Because grouped bar charts rely on a shared baseline for accurate comparison, starting the y-axis at anything other than zero distorts visual differences. In geoscience and engineering fields, truncating the axis is considered misleading because it exaggerates gaps between values.