Read the Before and After

The raw table has five rows, but Tuesday has NA for orders. For this first pass, the rule is simple: drop the missing row and do not fill in a guess. After the drop, the plot has four dots. The summary table names the effect: five raw rows become four clean rows, one row is dropped, and the mean of the kept values is 6.

A tiny cleaning readout: make one missing-value decision, then show how the clean rows and plot differ from the raw load.

highlighted = computed this step

Load before cleaning

Start with 5 pinned rows; only 4 rows have numeric orders.

raw rows before cleaning\text{raw rows before cleaning}
Loaded Toy DatasetdayordersMon6TueNAWed9Thu3Fri6

Choose the cleaning rule

Drop the 1 missing row; do not invent a replacement value.

drop missing row\text{drop missing row}
Cleaning DecisionrowdecisionreasonTuedroporders is NAother rowskeeporders is numeric

Show the clean plot

The clean plot has 4 dots: Mon 6, Wed 9, Thu 3, Fri 6.

clean plot after dropping missing\text{clean plot after dropping missing}
Clean order counts as dot marksdayorders02468106Mon9Wed3Thu6Fri

State the effect

Cleaning changes the readout from 5 raw rows to 4 clean rows, with mean 6 over kept values.

cleaning effect\text{cleaning effect}
Effect of Dropping the Missing Rowreadoutvalueraw rows5clean rows4dropped rows1mean of kept values6guessed values0