The median (as well as quartiles, deciles, and percentiles) of a grouped frequency distribution is determined graphically using Ogives (Cumulative Frequency Curves):
By drawing a horizontal line from $N/2$ on the cumulative frequency axis to intersect the ogive and dropping a perpendicular to the horizontal axis, or
By the point of intersection of the 'less than' ogive and the 'more than' ogive.
In contrast, the mode is determined graphically from a Histogram.

A continuous frequency distribution of grouped data is represented graphically by a Histogram, where class intervals are taken on the horizontal X-axis and class frequencies on the vertical Y-axis, with adjacent rectangles whose areas are proportional to frequencies.

Why Option 2 is correct: A histogram is specifically designed to graphically depict continuous grouped frequency distributions without gaps between intervals.
Why Option 1 is incorrect: A pie chart is used to show proportions/percentages of categorical or discrete data, not continuous grouped frequency distributions.
Why Option 3 is incorrect: Pictograms use pictures or icons to represent discrete counts or qualitative data.
Why Option 4 is incorrect: Cartograms represent statistical data on geographical maps.

Principles of histogram construction:

• If class widths are unequal, keeping height directly proportional to absolute frequencies distorts the visual area.
• For unequal class intervals, the area of each rectangle must represent the class frequency. Hence, $\text{Area} = \text{Width} \times \text{Height} \propto \text{Frequency}$, which implies $\text{Height} \propto \frac{\text{Frequency}}{\text{Class Width}}$ (frequency density).
• Succeeding class frequencies have no mathematical relevance to current rectangle height.
• Succeeding class widths are not used in adjusting heights.