What We Provide
A range of statistical consultancy services, including review work, provision of help desks, call-down contracts for extended work. We also advise on:

• statistical analysis and modelling
• planning and designing a study
• application of advanced statistical methodology
• use of statistical software

Partners
We are partners of The Enrichment unit at the University of Southampton.

Why Choose Us?
Our staff are accomplished consultants and trainers, with decades of experience providing support at all levels, from basic to advanced methodologies.
As applied statisticians, we are able to use the best statistical tools for the task. Our staff are experts in all major statistical packages, so we can easily adapt to your software needs.
Our clients have included:
AB Group
London Grid for Learning
National House Building Council
Therapeutic Solutions
Royal Horticultural Society

Contact Us
Please contact us to discuss how we can help:
Email: [email protected]
Telephone: +44 (0) 7914 383568

​Special offers
Introductory rates for PhD students in Higher Education.

We are experienced users of the following packages:

General applications:
R, SAS, SPSS, Stata

Power studies:
G*Power, nQuery, PASS.

Bayesian approach:
WinBUGS, OpenBUGS, JAGS

Spreadsheets:
Excel, Calc, Sheets

Linear Mixed Models for interval responses with clusters, i.e. random effects. Normal and lognormal distributions.

GLMM for counts and categorical responses with clusters. Negative binomial, Poisson, binomial and multinomial distributions. From subject-specific effects to population-averaged effects.

Time-to-event responses: Cox model, stratification, frailty. Counting process for time-varying covariates; piecewise Cox model. Weibull Proportional Hazards and Accelerated Failure Time model. Competing Risks model.

Sample Size for superiority, non-inferiority and equivalence studies in cross-over and longitudinal designs.

Models for Repeated Measures. Summary statistics approach, applications of Mixed Models: marginal (covariance pattern) and subject-specific (random coefficients) models.

Bayesian approaches: model fitting, MCMC diagnostics, posterior distributions, Credible and HPD Intervals. Model selection, model checking, predictive hierarchical models.

Evaluation of policy impact: causal approach for observational studies. Average Treatment Effects. Propensity Score Matching and doubly robust matching methods for continuous, binary, counts and time-to-event outcomes.

Model-based Small Area Estimation methods for normal, lognormal, counts, rate and binary data; ideal for mapping tasks.

Design of Experiments for assay optimisation: Central Composite, Box-Behnken, Fractional Factorials. Response Surface Methods.