Back to the on-screen lesson ·
Explain how re-aggregation can change a mapped conclusion.
Paper packet. Every task here also exists on screen, where it is checked automatically; answers written on paper are not assessed by Nydus. When you are back at a device, enter your answers there.
Recalculate indicators under changed boundaries and distinguish a zoning effect from a real change in people or places.
Recall that a percentage is a count divided by a relevant total, multiplied by one hundred. When combining areas, add numerators and denominators before dividing. Do not average percentages unless their denominators are equal or the intended statistic is explicitly an average of area rates. You will also preserve spatial positions while redrawing groups, because sorting values is not the same operation as changing boundaries.
| Term | What it means |
|---|---|
| MAUP | Dependence of an area-based result on the scale or zoning of aggregation. |
| Zoning effect | A changed result produced by a different arrangement of reporting boundaries. |
| Scale effect | A changed result produced by larger or smaller reporting units. |
| Ecological fallacy | An unsupported inference from an area relationship to individuals. |
The modifiable areal unit problem, often shortened to MAUP, arises when results depend on how observations are grouped into areas. A zoning effect changes the arrangement of boundaries while retaining a similar number of reporting areas. A scale effect changes the size or number of areas. Neither requires any person to move or any underlying measurement to change. The reporting model changes, and with it the pattern visible in the summary.
A boundary is therefore an analytical choice as well as a line. Administrative districts may be appropriate for allocating services, while watersheds suit drainage questions and commuting zones suit labor-market questions. There is no single grouping that answers every geographic question. The task is to choose a defensible unit, show how the indicator is calculated, and test whether a consequential conclusion survives another plausible grouping. All blocks, populations and service records here are invented, so that the effect of boundary choice can be isolated.
Another way: table
| Block position | Households | Without reliable transit |
|---|---|---|
| Northwest A | 100 | 80 |
| Northeast B | 100 | 20 |
| Southwest C | 100 | 80 |
| Southeast D | 100 | 20 |
West and east districts group A+C and B+D. North and south districts group A+B and C+D. These are two different zonings of the same four blocks.
Read the four-block table before interpreting any shaded map. The numerator counts households without reliable transit; the denominator counts all surveyed households in the same block. Block A has 80 out of 100, so its rate is 80 percent. Block B has 20 out of 100, or 20 percent. All blocks have equal denominators in this introductory example, which makes the arithmetic simple but should not be assumed in other datasets.
Combine A and C into a western district. Add 80 and 80 to get 160 households without reliable transit, then add 100 and 100 to get 200 households. The district rate is 160 divided by 200, or 80 percent. The eastern district has 40 out of 200, or 20 percent. A west-east map displays a strong disparity.
Now combine A with B and C with D. Each new district has 100 households without reliable transit out of 200, or 50 percent. The north-south map appears equal. No household gained a bus service. Equality at the reporting scale has concealed a pattern that cuts across the new districts.
Both district maps contain two areas, but their boundaries group different blocks. That is the zoning effect. Combining all four blocks into one city instead changes the scale of reporting. The city contains 200 households without reliable transit out of 400, giving 50 percent. A one-color city map cannot reveal the internal west-east contrast.
Subdividing a reporting area can reveal variation, but it also creates smaller denominators. A rate based on two households is more sensitive to one changed observation than a rate based on two thousand. More detailed reporting can therefore improve spatial visibility while reducing stability or threatening privacy. The decision is not simply to use the smallest possible units.
Keep the original counts whenever legitimate access permits. If only rounded district percentages survive, it may be impossible to reconstruct block patterns. In a reproducible workflow, preserve the source geography, the lookup from source units to target units, and the date of each boundary file. A changed district name does not necessarily mean the boundary stayed the same.
Consider a different hypothetical pair: a small block has 10 households, of which 8 lack reliable transit, while a large block has 90 households, of which 18 lack it. Their rates are 80 percent and 20 percent. Averaging the two rates gives 50 percent, but the combined household rate is 26 out of 100, or 26 percent. The large block contributes most households and cannot be assigned the same weight as the small one when estimating household experience.
The unweighted mean could be useful if the unit of concern is a block selected with equal probability. That is a different question. Specify whether the statistic describes people, households, establishments or areas. An apparently technical denominator choice encodes who or what counts equally.
Check whether numerator definitions also match. One survey may call transit reliable if a bus arrives within ten minutes of schedule, another if a route operates every day. Combining these counts without harmonization produces a rate whose meaning is unclear. Consistent geography cannot repair inconsistent measurement definitions. Document both kinds of comparability before presenting a merged map.
Suppose districts with higher mean income also have higher transit use. This could reflect denser employment, service frequency or a mixture of resident groups. It does not establish that higher-income individuals are the transit users. Within each district, lower-income residents could account for most transit journeys. An area-level relationship and an individual-level relationship can differ without any arithmetic contradiction.
To examine individual behavior, obtain appropriately protected individual or household evidence, or narrow the claim to areas. Do not infer identity, preferences or hardship from the shade assigned to a person's district. A neighborhood category is a property of a summary, not a diagnosis of everyone living there.
Cultural and political maps need the same care. A majority-language district can contain substantial linguistic diversity, and a district boundary can divide a shared cultural landscape. A map showing one label per district selects one aspect of identity. Use multiple sources and allow overlapping affiliations when the inquiry concerns lived experience rather than a single administrative category.
First verify conservation. The total number of households and the total number without reliable transit must stay the same across complete regroupings. In the four-block case, every valid partition totals 400 households and 200 without reliable transit. If a redraw changes these totals, a block was omitted, duplicated or split incorrectly.
Next inspect the geography. Are the alternative districts contiguous if contiguity is required? Do they represent plausible service areas, or were they designed only to erase the disparity? A sensitivity analysis should compare defensible alternatives, not arbitrary combinations selected after seeing a preferred result. When a block straddles a new boundary, assigning it by its centroid is an approximation; document that rule and assess whether it matters.
Finally, compare the conclusion, not only the numbers. The statement that half the city's households lack reliable transit survives both zonings. The statement that all districts have equal rates does not. Report robust findings separately from boundary-sensitive ones. If a funding decision depends on a threshold, show whether plausible regrouping moves places above or below it and recommend examining the underlying communities before allocating resources.
A fictional grant targets districts where more than 60 percent of households lack reliable transit. Under the west-east zoning, the western district qualifies at 80 percent. Under the north-south zoning, neither district qualifies because each is at 50 percent. The 160 underserved western households have not disappeared. A planner who relies only on the newer districts would miss a concentrated need visible in the underlying blocks.
A better review preserves the eligibility rule but asks whether additional small-area evidence is permitted and relevant. Present both zonings, the original counts and the service-network map. Explain which households the proposed route would actually reach. The calculation cannot decide the grant's legal rules, but it can expose a mismatch between the reporting geography and the service problem. This is a hypothetical allocation exercise, not advice about an actual program.
An invented municipality redraws two districts after annexation. A language survey later reports that the share using Language R at home rose from 40 to 55 percent in one district. Before describing cultural change, compare the old and new boundaries. If the new district includes a previously separate settlement where Language R is common, part of the change reflects composition.
Recalculate both years on a common geographic basis where data allow, and inspect migration, births, deaths and survey wording. Place names, religious buildings and market practices can enrich the interpretation, but none is a perfect substitute for people's own accounts of identity. A careful report distinguishes a changed statistical boundary from a changed cultural practice and states where evidence is insufficient to separate them.
New district colors can result from changed boundaries, changed class breaks, changed measurements or real changes in conditions. Inspect the metadata to distinguish them. MAUP does not mean every map is deceptive or every aggregate is useless. It means a geographic summary has a unit of analysis whose influence must be understood. An aggregate claim can be accurate at its own scale while becoming misleading when applied to a household or a different zoning.
Add western underserved counts.
80 + 80 = 160 households.
A and C form the western district.
Add their denominators.
100 + 100 = 200 households.
All households remain represented.
Divide the combined counts.
160 / 200 * 100 = 80%.
A district rate uses the district denominator.
Compare with the east.
40 / 200 * 100 = 20%.
The same definition reveals a west-east gap.
Form the northern district.
A+B gives 80+20 = 100 underserved.
Only the grouping changes.
Calculate its rate.
100 / 200 * 100 = 50%.
The two northern blocks contain 200 households.
Repeat for the south.
C+D also gives 100 / 200 = 50%.
The southern composition matches.
Check total conservation.
100+100 = 200 underserved citywide.
No underlying household count changed.
Interpret the apparent equality.
Two equal district rates conceal a west-east disparity.
This is a zoning effect.
Read the small block.
8 of 10 households: 80%.
A small denominator creates a high area rate.
Read the large block.
18 of 90 households: 20%.
Most households live here.
Pool the original counts.
8+18 = 26; 10+90 = 100.
Counts retain the correct weights.
Compute the household rate.
26 / 100 * 100 = 26%.
It differs from the unweighted area mean.
Identify the wrong target.
(80+20)/2 = 50% describes two area rates.
That does not describe a random household.
Report the unit explicitly.
26% of the combined households lack reliable transit.
The statement identifies its population.
Add the numerator.
70+30 = 100.
Count underserved households.
Add the denominator.
100+100 = 200.
Count every household.
Compute the percentage.
The same 35 blocks are grouped into new districts. District rates change although no records change. What has been demonstrated?
Complete the calculation. A fictional district contains 3 equal blocks, each with 24 households lacking reliable transit. What is the district numerator?
Multiply the equal block count by the number of blocks.
r
Add the underserved count from every included block.
Check that no block is counted twice.
Check the units and the stated comparison.
Regrouping must conserve household totals.
Do not treat the numerator alone as a percentage.
Keep the result within the supplied observations.
Keep underlying counts fixed when testing a boundary effect and state the unit to which the rate applies.
Order a valid comparison across a district redraw.
Number the steps in order (write the number in the box):
A new map shows less inequality after district boundaries change. Select valid checks.
This task has no paper form; do it on a device.
Match each change to its analytical implication.
| Zoning effect | Scale effect | Unsupported individual inference | |
|---|---|---|---|
| Two districts redrawn into two different districts | |||
| Ten blocks combined into one city value | |||
| District transit rate assigned to every resident |
A fictional district contains 2 equal blocks, each with 32 households lacking reliable transit. What is the district numerator?
Answer: households
Four equal blocks each contain 100 households. Both western blocks have 20 underserved; both eastern blocks have 100-20 underserved. Enter the western and eastern district rates in percent.
| percent | |
|---|---|
| Western district | |
| Eastern district |
A fictional language map redraws districts, then reports a rise from 52% to 70% in one district. No comparable household follow-up is supplied. Which inference is best?
Lesson test: one question per skill, one attempt each, no hints. Your answers are checked when you submit.
An invented area contains four blocks, each with 100 households. Blocks A and B each have 11 households without reliable transit; C and D each have 42. Zoning X groups A+B and C+D. Zoning Y groups A+C and B+D. Calculate the percentage without reliable transit for X's two districts and for Y's A+C district. The households do not move.
| percent of households | |
|---|---|
| X district A+B | |
| X district C+D | |
| Y district A+C |
Explain how the same four blocks can produce unequal district rates under one grouping and equal rates under another.
16. Two equal-sized blocks have 70 and 30 underserved households out of 100 each. Complete the combined rate., step 3
100/200*100 = 50%.
Combine counts before dividing.