Cleaning an untidy dataset
Deliverable: A genuinely messy real dataset cleaned with a documented script, so that running the script on the raw file reproduces the tidy result exactly.
Course
R Programming for Statistics
R taught for statistics rather than as a second Python — data wrangling, visualisation, testing and regression, with the discipline to interpret a result honestly.
- Duration: 3 months
- Classroom · Online live
- Level: beginner to advanced
What you will be able to do
- Write R using vectors, data frames and functions confidently
- Clean and reshape untidy data with dplyr and tidyr
- Produce publication-quality charts with ggplot2
- Run and correctly interpret the common hypothesis tests
- Fit and diagnose linear and logistic regression models
- Produce a reproducible report that regenerates from the raw data
Who this course is for
- Graduates in any discipline, statistics and economics backgrounds especially
- Researchers and academics analysing their own data
- Analysts who want stronger statistical grounding than Excel provides
- School-level mathematics is sufficient to begin
Syllabus
7 modules · 3 months
Module 1. R fundamentals
- Vectors, factors, lists and data frames
- Indexing, recycling and the gotchas that catch newcomers
- Functions, apply and writing your own
- Packages, projects and reproducible setup
Module 2. Data wrangling with the tidyverse
- Filter, select, mutate, arrange and summarise
- Grouping and per-group summaries
- Joins, and what each one keeps
- Pivoting between wide and long data
Module 3. Visualisation with ggplot2
- The grammar of graphics and why it is worth learning
- Aesthetics, geoms, scales and facets
- Choosing a chart that answers the question
- Publication-quality output, labelling and export
Module 4. Descriptive statistics and distributions
- Centre, spread and shape
- Distributions and what they imply
- Sampling, standard error and confidence intervals
- Reading a summary without over-reading it
Module 5. Hypothesis testing
- Null hypotheses and what a p-value does and does not mean
- t-tests, chi-square and ANOVA
- Assumptions, and checking them before trusting a result
- Multiple comparisons and the error nobody notices
Module 6. Regression
- Simple and multiple linear regression
- Interpreting coefficients honestly
- Residual diagnostics and what a bad plot is telling you
- Logistic regression for binary outcomes
Module 7. Reproducible reporting
- R Markdown and knitting a document
- Parameterised reports
- Version control for analysis
- Sharing work somebody else can rerun
Tools and technologies you will use
- R and RStudio
- tidyverse — dplyr, tidyr, ggplot2
- R Markdown and Quarto
- Statistical packages for testing and modelling
Projects you will build
Exploratory analysis with charts
Deliverable: A set of ggplot2 visualisations that answer four stated questions about a dataset, each with a written caption explaining what it shows and what it does not.
Statistical test with interpretation
Deliverable: A hypothesis test run correctly with assumptions checked, reported with the effect size and a written interpretation that avoids the common overclaims.
Regression report
Deliverable: A reproducible R Markdown report fitting a regression model, including residual diagnostics and an honest statement of the model's limitations.
Where this course can take you
- Data Analyst
- Statistical Analyst
- Research Associate
- Biostatistics Assistant
- Market Research Analyst
Duration, modes and fees
- Duration
- 3 months
- Delivery modes
- Classroom · Online live
- Fees
- Share your details for the current fee
- Fees vary by batch and delivery mode. R and RStudio are free and open source — there is no software cost.
Placement assistance
Every student gets placement assistance — that is what 100% placement assistance means. It is support for all, not a job for all. We do not promise a specific salary, a specific number of interviews, or placement at any named company, and you should be wary of anyone who does.
What is included
- A place in the monthly placement drive, held every third Saturday
- The readiness programme every second Saturday — mock interviews and preparation
- CV review against the specific roles you are targeting
- Portfolio review, so your project work is presented the way a reviewer will read it
- Access to the vacancy pool employers send directly to the Samyak network
- Guidance on which roles realistically fit your background and which do not
- A place in the next drive, with coaching, if you are not selected in this one
What is not included
- Any guarantee of a job, an interview, or a particular salary
- Placement at a named or partner company
- Applying to jobs on your behalf
- Support before you have completed the course and its project work
- Visa, relocation or overseas placement assistance
Taught as statistics, not as a second Python
There is a way to teach R that treats it as Python with different brackets. It produces people who can load a CSV and make a chart, and who cannot tell you whether their result means anything.
This course goes the other way. The language takes two modules; the rest is distributions, testing, regression and — most importantly — interpretation. That is what R is genuinely better at than the alternatives, and it is what an employer hiring an R user is actually buying.
What a p-value is not
Half of the hypothesis testing module is about restraint.
A p-value below 0.05 does not mean the effect is large, or important, or real. It does not mean the null hypothesis is false. Assumptions go unchecked, multiple comparisons go uncorrected, and a result gets reported with more confidence than the data can support — routinely, in published work.
Being the person in the room who spots that is worth more than any package.
ggplot2 is worth the learning curve
The grammar of graphics is strange for about a week and then it is difficult to work any other way.
Because charts are described rather than chosen from a menu, you can build something no chart wizard offers, faceted across a dimension, with consistent scales and honest labelling. Publication-quality output is the normal result rather than an achievement.
Reproducible, or it did not happen
Six months after an analysis, somebody asks how you arrived at a number.
If the answer involves remembering a sequence of manual steps, the analysis was never trustworthy. Every project here has to regenerate from the raw data by running a script, which is a habit that separates an analyst from someone who made a chart once.
Questions
R Programming for Statistics — frequently asked questions
Should I learn R or Python?
Python is the better general-purpose choice and dominates industry data science and machine learning. R is stronger for statistics, research and publication-quality visualisation, and is the standard in academia, pharmaceuticals and a lot of market research. If you are heading for industry, take Python. If your work is research or statistical, R is genuinely better at it.
Do I need a statistics background?
No. The statistics modules build from centre and spread upwards, and school mathematics is enough. What the course does insist on is interpretation — a p-value below 0.05 is not the end of an argument, and a good part of the teaching is about not overclaiming from a result.
Is R only useful in academia?
No, though academia is where it is most dominant. Pharmaceutical companies, market research firms, government statistical agencies and finance all use it substantially. The honest picture is that R has a narrower but deeper market than Python, and being genuinely good at statistics in it is a real differentiator.
What does reproducible reporting mean in practice?
That your analysis regenerates from the raw data with one command, rather than living in a spreadsheet nobody can retrace. It matters because six months later somebody will ask how you got a number, and "I remember doing some steps in Excel" is not an answer. The last module and project are built entirely around this.
Enquire about R Programming for Statistics
Three details is all we need. A course advisor will call you back.
Related courses
- Python Programming
Python taught as a working tool rather than a syntax tour — you finish able to read unfamiliar code, automate real tasks, call APIs and write tests that catch your own mistakes.
- Data Analytics
A structured path from spreadsheets to SQL, Power BI and Python, built around the four things a data analyst is actually paid to do — pull the data, clean it, analyse it and explain it.
- Data Science
Statistics, Python, SQL and machine learning taught as one connected discipline, with the emphasis on framing a problem correctly and knowing when a result is not real.
- Machine Learning
Classical machine learning done properly — feature engineering, honest evaluation, and the discipline to notice when a good score is an artefact rather than a result.
Next step
Talk to a course advisor
Tell us what you want to learn and we will help you pick the right course, batch and mode.