At a glance
By Adam Turner
1. R is the go-to language for statistical analysis
A versatile programming language for statistical analysis, R was first developed in the 1990s to teach introductory statistics.
Created by data scientists for data scientists, R programming language is exceptionally strong at statistical analysis and data visualisation.
Rival language Python has a wider range of applications than R because it is more of a general-purpose programming language, but the jury is still out on which language is better for statistical analysis.
2. R is free
As free, open-source software, R is supported by a large online community that publishes “packages” to an online library, which facilitates ongoing expansion of R’s functionality.
Being free and open source contributes to the popularity of both R and Python by enabling widespread access and encouraging continuous improvement and innovation through collaborative development.
"R is great for answering ‘why’ and ‘what if’ questions, particularly when you need to show your working out."
Use of R does rely on writing lines of code, but RStudio provides a user-friendly graphical interface and tools for visualising data analysis.
3. It is ideal for answering “why” and “if” questions
R is well-suited to analysing complex causality. This makes it useful for accountants looking to understand the impact of multiple variables on an event and forecast what is likely to occur next, says Fahim Khondaker, partner and leader of data analytics and insights at BDO in Australia.
Tools such as Excel or Power BI suffice when it comes to simply presenting sales figures, Khondaker says, but R’s strength is in understanding the impact of a range of variables that influence those sales figures.
“R is great for answering ‘why’ and ‘what if’ questions, particularly when you need to show your working out,” he says. “For example, at BDO we use it to explore the impact of multiple variables on an event, such as understanding the impact of the weather conditions on sales figures.”
4. It excels at predictive modelling
“R is also useful for predictive modelling, especially when you want to understand the drivers behind that model, rather than simply have an AI [artificial intelligence] black box spit out an answer without explaining how it came to that conclusion,” Khondaker says.
"As the one doing the complex analysis, R makes it easier for you to determine the impact of different variables, which in turn helps you assist decision makers in determining their priorities."
To facilitate this, R is well-equipped to import raw data from a wide range of sources such as spreadsheets, ERP systems and websites. It then cleanses and transforms the data to provide better quality data, allowing for easier analysis, fewer errors and more reliable outcomes.
5. R has powerful visualisation tools
When it comes to conveying insight to others, R’s visualisation tools are more flexible and customisable than those of other tools such as Power BI, argues data scientist Michael Spear of BDO in Australia.
As one of R’s key strengths is revealing the “signal” from the noise within data, R’s powerful visualisation tools make it easier to see and convey the complex insights uncovered, Spear says.
“As the one doing the complex analysis, R makes it easier for you to determine the impact of different variables, which in turn helps you assist decision makers in determining their priorities,” he explains.
“Data journalists at media outlets use R to create visuals and explain their analysis to their readers, because it is so useful for clearly and elegantly conveying very technical analysis to a non-technical audience.”
6. Excel and Power BI skills can be applied to R
Accountants familiar with data manipulation in Excel will find many concepts transferrable to R.
Additionally, R’s extensive library of packages can streamline specific accounting-related tasks such as financial ratio analysis, reconciliations, predictive modelling and data visualisations, says Dr Richard Busulwa, senior lecturer in accounting at Swinburne University of Technology.
Similar to the plugins available for other software, R’s package library includes tools for more easily importing, cleansing, manipulating, analysing and visualising datasets.
“The fact that R is free, easy to set up and benefits from Excel or Power BI knowledge makes it easier to start. However, to fully harness its vast power, you need to invest more time learning its syntax and methods,” Busulwa says.
“The great range of free online tutorials further ease getting started with R, while the extensive range of R packages often eliminates the need to build solutions from scratch, easing or streamlining challenging analysis tasks.”
7. It may help bridge the data science and decision-making gap
While R helps accountants take their quest for insights to the next level, realistically most accountants do not require R’s advanced level of statistical analysis to perform their role, Khondaker says.
Even so, while they might not need to learn how to use R, it is important to understand its capabilities in order to work more effectively with data scientists in support of the business.
Within every organisation, there are people who speak the language of business and people who speak the language of data, Khondaker says, and the modern accountant needs the skills to translate between them.
"The fact that R is free, easy to set up and benefits from Excel or Power BI knowledge makes it easier to start. However, to fully harness its vast power, you need to invest more time learning its syntax and methods."
“We need accountants who don’t shy away from complexity and are advocates of complex analysis where appropriate,” he says.
“Having an appreciation for the power of R, even if they are not experts in its use, allows accountants to be that critical translation layer who bridges the gap between complex data science and effective decision-making.”
