Research and statistical analysis
The right technique, the assumptions checked, the result defensible.
Who this is for
International research organisations, institutional research teams, and doctoral and masters researchers who need analysis that will survive review.
The work
What we actually do
Choosing a statistical technique is the easy part. Knowing which assumption your data has quietly violated, and what that does to your conclusion, is the part that takes years.
We work across hypothesis testing and experimental design, linear and logistic regression, multilevel and mixed effects models for clustered survey data, survival analysis, factor analysis and principal components, structural equation modelling, time series and forecasting, propensity score matching and difference in differences, complex survey weighting and design based estimation, and non parametric methods when the assumptions do not hold.
You receive the analysis, the code that produced it, the diagnostics, and a written account of what the data can and cannot support. If a reviewer challenges the method, you will be able to answer.
What you receive
- Analysis plan agreed before any modelling begins
- Cleaned, documented analysis dataset
- Reproducible code in R, Python or Stata
- Assumption diagnostics, stated plainly
- Results write up with tables and figures ready for publication
- A walkthrough call, so you can defend every number yourself
Tools
Start with a conversation
Describe the problem and the decision it has to support. We will tell you honestly whether we can help.
Start a project