By Figran Team. We are associated with Figran, a commercial AI-assisted visual creation service. This original paper-and-table teaching activity was prepared with AI assistance and checked for internal arithmetic consistency. All cards, counts and coordinate ranges below are invented. They are not measurements, a classroom trial, or evidence of a product or learning benefit. No software account is needed.
A tall histogram bar can represent a high number of cards per coordinate unit without containing more cards than a wider, shorter bar. To read it, retain the bin boundaries and the vertical-unit rule with the source counts. This activity makes those choices explicit, then changes one bin without inventing the positions of its individual cards.
1. Declare what the rectangle represents
Use a coordinate measured in units called u. Each card is assigned one coordinate, but the individual coordinates are not supplied. Each card has equal weight. Source Card H assigns all 12 cards to the three disjoint bins below; there are no additional cards outside their combined range [0, 7).
The interval notation [a, b) includes a and excludes b. Its width in coordinate units is w = b - a, with w greater than zero. These are numeric intervals, not three categories with arbitrary decorative bar widths.
For this activity's count-density histogram, the height is h = c / w, where c is the assigned integer count in that bin. Label the horizontal axis with its coordinate unit and the vertical axis "cards per u". The rectangle's area in these coordinate units is w × h = c.
The visual area comparison assumes linear axes and a zero baseline. Coordinate area is not a count of screen pixels. A cropped baseline or a transformed axis needs its actual rules checked; multiplying pixel lengths does not recover the source count. This is a declared drawing contract, not a claim that every chart uses the same height convention.
2. Retain the complete invented source
| Bin | Coordinate interval (u) | Width w (u) | Assigned cards c | Count-density height h (cards per u) | Relative-frequency density (per u) |
|---|---|---|---|---|---|
| A | [0, 1) | 1 | 2 | 2 | 1/6 |
| B | [1, 3) | 2 | 6 | 3 | 1/4 |
| C | [3, 7) | 4 | 4 | 1 | 1/12 |
The counts add to 12. The count-density heights add to 2 + 3 + 1 = 6, which is not the count total. Recover the counts using each bin's own width: 1 × 2 = 2, 2 × 3 = 6, and 4 × 1 = 4. Their areas add to 12.
Bin A is twice as high as Bin C under this contract, yet A contains 2 cards and C contains 4. The first bin has half as many cards. Bin B contains the most cards here, but the example does not establish that the tallest bar always contains the most cards when bin widths differ.
Keep the source table available even if you draw rectangles on paper. The heights are derived display values; the assigned counts and boundaries are the source for this exercise. Do not replace those fields with a screenshot.
3. Normalize the areas, not the height sum
For a relative-frequency density display, use the known total N = 12 and set the height to g = c / (N × w). Its vertical unit is "relative frequency per u". Its area is w × g = c / N.
The final column gives exact density heights 1/6, 1/4 and 1/12. Multiply by the widths to obtain areas 1/6, 1/2 and 1/3. These areas sum to one because the bins exhaust the declared inventory. The density heights themselves sum to 1/2, which is not an error and is not the total relative frequency.
Count density and relative-frequency density have different vertical units. Changing the normalization divides every count-density height by 12 in this source. It does not change the boundaries or the assigned counts. Keep fractions exact for this exercise; if you display decimal approximations, state the rounding rule and retain the exact source.
These relative frequencies describe this invented finite inventory. They are not estimates of an unknown population density, a confidence statement, or an explanation of why cards were assigned to a bin.
4. Try seven review tasks
- For Card H, calculate each bin's width, count-density height and count recovered from its rectangle area. Compare the height sum with the area sum.
- Repair this caption: "A is twice as high as C, so A has twice as many cards." Give the correct height and count comparisons, with their units.
- Calculate the three normalized density heights and their areas. Explain which sum equals one and why the other sum need not do so.
- Under the declared interval convention, where would coordinates exactly 1, 3 and 7 belong? Explain what must change if a later source is to include a card at 7. Do not silently add it to the existing inventory.
- An extension explicitly supplies 1 card in [3, 5) and 3 cards in [5, 7), partitioning C's 4 cards. Calculate the two new count-density heights and areas. Could Card H's original aggregate count alone have supplied these two counts?
- A separate Source Card E assigns counts 2, 6 and 4 to equal-width bins [0, 2), [2, 4) and [4, 6). Calculate its count-density heights. Why are height ratios proportional to count ratios in E? Do not treat E as a re-binning reconstructed from H's unknown individual positions.
- Another source explicitly declares relative-frequency density, with the same widths 1, 2 and 4 and exact heights 1/6, 1/4 and 1/12, but omits its total N. Do these fields uniquely identify counts 2, 6 and 4? Give two different consistent integer-count inventories and identify the missing information.
Write the source version, vertical-unit rule and any extra information beside each answer. An unlabelled vertical axis would leave even more unresolved information than Task 7, which explicitly supplies its normalization rule.
5. Worked answers
1. Widths 1, 2 and 4 give count-density heights 2, 3 and 1. Areas 2, 6 and 4 recover the assigned counts and sum to 12. Heights sum to 6. Adding values in "cards per u" does not substitute for adding counts; each bin contributes its height multiplied by its own width.
2. A's height is 2 cards per u and C's is 1 card per u, so the height ratio is 2:1. Their counts are 2 and 4 cards, so the count ratio is 1:2. One repair is: "A has twice C's count density per coordinate unit, but half C's number of cards, because C is four times as wide." This comparison is about the declared fictional bins, not their quality or a product result.
3. The normalized heights are 2/(12 × 1) = 1/6, 6/(12 × 2) = 1/4, and 4/(12 × 4) = 1/12. The corresponding areas are 1/6, 1/2 and 1/3, which sum to one. Heights sum to 1/2. Exhaustive disjoint bins make the relative-frequency areas add to one; they do not require the height values to add to one.
4. Coordinate 1 belongs to B, and coordinate 3 belongs to C. Coordinate 7 belongs to none of H's bins, because the last upper boundary is excluded. A new source that includes 7 must explicitly revise the covered range or its final endpoint convention and its source inventory. Some reporting systems use a different final-bin boundary rule; that possibility does not silently change this activity's written [3, 7) rule or add a thirteenth card to H.
5. Both new bins have width 2. Their heights are 1/2 and 3/2 cards per u, and their areas are 1 and 3 cards. These areas retain C's total of 4. The original table's count 4 did not tell us the within-bin positions, so it could not determine the partition 1 and 3. Other assignments, such as 2 and 2, would also preserve that aggregate count. The extension supplies extra fictional source information; it is not evidence recovered from C's former height.
| Extension bin | Coordinate interval (u) | Width (u) | Supplied cards | Count-density height (cards per u) |
|---|---|---|---|---|
| C-one | [3, 5) | 2 | 1 | 1/2 |
| C-two | [5, 7) | 2 | 3 | 3/2 |
The other bins remain assigned 2 and 6 cards, so the extended source still totals 12. The four heights now sum to 2 + 3 + 1/2 + 3/2 = 7, while the four areas still sum to 12. Changing a partition can change the height sum without changing its total inventory.
6. E's widths are all 2, so its heights are 1, 3 and 2 cards per u. Each count is its height multiplied by the same factor 2. Thus height ratios match count ratios under this count-density rule when widths are equal. That proportionality does not make the height values themselves counts. E is a separate invented assignment; H's aggregate bins do not determine counts for these changed boundaries.
7. The three relative-frequency areas are still 1/6, 1/2 and 1/3, but their total count is unknown. N = 6 with counts 1, 3 and 2 is consistent, as is N = 24 with counts 4, 12 and 8. N = 12 with counts 2, 6 and 4 is another possibility. These are compatible arithmetic assignments, not proof of what the source actually contained. Request the actual total and original counts if exact reconstruction is needed. Do not fill the missing total from H simply because the density labels look the same.
6. Leave a recoverable caption and review record
For H's count-density display, one supported caption is: "Source Card H: 12 invented cards in three disjoint intervals covering [0, 7). Bar heights are assigned counts divided by bin widths in u; rectangle areas in the declared coordinate units represent counts. The individual card coordinates are not supplied."
For its normalized display, replace the height definition with "assigned count divided by total count 12 and bin width" and identify areas as relative frequencies. The count total remains part of the source even though areas now sum to one.
Retain source version, boundaries and endpoint rule; counts and total; horizontal unit and widths; vertical-unit and normalization rule; baseline and axis scale; and any supplied within-bin information. Use these fields to review a figure before interpreting a tall rectangle. Do not infer individual card positions, a within-bin pattern, or an average coordinate from these aggregate counts.
7. Credit, limits and reuse
This is a proposed arithmetic and source-reading activity. It supplies no real measurements, sampling claim, classroom outcome or evidence of improved scientific or product performance. A real histogram needs its actual data definitions and binning rules, and may require subject-specific statistical advice. This fictional contract cannot validate it.
Creator's commercial website: figran.net. Public contact: support@figran.net. We are affiliated with Figran; this is not an independent review of the service.
The original teaching text and invented source tables are offered under Creative Commons Attribution 4.0 International. Credit Figran Team and this article title, link the licence and indicate changes. This licence does not cover Figran software, brand assets, Mataroa software or the platform interface.