Coding classes for kids in Puno, Peru

Source: MINEDU, Evaluación Censal de Estudiantes, published in INEI, Compendio Estadístico Perú 2026, tomo 1, table 5.53. Region-wide results; they describe groups of pupils, not any single child. In 2019 Puno's maths figure was 45.6 per cent.

From INEI: tables 1 and 3 of the 2025 census demographic indicators (the census was referenced to 15 September 2025, and table 3 lists those actually enumerated), the 2017 sociodemographic profile, table 1.9, for the city, and the 2017 directory of populated places for the settlement and its altitude (census day 22 October 2017). The rows overlap and are never added.

In 2025 the department counted 100,965 children aged 6 to 11. Many live far from the city: the province of Puno alone stretches from the lake shore to districts such as Acora, with 24,734 people. For a child in Acora or Chucuito, a class in the city centre is a long trip; a live online class is the same distance from everywhere. Families in Juliaca, a separate city in the province of San Román, can join in exactly the same way.

MINEDU, Edades normativas (age on 31 March of the school year); Ley N.° 28044 (primaria lasts six years); CNEB (no programming area). The last column is our own plan and is adjusted to each child.

In every class the mentor asks questions before giving answers: what did you expect, what happened, which block or line made the difference? We teach children to think a problem through, not to copy a solution, and a child who can explain a fix has learned more than one who was handed it. Families comparing options across the country can read the national page on online coding classes for kids in Peru.

import csv

riders = {}
with open("202608-PDE-REP-00301.csv", encoding="utf-8-sig") as f:
    for row in csv.DictReader(f, delimiter=";"):
        if row["ANIO"] == "2025" and row["TRAMO"] == "SUR":
            trip = (row["ESTACION_ORIGEN"], row["ESTACION_DESTINO"])
            riders[trip] = riders.get(trip, 0) + int(row["CANTIDAD PASAJEROS"])

top = sorted(riders.items(), key=lambda item: item[1], reverse=True)
for trip, people in top[:4]:
    print(trip[0], "->", trip[1], people)

The program did exactly what it was told: it treated Cusco to Puno and Puno to Cusco as two different things, because the pair (Cusco, Puno) is not the same as the pair (Puno, Cusco). For a ticket that is true. For a question about routes, it is not: both trains run on the same track between the same two stations. The children need a way to make the two pairs into one key.

import csv

riders = {}
one_way = {}
with open("202608-PDE-REP-00301.csv", encoding="utf-8-sig") as f:
    for row in csv.DictReader(f, delimiter=";"):
        if row["ANIO"] == "2025" and row["TRAMO"] == "SUR":
            a, b = row["ESTACION_ORIGEN"], row["ESTACION_DESTINO"]
            people = int(row["CANTIDAD PASAJEROS"])
            route = tuple(sorted((a, b)))
            riders[route] = riders.get(route, 0) + people
            one_way[(a, b)] = one_way.get((a, b), 0) + people

top = sorted(riders.items(), key=lambda item: item[1], reverse=True)
for route, people in top[:2]:
    print(route[0], "<->", route[1], people)
print("To Puno:", one_way[("CUZCO WANCHAC", "PUNO ESTACION")])
print("From Puno:", one_way[("PUNO ESTACION", "CUZCO WANCHAC")])

The trick is one line: tuple(sorted((a, b))). Sorting the two station names puts them in alphabetical order, so Cusco to Puno and Puno to Cusco both become (CUZCO WANCHAC, PUNO ESTACION), and their passengers land in the same box: 24,714 people in 2025. Version 2 also keeps the one-way counts, and they tell a second story. More people rode towards Puno, 14,413, than back to Cusco, 10,301. The file does not say why. Children suggest that some travellers leave Puno another way, by bus or by plane from Juliaca, which is a good guess to test, not a fact.

Data: OSITRAN, Tráfico de Pasajeros, FERROCARRIL, datosabiertos.gob.pe, licence ODC-By. The sums are ours, over every service type on the pair. The file has no rows for this pair in 2021, and we do not guess the reason. In 2025 the busiest month towards Puno was October, with 1,682 riders, and the quietest February, with 504. A second Puno stop, which the file calls PUNO MUELLE (muelle means pier), had 197 riders from Cusco and 58 back in 2025.

1. Six and seven, paper and pencil. Draw Cusco and Puno with a line between them. Put five counters on trains going one way and three coming back. How many people used the line?
2. Eight and nine, Scratch. Make a train sprite that goes to Puno and back. Count each trip with one variable, then with two, one for each direction.
3. Ten to twelve, Python. Start with Version 1 and say in your own words why the same route appears twice. Then type the sorted line yourself before reading Version 2.
4. A step further. Use the monthly rows to find the month in 2025 when the two directions were closest. What might make people travel both ways in the same month?

Peru: UTC minus five hours throughout the year, without daylight saving (tz database, America/Lima; INACAL is the official timekeeper). India: UTC plus five hours thirty minutes, also unchanged through the year.

Young children usually concentrate more easily earlier in the day, so many families pick a morning slot during the January to March holidays and an after-school slot later. Neither country changes its clocks, so the hour you choose in February is the same hour in November. When you book, you pick the time in Puno time and the system shows the mentor theirs.