Interpret figures without overstating the evidence
Was eine Kennzahl zeigt – und was nicht
Explain a numerical change accurately, identify the reference group and distinguish an observed association from a demonstrated cause.
A numerical statement depends on its denominator, the people included and the way observations were collected. A change from 10% to 5% is a fall of five percentage points and a relative reduction of 50%. Both descriptions are correct, but they answer different questions. Expressions such as lässt darauf schließen, deutet darauf hin and belegt differ in strength. If other conditions changed, describe the observation first and qualify any causal conclusion. A sample that excludes people who dropped out cannot automatically represent all potential participants.
Der Anteil sank von zehn auf fünf Prozent.
The proportion fell from ten to five percent.
Das entspricht einem Rückgang um fünf Prozentpunkte.
That corresponds to a decline of five percentage points.
Relativ zum Ausgangswert hat sich die Zahl halbiert.
Relative to the starting value, the number has halved.
Die veränderte Zusammensetzung schränkt die Vergleichbarkeit ein.
The changed composition limits comparability.
Daraus lässt sich noch kein eindeutiger Ursache-Wirkungs-Zusammenhang ableiten.
A clear cause-and-effect relationship cannot yet be inferred from that.
Die Befragung bildet die Erfahrungen der Abbrechenden nicht ab.
The survey does not reflect the experiences of those who dropped out.
A numerical statement depends on its denominator, the people included and the way observations were collected. A change from 10% to 5% is a fall of five percentage points and a relative reduction of 50%. Both descriptions are correct, but they answer different questions. Expressions such as lässt darauf schließen, deutet darauf hin and belegt differ in strength. If other conditions changed, describe the observation first and qualify any causal conclusion. A sample that excludes people who dropped out cannot automatically represent all potential participants.
Explain a numerical change accurately, identify the reference group and distinguish an observed association from a demonstrated cause.
A 50% relative fall is not a fall of 50 percentage points. A favourable figure does not by itself establish why the change occurred or whose experience was omitted.
Percentage points compare two percentages directly; relative change compares the difference with the starting value. In formal prose, distinguish a finding (Es wurde beobachtet …), a cautious inference (Dies könnte darauf hindeuten …) and a claim requiring stronger evidence (Dies belegt …).
Write a 250–300-word assessment of the German report below for a course-planning team. Explain the two correct ways to express the decline, identify two limits of the comparison and propose one additional observation. Separate evidence, inference and recommendation. Do not invent a control group or a proven effect.