Technologies / Data & Analytics

Statistics for scientists

The five-day Statistics for Scientists training course offers a comprehensive introduction to statistical methods used in scientific research. Participants will learn to design experiments, analyze data and interpret results using modern statistical tools.

Issues

Fundamentals of probability and statistics
Design of experiments and sampling
Parametric and non-parametric tests
Analysis of variance (ANOVA)
Linear and non-linear regression
Multidimensional methods
Time series analysis
Monte Carlo methods and bootstrap
Visualization of scientific data

Benefits

Ability to design scientific experiments
Ability to select and apply appropriate statistical methods
Knowledge of advanced data analysis techniques
Ability to interpret and present statistical results

Who is this training for?

Scientists and researchers from various fields

Doctoral and master's students

Data analysis specialists at research institutions

Those involved in research projects requiring statistical analysis

Prerequisites

Basic knowledge of mathematics at the university level
Experience in conducting scientific research

Training program

Probability — distributions (normal, t-Student, chi-squared, F), central limit theorem, parameter estimation
Descriptive statistics — measures of location and dispersion, quartiles, box plot, outlier detection
Experiment design — DOE (Design of Experiments), randomization, blocking, control groups, test power
Sample selection — sampling techniques (random, stratified, cluster), calculating minimum sample size (power analysis)
Statistical hypotheses — H0 vs H1, Type I and Type II errors, p-value, confidence intervals, effect size
Exercises: designing an experiment with power analysis and sample selection

Parametric tests — Student's t-test (one sample, two samples, paired), assumptions (normality, variance homogeneity)
Non-parametric tests — Mann-Whitney, Wilcoxon, Kruskal-Wallis, Friedman — when to apply instead of parametric
Chi-squared test — independence test, goodness-of-fit test, Yates correction, Fisher's exact test
Assumption verification — Shapiro-Wilk test (normality), Levene's test (homogeneity), data transformations
Multiple comparison correction — Bonferroni, Holm-Bonferroni, FDR (Benjamini-Hochberg), when to apply which
Exercises: selecting and conducting statistical tests on scientific data sets

One-way ANOVA — variance decomposition, F-test, post-hoc tests (Tukey, Scheffé, Dunnett), effect size (eta-squared)
Multi-factor ANOVA — interactions, main effects, factorial, repeated measures ANOVA, mixed design
ANCOVA — covariates, confounding variable control
Linear regression — simple and multiple, assumptions, model diagnostics (residuals, VIF, Cook's distance)
Nonlinear regression — logistic regression, Poisson regression, generalized models (GLM)
Exercises: building ANOVA and regression models on research data, result interpretation

Principal Component Analysis (PCA) — dimensionality reduction, scree plot, component interpretation, biplot
Factor analysis — exploratory (EFA) vs confirmatory (CFA), rotation (Varimax, Promax), factor loadings
Cluster analysis — k-means, hierarchical, DBSCAN, selecting the number of clusters (silhouette, elbow method)
Discriminant analysis — classification, LDA, cross-validation
Time series — decomposition (trend, seasonality, noise), autocorrelation, ARIMA models, forecasting
Exercises: PCA and cluster analysis on multivariate scientific data

Monte Carlo methods — simulation, confidence interval estimation, permutation tests
Bootstrap — non-parametric confidence interval estimation, bootstrap for regression, jackknife
Bayesian methods — Bayesian inference fundamentals, prior and posterior, comparison with frequentist approach
Scientific data visualization — effective visualization principles, publication charts (R/ggplot2, Python/matplotlib)
Statistical results reporting — APA standards, result tables, effect size reporting, reproducibility
Exercises: visualization and reporting of statistical analysis results in publication format

Delivery Methods

Online

Convenience of participating from anywhere
Interactive live sessions with trainer
Materials available for 30 days
No travel costs

On-site

Direct contact with trainer and group
Intensive hands-on workshops
Networking with other participants
Full focus on learning

Frequently asked questions

What are the prerequisites for this training?

For Statistics for scientists we recommend: Basic knowledge of mathematics at the university level; Experience in conducting scientific research.

What is the format and duration of this training?

The training lasts 5 days and is available in online and on-site format. Sessions run from 9:00 AM to 4:00 PM. We can also customize the schedule to fit your team's needs.

Who is this training designed for?

This training is designed for: Scientists and researchers from various fields; Doctoral and master's students; Data analysis specialists at research institutions.