Data Analysis

Give your data legs

Data formatting, programming, statistical analyses, graphics, and publication submissions can be overwhelming. Will your results stand up to peer review?

We give your data legs with advanced statistical analysis and publication support. Our specialties range from database management and programming to GIS geospatial data processing, advanced modeling applications, and sophisticated analytics using MS Access, ESRI ArcGIS and R statistical software.

Examples of our analytical work:

A repeated measures mark-resight approach to evaluating winter distribution and survival of Stone’s sheep

Applying zero-inflated mixed models to evaluate factors associated with seabird bycatch in commercial salmon fisheries

Linear regression mixed-effects modeling of horn growth in 2 bighorn sheep ecotypes to evaluate potential selective effects of trophy hunting

Analytical experience includes database management and programming; geospatial data processing; advanced habitat modeling applications with MS Access, ESRI ArcGIS and R; population estimation based on stratified sampling and mark-resight analyses;  survival and mortality risk analyses using Kaplan-Meier staggered entry methods;  estimation of seasonal and annual home ranges using minimum convex polygon and fixed probability kernel methods; habitat supply modeling and risk/disturbance modeling using Bayesian Belief Networks; habitat use assessments based on multivariate logistic regression and resource selection functions; and multivariate mixed-effects modeling using R statistical software to understand ecology and population dynamics of focal species.


Our Analytical Approach

We use advanced statistical techniques to analyze large, complex data sets, giving your data legs to stand up to peer review in scientific publications.
Our analytical specialties include:
▪    Advanced software applications in R and ArcGIS – useful when a cookie-cutter, out-of-the-box tool just won’t do the job

▪    Non-parametric statistics – hardly anything is normally distributed!

▪    Complex data distributions:

▪    Mixed-effects analyses for managing imbalanced data sets and autocorrelation in repeated measures

▪    Variants of Poisson data distributions and logistic regression for count and presence-absence analyses

▪    Mixture models for zero-inflated continuous data, typical in counts and occupancy modeling of aggregated species with patchy spatial distributions

▪    Resource selection functions that combine all of the above!


Comments are closed.