Regression Analysis

Causal Thinking in Thermal Comfort

September 17, 2024
Ruiji Sun
by Ruiji Sun. We demonstrate the importance of causal thinking by comparing two linear regression approaches used in thermal comfort research: Approach (a), which regresses thermal sensation votes (y-axis) on indoor temperature (x-axis); Approach (b), which does the reverse, regressing indoor temperature (y-axis) on thermal sensation votes (x-axis). From a correlational perspective, they may appear interchangeable, but causal thinking reveals substantial and practical differences between them. Using the same data, we found Approach (b) leads to a 10 °C narrower than the conventionally derived comfort zone using Approach (a). This finding has important implications for occupant comfort and building energy efficiency. We highlight the importance of integrating causal thinking into correlation-based statistical methods, especially given the increasing volume of data in the built environment.

Theo Snow

Availability: By appointment only

Consulting Areas: Python, R, SQL, SAS, Databases & SQL, Data Manipulation and Cleaning, Data Science, Data Visualization, Geospatial Data, Maps & Spatial Analysis, Machine Learning, Mixed Methods, Qualitative methods, Surveys, Sampling & Interviews, Regression Analysis, Means Tests, Software Output Interpretation, Other, Excel, Git or Github, RStudio, RStudio Cloud, SAS, Tableau

Sakina Dhorajiwala

Availability: By appointment only

Consulting Areas: Python, R, Stata, LaTeX, Data Manipulation and Cleaning, Data Visualization, Mixed Methods, Qualitative Methods, Surveys, Sampling & Interviews, Regression Analysis, Excel, Git or Github, RStudio

Iñigo Parra

Availability: By appointment only

Consulting Areas: Python, R, LaTeX, Data Manipulation and Cleaning, Data Science, Data Visualization, Deep Learning, Digital Humanities, Machine Learning, Natural Language Processing, Social Network Analysis, Regression Analysis, Means Tests, Bash or Command Line, Excel, Gephi, Git or Github, Qualtrics, RStudio, Overleaf

Emma Lasky

Availability: By appointment only

Consulting Areas: Python Programming, R Programming, Data Manipulation and Cleaning, Data Science, Data Sources, Data Visualization, Geospatial Data, Maps & Spatial Analysis, Mixed Methods, Regression Analysis, ArcGIS Desktop, Online or Pro, Excel, Git or Github, QGIS, RStudio, RStudio Cloud

Anusha Bishop

Availability: By appointment only

Consulting Areas: Python, R, Cloud & HPC Computing, Data Sources, Data Visualization, Geospatial Data, Maps & Analysis, Machine Learning, Research Design, Cluster analysis, Experimental design, Hierarchical Models, High dimensional statistics, Means Tests, Nonparametric methods, Regression Analysis, Software Output Interpretation, Spatial statistics, Bash or Command Line, Excel, Git or Github, RStudio

Kurt Soncco Sinchi

Consultant
Civil Engineering

First generation student and looking to improve and apply Data Science core concepts into social impactful projects, as well as trying to leverage the information from previous cases for better insights of society. Focused on infrastructure and its impact under natural disasters.

Pratik Sachdeva, Ph.D.

Data/Research Scientist, Senior Consultant, and Senior Instructor

I am staff at the Social Sciences D-Lab. I received my Ph.D. in the Physics department at Berkeley. My research lies in the realm of theoretical/computational neuroscience, which aims to use mathematical and computational tools to better understand how neural systems operate and process information. My projects include using information-theoretic techniques to study how neural variability impacts information processing in neural circuits and investigating the statistical issues that impede the interpretation of parametric models of neural activity.

Beyond research, I'm...

Yue Lin

Data Science Fellow 2024-2025
Political Science

Yue is a Ph.D. student in Political Science at the University of California, Berkeley, with a Designated Emphasis on Political Economy. Using mixed methods, she studies foreign lobbying, geopolitical risk, and economic security to understand when, how, and why multinational corporations become the targets and weapons of state power rivalry.

Sohail Khan

Data Science Fellow 2024-2025
School of Information

Hey everyone, I’m Sohail - a 1st years Master’s student studying Data Science at the I-School. I am interested in the intersection between Computer Science, Data Science, and Cognitive Psychology and using these tools to understand, discover, and drive the development of assistive technologies.

I have experience building with brain computer Interfaces, developing distributed data processing applications, and am currently working on a large scale archival project aimed at preserving the history and memory of resistance movements through an embedding based...