Combine filters and keep separate group summaries

Some questions require more than one condition. For example, a record must have group F AND city Chennai. Check both fields, then update only when the full condition matches.

AND, OR and NOT #

A AND B requires both. A OR B requires at least one, including the case where both are true. NOT A reverses the truth value. Use parentheses to make combined conditions clear. A comparison such as letterCount > 4 already produces True or False.

Worked example: F records from Chennai #

The matching IDs are S01 and S03, so count = 2. Their maths sum is 125 and average is 62.5. The original lecture finds 5 F cards from Chennai in its 30-card dataset; that is a different dataset.

Applying the two AND filters in either order selects the same records. In one pass, once the first AND condition is false, the second does not need to be evaluated to decide the result. For OR, a true first condition is enough to accept the card.

Date ranges include their boundaries #

The first half of a year runs from 1 January through 30 June, inclusive. With valid month/day values, months 1–6 meet this condition. The lecture reports 8 M and 7 F records in that half-year, totaling 15. The answer is computed from data, not assumed to be half just because the interval is half a year.

Two averages require four group variables #

FCount ← 0; FSum ← 0
MCount ← 0; MSum ← 0
For each valid marks card X:
    If X.group = "F":
        FCount ← FCount + 1
        FSum ← FSum + X.mathsMarks
    Else if X.group = "M":
        MCount ← MCount + 1
        MSum ← MSum + X.mathsMarks
Compute each average only if its own count > 0

For the lecture data, MCount = 17, MSum = 1220, FCount = 13 and FSum = 951. The averages are approximately 71.7647 and 73.1538. These describe that dataset; they do not establish a universal group difference. One exceptional maximum is not a substitute for comparing all observations.

Tutorial examples use exactly the same patterns #

The store totals come from 567+276 = 843 and 340+354 = 694. A zero total in one store means no matching spending was accumulated there; it is different from an unknown or missing amount.

Source: Lecture 1.3; Tutorial 1.2 — Tutorial for Lecture 3. Small teaching datasets and diagrams are prepared for these notes; lecture and tutorial totals are identified explicitly.