Easy R Programming Interview Questions

Show answer

R is an open-source language and environment built for statistics, data analysis and graphics. It is popular because of CRAN, which hosts over 20,000 packages, strong visualization with ggplot2 and built-in statistical functions. Analysts, biostatisticians and data scientists use it for reporting, modelling and research. For example, a pharma analytics team may use R for clinical trial analysis because many statistical methods are first released as R packages.

What are the atomic data types in R?

Show answer

R has six atomic types: logical (TRUE/FALSE), integer (5L), double or numeric (3.14), character ('text'), complex (2+3i) and raw (bytes). typeof() returns the internal type while class() returns the class attribute used for method dispatch. For example, typeof(5) is 'double' but typeof(5L) is 'integer'. Knowing types matters because mixing them in one vector triggers coercion to the most flexible type.

What is the difference between <- and = for assignment in R?

Show answer

Both assign values at the top level, but <- is the conventional assignment operator in R style guides, while = is mainly used to pass named arguments inside function calls. For example, mean(x = 1:10) passes an argument and does not create x in the workspace, but mean(x <- 1:10) creates x as a side effect. Using <- for assignment keeps code readable and avoids confusion between assignment and argument matching.

How do you install and load a package, and what is the difference between library() and require()?

Show answer

install.packages() downloads from CRAN once; library() loads it in every session. library() throws an error if the package is missing, while require() returns FALSE with a warning, so require() suits conditional checks. install.packages('dplyr') library(dplyr) if (!require('data.table')) { install.packages('data.table') library(data.table) }

What is a factor in R and when would you use one?

Show answer

A factor stores categorical data as integer codes with a set of labels called levels. Factors are used for variables like gender, city or rating, and modelling functions such as lm() turn them into dummy variables automatically. Ordered factors, created with factor(x, levels = c('Low', 'Medium', 'High'), ordered = TRUE), support comparisons. A common pitfall is as.numeric() on a numeric factor, which returns codes; use as.numeric(as.character(f)) instead.

A numeric column in your data has been read as character. How do you fix it?

Show answer

I first find the offending values, for example with unique(df$amount[is.na(as.numeric(df$amount))]), which reveals entries like '1,200', 'NA ' or '-'. Then I clean them, removing commas with gsub(',', '', x) and spaces with trimws(), and convert with as.numeric(). With readr I can also pass na = c('', '-', 'NA') and col_types so the column is parsed correctly at load time instead of being fixed later.

How do you create sequences and repeated values in R?

Show answer

seq(), rep() and the colon operator build most vectors. 1:10 # integers 1 to 10 seq(0, 100, by = 10) # 0, 10, ..., 100 seq(1, 2, length.out = 5) # 5 evenly spaced values rep(c('A', 'B'), times = 3) # A B A B A B rep(c('A', 'B'), each = 2) # A A B B seq_along(c('x', 'y', 'z')) # 1 2 3, safe for empty vectors seq_len(0) # empty, unlike 1:0

How do you handle missing values in calculations in R?

Show answer

Most summary functions return NA if any value is missing, so pass na.rm = TRUE or remove NAs first. x <- c(10, NA, 30, 40) mean(x) # NA mean(x, na.rm = TRUE) # 26.67 sum(is.na(x)) # count missing: 1 x[!is.na(x)] # drop NAs x[is.na(x)] <- median(x, na.rm = TRUE) # impute df[complete.cases(df), ] # rows with no NA

Why did you choose to learn R?

Show answer

I chose R because I enjoy working with data and statistics, and R is designed exactly for that. During my coursework I used it for a project analysing student performance, where dplyr and ggplot2 helped me move from raw data to clear insights quickly. I like that results are reproducible with R Markdown. I am also learning SQL and basic Python, so I can work comfortably in a mixed team and pick the right tool for each task.

How do you manage working directories and file paths in an R project?

Show answer

getwd() shows and setwd() changes the working directory, but a hardcoded setwd() breaks on other machines. RStudio projects with the here package build paths relative to the project root. getwd() library(here) sales <- read.csv(here('data', 'sales.csv')) write.csv(summary_df, here('output', 'summary.csv'), row.names = FALSE) file.exists(here('data', 'sales.csv')) list.files(here('data'), pattern = 'csv$')

Your analysis gives slightly different results every time you run it. How do you make it reproducible?

Show answer

Functions like sample(), rnorm() and random train-test splits depend on the random number generator. I call set.seed(42) at the start of the script so the same random sequence is generated every run, which lets reviewers reproduce results. For parallel code I use reproducible streams, for example future with seed = TRUE. I also record the R version, because the default sampling algorithm changed in R 3.6.0 and can change results between versions.

What is the difference between a vector and a list in R?

Show answer

An atomic vector holds elements of a single type, such as all numbers or all strings, while a list can hold elements of any type, including other lists, data frames or functions. c(1, 'a') coerces to character, but list(1, 'a', TRUE) keeps each type. Lists are used for nested data such as JSON responses and for returning multiple results from a function, which is why model objects like lm() output are lists.

What is a data frame in R?

Show answer

A data frame is a list of equal-length vectors, where each vector is a column with its own type. It is R's main structure for tabular data, similar to a spreadsheet or SQL table. str(df) shows column types, nrow() and ncol() give dimensions and summary() gives quick statistics. Tibbles from the tidyverse are modern data frames that print neatly, never convert strings to factors and do not partially match column names.

How is a matrix different from a data frame?

Show answer

A matrix is a two-dimensional atomic vector, so all elements share one type, usually numeric. A data frame allows a different type per column. Matrices are faster for linear algebra, such as %*% multiplication, solve() and distance calculations, while data frames suit mixed real-world data. For example, the output of cor() is a matrix, while customer records with names, dates and amounts belong in a data frame. as.matrix() and as.data.frame() convert between them.

Show different ways to index vectors and data frames in R.

Show answer

R indexes from 1 and supports positive, negative, logical and name-based indexing. x <- c(a = 10, b = 20, c = 30, d = 40) x[2] # 20 x[-1] # drop first x[c('a', 'd')] # by name x[x > 15] # logical filter which(x > 15) # positions df[df$age > 30, c('name', 'city')] df[order(-df$salary), ][1:5, ] # top 5 salaries

Dates in your file look like 25/12/2025 and as.Date() returns NA. What is wrong and how do you fix it?

Show answer

Indian files often store dates as dd/mm/yyyy, but as.Date() expects yyyy-mm-dd by default, so values become NA. I specify the format, as.Date(x, format = '%d/%m/%Y'), or use lubridate::dmy(x), which handles different separators automatically. Then I count NAs before and after conversion to confirm nothing failed silently. If the column came from Excel as serial numbers, I use as.Date(x, origin = '1899-12-30') instead.

What is an array in R?

Show answer

An array is a multi-dimensional generalization of a matrix, created with array(data, dim = c(rows, cols, layers)). All elements share one type. For example, array(1:24, dim = c(2, 3, 4)) has four layers of 2 by 3 matrices, and you index with arr[1, 2, 3]. Arrays are used for image data, simulation outputs and time by region by product cubes, though for business analysis long-format data frames are usually easier to work with.

How do you explain technical results to non-technical managers?

Show answer

I start with the business question and the answer in one sentence, then show one or two clear charts instead of tables of coefficients. For example, in a college project on customer churn, I said that customers with more than three complaints were twice as likely to leave, showed a simple bar chart and suggested a follow-up call process. I keep technical details in an appendix and invite questions so managers can decide confidently.

What is the tidyverse?

Show answer

The tidyverse is a collection of R packages that share a common design and grammar for data science. Core packages include readr for importing, dplyr for transformation, tidyr for reshaping, ggplot2 for visualization, purrr for functional programming, stringr for text, forcats for factors and lubridate for dates. library(tidyverse) loads the core set. Its consistent function names and the pipe make analysis code readable, which is why most R analytics teams use it.

What are the main dplyr verbs?

Show answer

dplyr has a small set of verbs for data transformation: filter() keeps rows that meet conditions, select() picks columns, mutate() creates or changes columns, arrange() sorts rows, summarise() reduces groups to summary rows and group_by() defines groups for the next operations. Others include distinct(), rename(), relocate(), count() and slice(). Each takes a data frame as the first argument and returns a data frame, so verbs chain naturally with the pipe.

What are the principles of tidy data?

Show answer

Tidy data follows three rules: each variable is a column, each observation is a row and each type of observational unit forms a table. For example, a sales sheet with separate columns for Jan, Feb and Mar is not tidy, because month is a variable spread across columns. Reshaping it into month and sales columns makes it tidy. Tidy data works directly with dplyr, ggplot2 and modelling functions, reducing reshaping effort later.

Write dplyr code to find total and average sales by city, sorted by total.

Show answer

Group, summarise and sort: library(dplyr) city_summary <- sales %>% filter(!is.na(amount)) %>% group_by(city) %>% summarise( orders = n(), total = sum(amount), avg_order = mean(amount), .groups = 'drop' ) %>% arrange(desc(total)) head(city_summary, 10)

How do you split one column into several, or combine columns into one?

Show answer

tidyr provides separate_wider_delim() to split and unite() to combine. library(tidyr) df <- tibble(full_name = c('Amit Kumar', 'Neha Singh'), dob = c('12-05-1999', '03-11-2000')) df %>% separate_wider_delim(full_name, delim = ' ', names = c('first', 'last')) df %>% separate_wider_delim(dob, '-', names = c('d', 'm', 'y')) %>% unite('dob_iso', y, m, d, sep = '-')

Finance says 120 transactions are missing from your report compared with their system. How do you find them?

Show answer

I load both datasets and standardise the key, for example trimming spaces and matching data types for transaction_id. Then anti_join(finance, report, by = 'transaction_id') lists transactions in finance but not in my report, and the reverse anti_join shows extras. I examine the missing rows for patterns such as a date range, status or filter in my code. Usually the cause is a timezone cut-off or an excluded status.

What is the difference between mutate() and summarise()?

Show answer

mutate() adds or modifies columns and keeps the same number of rows, so each row gets a value. summarise() collapses each group into one row of summary values. For example, group_by(city) %>% mutate(city_avg = mean(sales)) adds the city average to every row, useful for comparing each row with its group, while group_by(city) %>% summarise(city_avg = mean(sales)) returns one row per city.

What is the grammar of graphics in ggplot2?

Show answer

The grammar of graphics describes a plot as layers built from components: data, aesthetic mappings that link variables to visual properties like x, y and colour, geoms that draw shapes, stats that transform data, scales, coordinate systems, facets and themes. In ggplot2 you start with ggplot(data, aes(x, y)) and add layers with +, such as geom_point(). This makes complex charts systematic and easy to modify.

Which chart types would you use for comparing categories, trends and distributions?

Show answer

For comparing categories I use a bar chart, sorted by value, or a dot plot when there are many categories. For trends over time I use a line chart with time on the x-axis. For distributions I use histograms or density plots for one variable and boxplots or violin plots to compare groups. For relationships between two numeric variables I use a scatter plot. I avoid pie charts beyond three or four slices.

Write ggplot2 code for a monthly sales line chart with points.

Show answer

A basic time-series line chart: library(ggplot2) ggplot(monthly, aes(x = month, y = sales)) + geom_line(colour = 'steelblue', linewidth = 1) + geom_point(size = 2) + scale_y_continuous(labels = scales::label_comma()) + scale_x_date(date_labels = '%b %Y', date_breaks = '2 months') + labs(title = 'Monthly Sales', x = NULL, y = 'Sales (INR)') + theme_minimal()

Write code to show the distribution of order values using a histogram and a boxplot by category.

Show answer

Histogram for one variable, boxplot to compare groups: ggplot(orders, aes(x = order_value)) + geom_histogram(binwidth = 500, fill = 'grey60', colour = 'white') + labs(x = 'Order value (INR)', y = 'Orders') ggplot(orders, aes(x = category, y = order_value)) + geom_boxplot(outlier.alpha = 0.3) + coord_flip() + labs(x = NULL, y = 'Order value (INR)')

How do you save a ggplot chart for a report or presentation?

Show answer

ggsave() saves the last plot or a named plot with exact size and resolution. p <- ggplot(df, aes(city, sales)) + geom_col() ggsave('output/sales_by_city.png', plot = p, width = 8, height = 5, dpi = 300) ggsave('output/sales_by_city.pdf', plot = p, width = 8, height = 5) # vector, sharp at any zoom # svg for web use ggsave('output/sales_by_city.svg', plot = p)

Bars in your chart appear in alphabetical order instead of by value. How do you fix it?

Show answer

ggplot2 orders discrete axes by factor levels, which default to alphabetical. I reorder the factor by the value with forcats: aes(x = fct_reorder(city, sales), y = sales), and add coord_flip() or put the category on the y-axis for long labels. For a fixed custom order, such as months or ratings, I set levels explicitly with factor(x, levels = c(...)). fct_infreq() orders by frequency for count bars.

What is the difference between base R graphics and ggplot2?

Show answer

Base graphics use functions like plot(), hist() and barplot() that draw directly on a device, and you add elements with lines() or legend() step by step. They are quick for exploration. ggplot2 builds plots as objects from layers using the grammar of graphics, handles legends and facets automatically and is easier to style consistently. Base plots are fine for quick checks, while ggplot2 is preferred for reports and publications.

Your manager says your charts are confusing. How do you respond?

Show answer

I would thank them and ask which part was unclear, such as too many colours, missing labels or an unclear message. In my internship this happened with a crowded chart of twelve product lines. I simplified it by highlighting the top three, greying the rest, adding a clear title stating the finding and labelling values directly. My manager found the new version easy to present, and I now review every chart for one clear message.

When would you report the median instead of the mean?

Show answer

The mean uses every value and is sensitive to outliers, while the median is the middle value and is robust to them. I report the median for skewed data such as salaries, house prices, order values or delivery times, where a few extreme values pull the mean up. For example, if most employees earn 6 LPA and a few earn 80 LPA, the median describes a typical employee better. The mode suits categorical data.

What is the difference between variance and standard deviation?

Show answer

Variance is the average squared deviation from the mean, and standard deviation is its square root. Because variance is in squared units, standard deviation is easier to interpret since it uses the same units as the data. In R, var() and sd() both use the sample formula with n - 1 in the denominator. For example, delivery times with mean 30 minutes and SD 5 minutes are far more consistent than with SD 15.

Why does correlation not imply causation?

Show answer

Two variables can move together because of a third confounding variable, reverse causation or pure chance. For example, ice cream sales and drowning incidents are correlated because both rise in summer. To claim causation you need a randomized experiment such as an A/B test, or careful causal methods like difference-in-differences or regression with controls. In interviews, mention that observational correlations are useful for hypotheses but not proof of impact.

How do you produce descriptive statistics for each group in R?

Show answer

Combine summary functions with dplyr, or use quick helpers for exploration. summary(df$salary) quantile(df$salary, c(0.1, 0.5, 0.9), na.rm = TRUE) df %>% group_by(department) %>% summarise(n = n(), mean = mean(salary, na.rm = TRUE), median = median(salary, na.rm = TRUE), sd = sd(salary, na.rm = TRUE), iqr = IQR(salary, na.rm = TRUE)) skimr::skim(df)

What are skewness and kurtosis?

Show answer

Skewness measures asymmetry of a distribution. Positive or right skew has a long right tail, typical of income and order values, so the mean exceeds the median. Negative skew has a long left tail. Kurtosis measures how heavy the tails are compared with a normal distribution; high kurtosis means more extreme values, common in financial returns. In R, the moments or e1071 packages provide skewness() and kurtosis() functions.

How do you keep improving your statistics knowledge?

Show answer

I revise core concepts from a standard textbook and practise each method in R on public datasets, so I understand both the theory and the code. For example, after learning hypothesis testing, I reproduced published A/B test results from Kaggle datasets. I follow the R Weekly newsletter and read package vignettes. When I meet a method at work that I do not understand, I note it and study it that weekend.

How do you write a function in R with default arguments?

Show answer

A function is created with the function keyword and returns the value of its last expression, or an explicit return(). gst_price <- function(price, rate = 0.18, round_to = 2) { if (!is.numeric(price)) stop('price must be numeric') round(price * (1 + rate), round_to) } gst_price(1000) # 1180 gst_price(1000, rate = 0.05) gst_price(c(100, 250)) # vectorized

What is the difference between if, ifelse() and dplyr's if_else()?

Show answer

if is a control-flow statement that checks a single TRUE or FALSE value and runs one branch, used for program logic. ifelse() is vectorized and returns a vector of the same length as the condition, used for transforming data. dplyr::if_else() is a stricter vectorized version that requires both outputs to have the same type, preserves classes like dates and has a missing argument for NA conditions, making it safer than ifelse().

Explain for, while and repeat loops in R.

Show answer

A for loop iterates over elements of a vector, for (i in seq_along(x)), and is used when the number of iterations is known. A while loop runs while a condition stays TRUE, useful for convergence or polling. repeat runs indefinitely until break is called. next skips to the next iteration. In R, loops are fine when each iteration is heavy, but for element-wise work vectorized functions or purrr::map() are usually faster and cleaner.

Write a recursive function in R to calculate factorial and Fibonacci numbers.

Show answer

A recursive function calls itself with a smaller input and needs a base case. fact <- function(n) { if (n <= 1) return(1) n * fact(n - 1) } fact(5) # 120 fib <- function(n) { if (n <= 2) return(1) fib(n - 1) + fib(n - 2) } fib(10) # 55 # vectorized and efficient alternative factorial(5); cumprod(1:5)

How does the switch() function work in R?

Show answer

switch() selects one of several alternatives based on a value, which is cleaner than a long chain of if-else statements. With a character input, it matches names: switch(type, mean = mean(x), median = median(x), stop('Unknown type')). The final unnamed argument acts as the default. Empty arguments fall through to the next value, so switch(x, a = , b = 'AB') returns 'AB' for both. It is not vectorized, so use it for single values.

Describe a time you learned something from a teammate.

Show answer

Situation: during my final-year project, our group's R code ran very slowly on a large dataset. Task: I was responsible for the cleaning step. Action: a teammate showed me how to replace my nested for loops with vectorized dplyr operations and explained why loops were slow. I practised the approach on other parts of the code. Result: processing time fell from twenty minutes to under one, and I shared the lesson with the rest of the group.

How do you build a linear regression model in R?

Show answer

lm() fits linear models using formula syntax; summary() shows coefficients and fit. fit <- lm(sales ~ ad_spend + price + region, data = stores) summary(fit) coef(fit) confint(fit) # 95% CIs for coefficients predict(fit, newdata = new_stores, interval = 'prediction') par(mfrow = c(2, 2)); plot(fit) # diagnostic plots broom::tidy(fit) # coefficients as a tibble

Why do we split data into training and test sets, and what is cross-validation?

Show answer

A model must be evaluated on data it has not seen, otherwise performance looks better than reality. We train on one part, typically 70 to 80 percent, and evaluate on the held-out test set. Cross-validation, such as 5-fold or 10-fold, splits training data into folds, trains on all but one and validates on the remaining fold in turn, giving a more stable estimate for tuning. In tidymodels, initial_split() and vfold_cv() handle this.

What is overfitting and how do you prevent it?

Show answer

Overfitting happens when a model learns noise in the training data, so it performs very well on training data but poorly on new data. Signs are a large gap between training and validation scores. Prevention includes using simpler models, regularization such as lasso or ridge, cross-validation for tuning, limiting tree depth, getting more data and removing leaky or irrelevant features. Underfitting is the opposite: the model is too simple to capture the pattern.

Explain the confusion matrix and the metrics derived from it.

Show answer

A confusion matrix compares actual and predicted classes: true positives, false positives, true negatives and false negatives. Accuracy is the share of correct predictions. Precision is TP / (TP + FP), how many predicted positives are correct. Recall is TP / (TP + FN), how many actual positives were found. F1 is their harmonic mean. For a loan default model, recall matters if missing defaulters is costly; precision matters if rejecting good customers is costly.

What are the assumptions of linear regression?

Show answer

The main assumptions are a linear relationship between predictors and the outcome, independent errors, constant variance of errors (homoscedasticity), normally distributed errors for valid inference and no severe multicollinearity. In R, plot(fit) shows residuals versus fitted values, a Q-Q plot and leverage, and car::vif() checks collinearity. Violations can be addressed with transformations like log, robust standard errors or different models such as GLMs.

Why should we hire you as a fresher for an R analyst role?

Show answer

I have practical R skills, not just theory. In my projects I cleaned real datasets with dplyr, built ggplot2 dashboards and ran regression and hypothesis tests, and I can explain results simply. I also know SQL, which most analytics work needs. I learn quickly, accept feedback well and care about accuracy, so I double-check numbers before sharing them. I am ready to start contributing to reporting tasks from the first month.

What is R Markdown and why is it useful?

Show answer

R Markdown is a file format that combines narrative text written in Markdown with R code chunks. When rendered with knitr, the code runs and its output, such as tables and charts, is inserted into an HTML, PDF or Word document. This makes reports reproducible: when data changes, you re-render instead of copying charts manually. It is used for monthly reports, research papers, client deliverables and teaching material.

What is Shiny and how is a Shiny app structured?

Show answer

Shiny is an R package for building interactive web applications without writing HTML or JavaScript. An app has a ui that defines layout and inputs such as dropdowns and sliders, and a server function that contains R logic and creates outputs such as plots and tables. shinyApp(ui, server) runs the app. Analytics teams use Shiny for internal dashboards, what-if calculators and tools that let business users explore data.

What is renv and why should you use it?

Show answer

renv is a package that creates a project-specific library and records exact package versions in a renv.lock file. renv::init() sets up the project, renv::snapshot() updates the lockfile and renv::restore() installs the same versions on another machine or server. This prevents code from breaking when packages update, makes analysis reproducible months later and is essential before deploying scripts or Shiny apps to production.

How do you connect R to a SQL database and run queries?

Show answer

DBI provides a common interface, with drivers such as RPostgres, RMariaDB or odbc. library(DBI) con <- dbConnect(RPostgres::Postgres(), host = Sys.getenv('DB_HOST'), dbname = 'sales', user = Sys.getenv('DB_USER'), password = Sys.getenv('DB_PASS')) orders <- dbGetQuery(con, 'SELECT * FROM orders WHERE order_date >= $1', params = list('2026-01-01')) dbWriteTable(con, 'daily_summary', summary_df, append = TRUE) dbDisconnect(con)

Why should R projects use Git and how does RStudio support it?

Show answer

Git tracks every change to code, lets you revert mistakes, work on branches and collaborate through pull requests and reviews. RStudio has a built-in Git pane to stage, commit, diff, pull and push, and usethis::use_git() and use_github() set up repositories. Keep data files, secrets and outputs out of Git with .gitignore, commit the renv.lock file and write clear commit messages. Version control is expected even for analysis work.

The finance team wants the daily output as a formatted Excel file, not a CSV. How do you produce it?

Show answer

I use openxlsx2 or openxlsx to write a workbook with multiple sheets, formatted headers, number formats for INR, column widths, frozen panes and conditional formatting, all from R. writexl is a lightweight option for plain sheets without formatting. I create a template once with the team's preferred layout and generate the file in the scheduled script, then email or upload it to a shared drive. This removes manual formatting every day.

Where do you see yourself in three years in an R or analytics career?