This section is based on this repository: https://github.com/MitchellAcoustics/Soundscapy
Under this folder you can find a simple tutorial that helps you:
-
- Scatter Plots
- Distribution (Density) Plots
- Jointplot
-
- Radar/Spider Plot
This section presents a case study using Python to analyze and represent data collected via field measurements and recordings, as well as data collected in a preliminary laboratory experiment. The study "Python Workflow for Soundscape Assessment: Case Study of an Adaptive Façade in Detmold, Germany" was developed as a contribution to the "IDS Report on Sustainable Façades - Summer Semester 2023" as documentation for a digital workflow - therefore focused on software features rather than on answering the broader research question regarding the influence of adaptive façades on the urban soundscape.
EFA is a statistical technique used in data analysis to uncover the underlying structure of a set of observed variables. The main goal of EFA is to understand the relationships among a large number of variables by reducing them into a smaller set of underlying factors. These factors represent the shared variance among the observed variables and can provide insights into the underlying constructs or dimensions being measured.
CFA is a subset of SEM that focuses on evaluating the measurement properties of latent constructs or factors. It is primarily used to assess the validity and reliability of a measurement instrument or questionnaire. CFA allows researchers to test whether the observed variables (indicators) in a dataset adequately reflect the underlying latent constructs they are intended to measure. CFA examines the relationship between the observed variables and the latent factors and estimates factor loadings, which indicates the strength of the relationship between the observed variables and the latent factors.
SEM is a more comprehensive statistical framework that encompasses CFA but goes beyond it. SEM includes not only the measurement model (CFA) but also a structural model that examines the relationships between latent variables (factors) and other observed variables. In other words, SEM allows researchers to simultaneously test the measurement model and the structural model, thereby examining the relationships and causal pathways between variables.