Coding classes in Mawaleh that put every observation back into the median

The shortcut that returns 4

Stall A contributes USD 1, USD 2 and USD 100, so its median is USD 2. Stall B contributes USD 3, USD 4 and USD 5, so its median is USD 4. Stall C contributes the nine values from USD 6 through USD 14, so its median is USD 10.

The three stall medians are 2, 4 and 10. Taking their median returns USD 4. That calculation gives each stall one vote even though Stall C contributes three times as many observations as each smaller stall.

stalls = {
  "A": [1, 2, 100],
  "B": [3, 4, 5],
  "C": [6, 7, 8, 9, 10, 11, 12, 13, 14]
}

stall_medians = [2, 4, 10]
median_of_medians = 4

The pooled median is 8

Pooling all 15 prices produces the ordered sequence 1 through 14 followed by 100. The eighth observation is USD 8, so USD 8 is the overall median. The group summaries cannot reconstruct that order because they discarded twelve of the fifteen observations.

from statistics import median

all_prices = [x for prices in stalls.values() for x in prices]
overall_median = median(all_prices)  # 8

A weighted average of the stall medians would not repair this. Medians are order statistics, not totals that combine through group weights. The correct approach is to retain the raw observations or use a valid distributed quantile method with sufficient rank information.

Free first class

Placement is tested before payment.

Group size

Five to eight compatible learners work together.

Private format

One student works with one teacher.

Frequency

Two weekly lessons normally make eight monthly.

Time difference

Oman is one and a half hours behind India.

Equipment

A laptop or desktop and stable internet are required.