve their reports and skills.
Duration 3 hours
Audience Sensory and consumer scientists who are interested in going faster and further easily in their analysis by using XLSTAT efficiently.
Background Basic understanding of statistics is helpful but not required. We will email registered participants before the workshop with some basic setup requirements (XLSTAT installation).
Laptop This is a workshop on XLSTAT, and so we ask all participants to bring a laptop with access to XLSTAT. If necessary, a temporary license can be given.