Bray-Curtis Dissimilarity Calculator
This page calculates Bray-Curtis Dissimilarity on time series data (Baselga, 2013; Lo et al., 2024), useful in detecting compensatory changes between multiple variables. The dissimilarity is calculated successively between each pair of consecutive time points, and everything updates as you type. Click a column heading to rename the variable.
| Time 1 | ||||||
|---|---|---|---|---|---|---|
| Time 2 | ||||||
| Time 3 | ||||||
| Time 4 |
Only accepts non-negative values.
Bray-Curtis dissimilarity
= replacement + nestedness; captures changes in multiple variables between 2 measurements.
Replacement (compensatory changes)
Lines that cross indicate replacement: one variable rises while another falls.
Nestedness-resultant (unidirectional changes across all variables)
Lines that change in the same direction, beyond what replacement can account for.
What is Bray-Curtis Dissimilarity?
Bray-Curtis dissimilarity's replacement component (also known as Simpson's dissimilarity) detects compensatory changes. In ecology, this may describe a simultaneous increase in one species (e.g., wolves) and decrease in another (e.g., rabbits) over time. In psychology, it may describe a person's switch from one emotion regulation strategy (e.g., distraction) to another (e.g., social sharing) over time.
The index ranges from 0, when two time points are identical, to 1, when they share nothing. It splits into two subcomponents that sum back to it: the replacement component described above, and the nestedness-resultant component, which captures unidirectional change, where every variable moves in the same direction. That decomposition follows Baselga (2013); its application to psychological time series follows Lo et al. (2024).
Applying the Dissimilarity to Your Time Series Data?
Have more than 6 variables or more than 4 time points? Use the R code in my tutorial for experience sampling data, which handles data of any size.
If you find this calculator useful for your research, please consider citing both the paper that first outlined the methodology in ecology (Baselga, 2013) and my paper (Lo et al., 2024) that translated it for use in psychology and time-series data:
- Baselga, A. (2013). Separating the two components of abundance-based dissimilarity: Balanced changes in abundance vs. abundance gradients. Methods in Ecology and Evolution, 4(6), 552–557. https://doi.org/10.1111/2041-210X.12029
- Lo, T. T., Van Lissa, C. J., Verhagen, M., Hoemann, K., Erbaş, Y., & Maciejewski, D. F. (2024). A theory-informed emotion regulation variability index: Bray–Curtis dissimilarity. Emotion, 24(5), 1273. https://doi.org/10.1037/emo0001344