Python vs R for Python Developers A Practical Comparison

scores = [72, 85, 90, None, 64]

valid = [s for s in scores if s is not None]
print(valid[0])                     # first item: index 0
print(sum(valid) / len(valid))      # mean

doubled = [s * 2 for s in valid]   # needs a comprehension
print(doubled)

def grade(s):
    return "Pass" if s >= 70 else "Retry"

print([grade(s) for s in valid])

scores <- c(72, 85, 90, NA, 64)

valid <- scores[!is.na(scores)]
print(valid[1])                     # first item: index 1
print(mean(scores, na.rm = TRUE))   # mean, ignoring NA

doubled <- valid * 2               # vectorised, no loop
print(doubled)

grade <- function(s) ifelse(s >= 70, "Pass", "Retry")
print(grade(valid))                 # works on the whole vector

import pandas as pd

sales = pd.DataFrame({
    "city": ["Delhi", "Noida", "Delhi", "Pune", "Noida"],
    "amount": [1200, 800, 1500, 950, 1100],
})

summary = (sales
    .query("amount >= 900")
    .groupby("city", as_index=False)
    .agg(total=("amount", "sum"), orders=("amount", "count"))
    .sort_values("total", ascending=False))
print(summary)

library(dplyr)

sales <- tibble(
  city   = c("Delhi", "Noida", "Delhi", "Pune", "Noida"),
  amount = c(1200, 800, 1500, 950, 1100)
)

summary <- sales |>
  filter(amount >= 900) |>
  group_by(city) |>
  summarise(total = sum(amount), orders = n()) |>
  arrange(desc(total))
print(summary)

import pandas as pd
import statsmodels.formula.api as smf

df = pd.DataFrame({
    "ad_spend": [10, 15, 20, 25, 30, 35, 40],
    "sales":    [52, 60, 71, 75, 88, 92, 101],
})

model = smf.ols("sales ~ ad_spend", data=df).fit()
print(model.summary().tables[1])   # coefficients, p-values

library(ggplot2)

df <- data.frame(
  ad_spend = c(10, 15, 20, 25, 30, 35, 40),
  sales    = c(52, 60, 71, 75, 88, 92, 101)
)

model <- lm(sales ~ ad_spend, data = df)
summary(model)                       # coefficients, p-values, R-squared

ggplot(df, aes(ad_spend, sales)) +
  geom_point() +
  geom_smooth(method = "lm") +
  labs(title = "Sales vs ad spend")

Should I learn R as a Python developer?

[ ] My target job listings mention R
[ ] My team or research group already uses R
[ ] I need advanced or niche statistical methods
[ ] I produce statistical reports for non-developers
[ ] I work in research, biostatistics or academia

0-1 ticks: stay Python-first; learn R only to read it
2-3 ticks: learn R basics + tidyverse in a few weeks
4-5 ticks: invest properly in R alongside Python

Week 1
- Install R and RStudio (or VS Code with R support)
- Vectors, data frames, functions, NA handling
- dplyr: filter, mutate, group_by, summarise

Week 2
- ggplot2: points, lines, bars, facets
- lm() and glm() basics
- Quarto or R Markdown report from one dataset

Redo one pandas project in R and compare.

pandas                     -> dplyr / base R
df.head()                  -> head(df)
df.describe()              -> summary(df)
df[df.x > 5]               -> filter(df, x > 5)
df.assign(y = ...)         -> mutate(df, y = ...)
df.groupby('g').agg(...)   -> group_by(df, g) |> summarise(...)
df.sort_values('x')        -> arrange(df, x)
df.merge(other, on='id')   -> left_join(df, other, by = 'id')
len(df)                    -> nrow(df)