Vertex Learn

2026 NCEA AS 91946 exam answers

Question ONE

Question ONE

a
Achieved

Using the travel-to-work line graph, estimate how many of the 3 246 047 working people worked from home in 2023.

Answer

About 17% of 3 246 047, which is roughly 550 000 people (accept about 535 000 to 570 000).

Step by step
  1. The graph shows percentages of people who answered the travel-to-work question, so use the number of working people, not the total census count.
  2. Read the work-from-home line at 2023: it is at about 17%.
  3. Multiply the number of working people by 0.17.
Full working
  1. Work from home in 2023 ≈ 17%read from graph
  2. 0.17 × 3 246 047 = 551 828percentage of workers
  3. ≈ 550 000 people work from home
b
Merit

A headline says driving to work has dropped a lot and working from home is rising. Agree or disagree, using values from the line graph.

Answer

Partly agree. Working from home has clearly risen (about 11.7% in 2018 to about 17% in 2023, its highest ever). Driving has fallen, but only from about 69% to about 66.5%, so calling that drop significant is an overstatement: driving is still by far the main way people get to work.

Step by step
  1. Look at both lines between 2018 and 2023, the latest two censuses.
  2. Work from home jumped by about 5 percentage points, an increase of nearly half of its 2018 value.
  3. Driving fell by only about 2.5 percentage points, and it is still about four times more common than working from home.
  4. Over the long term, driving rose from about 45% in 1971 to a peak of about 69% in 2006, then stayed flat until 2018 before dipping.
  5. So the second half of the headline is supported, but the word significantly for driving is too strong.
  6. A likely reason for the rise in working from home is the change in work habits after COVID-19 lockdowns.
Full working
  1. Work from home: 2018 ≈ 11.7%, 2023 ≈ 17%read from graph
  2. Increase ≈ 5.3 percentage points, about 45% more than in 2018
  3. Drove: 2018 ≈ 69%, 2023 ≈ 66.5%read from graph
  4. Decrease ≈ 2.5 percentage points, about 4% of its 2018 value
  5. Driving is still about 66.5% ÷ 17% ≈ 4 times as common as working from home
  6. Conclusion: work from home is clearly rising; driving has only decreased slightly, not significantly
c
Achieved

Give and justify one reason to be careful drawing conclusions about travel to work from the line graph.

Answer

The census records how people travelled on one single day. That day may not be typical, for example because of weather, a public holiday, school holidays or (in 2023) recent changes after COVID-19, so the graph may not show how people usually travel to work.

Step by step
  1. The census asks how you travelled on census day only.
  2. Bad weather would push people from walking or cycling into cars, and a strike or disruption could change public transport use.
  3. So one day's answers may not represent normal travel patterns.
  4. Other acceptable reasons: only the main method is recorded (a trip that uses a car and a train counts once); people who did not answer or did not work that day are not included; the gap between 2006 and 2013 is 7 years, not 5, so the time axis is uneven and can make trends look steeper or flatter than they are.
Full working
  1. Reason: data comes from one day every five yearsfrom the resource description
  2. Justification: that day may not be typical (weather, holidays, disruptions)
  3. Therefore conclusions about usual travel to work should be made with caution
d
Excellence

Using both travel graphs (work, and school or university), describe two similarities or differences and explain what each shows about travel patterns.

Answer

Similarity: staying home has grown in both. Work from home rose from about 11.7% (2018) to about 17% (2023), and study at home rose from about 5.4% to about 7.6%. Difference: workers mostly drive themselves (about 66.5% in 2023, with only about 4% as car passengers), while students are mostly car passengers (about 48% in 2023, up from about 39%), with only about 9.6% driving.

Step by step
  1. Similarity 1: both groups are more likely to stay at home in 2023 than in 2018.
  2. This suggests that after COVID-19 lockdowns, working and studying online became more normal across the population.
  3. Difference 2: the main way adults get to work is driving themselves, but the main way students get to school or university is being driven by someone else.
  4. This makes sense because most school students are too young to drive or do not own a car, so parents drop them off, often on their way to work.
  5. Another valid difference: students walk far more (about 16.5% in 2023) than workers use active travel (about 6%), because schools are usually closer to home than workplaces.
  6. Another valid similarity: public transport and active travel fell in both groups between 2018 and 2023 (work public transport about 6.6% to 5.2%; school walking about 20.5% to 16.5%), showing a general move towards cars and staying home.
Full working
  1. Work from home: 11.7% → 17%Resource 1A
  2. Study at home: 5.4% → 7.6%Resource 1B
  3. Similarity: both increased by about 2 to 5 percentage points
  4. Drove to work 2023 ≈ 66.5%, car passenger to work ≈ 4%Resource 1A
  5. Driving to school 2023 ≈ 9.6%, passenger to school ≈ 48% (2018 ≈ 39%)Resource 1B
  6. Difference: workers mainly drive themselves, students mainly ride as passengers
Question TWO

Question TWO

a
Achieved

From the licence two-way table, find the probability that a randomly chosen licence holder has a full licence.

Answer

64160 = 0.4 (40%)

Step by step
  1. Use the total row and column: 64 full licence holders out of 160 in the sample.
  2. Divide and simplify.
Full working
  1. P(full) = 64160full licence total over grand total
  2. = 25 = 0.4
  3. P(full) = 0.4
b
Merit

Is a rural licence holder more likely to have a learner licence than a city one? Decide using the table, and comment on the reliability of the sample.

Answer

True for this sample: P(learner | rural) = 2896 ≈ 0.29, which is higher than P(learner | city) = 1664 = 0.25. But the difference is small (about 4 percentage points) and the sample has only 160 people (64 in cities), so the claim is only weakly supported and may not hold for all of New Zealand.

Step by step
  1. Compare proportions, not counts: there are more rural people in the sample (96 versus 64), so 28 being bigger than 16 does not prove anything on its own.
  2. Find the fraction of city holders with a learner licence and the fraction of rural holders with one.
  3. The rural proportion is higher, so the claim is true for this sample.
  4. Reliability: the sample is random, which is good, but 160 is quite small and the gap is only about 4 percentage points, so another sample could easily show a different result.
  5. The sample only includes people who already hold a licence aged 16 to 26, so it says nothing about people with no licence at all.
  6. It is one year (2024) of data from one source.
Full working
  1. P(learnercity) = 1664 = 0.25 | 25%
  2. P(learnerrural) = 2896 ≈ 0.292 | about 29%
  3. 0.292 > 0.25, so rural is more likely in this sample
  4. Difference ≈ 0.04, small for a sample of 160
  5. Claim true for the sample, but the evidence is weak
c
Achieved

Using the licence-rate map, name two places where under 25% of under-26s have a full licence.

Answer

Gisborne and Whangārei (Northland). The Hawke's Bay area around Napier is also shaded in the lowest band.

Step by step
  1. Find the darkest shading in the key: less than 25%.
  2. The whole of Northland around Whangārei and the whole East Coast around Gisborne are in this darkest shade.
  3. All of the South Island is in the two lighter bands, so no South Island place qualifies.
Full working
  1. Darkest colour = less than 25%map key
  2. Whangārei (Northland): darkest shade
  3. Gisborne: darkest shade
  4. Answer: Whangārei and Gisborne
d
Excellence

Using the table and the map, discuss one similarity or difference between the North and South Islands in full-licence rates for under-26s, give reasons, and link to the city and rural probabilities.

Answer

Difference: the South Island has higher full-licence rates than the North Island. Almost all of the South Island is above 34% (only the West Coast and Nelson areas are 25% to 34%), while large parts of the North Island (Northland, Gisborne and the East Coast, Hawke's Bay, parts of the central North Island and the Wellington region) are below 25%. The table shows a full licence is more likely in cities (2964 ≈ 0.45) than in rural areas (3596 ≈ 0.36). This fits the North Island, where the low areas are mostly rural and the main city areas are lighter. But the South Island is mostly rural and still has high rates, so being rural is not the only factor.

Step by step
  1. Read the map: the South Island is mostly the lightest shade, the North Island has many areas in the darkest shade.
  2. From the table, P(full | city) = 2964 ≈ 0.45 and P(full | rural) = 3596 ≈ 0.36, so city licence holders are more likely to have a full licence.
  3. In the North Island the darkest areas are mostly rural (Northland, the East Coast), which agrees with the table.
  4. In the South Island, rural areas like Canterbury and Otago still have rates above 34%, which does not match the city versus rural pattern.
  5. Possible reasons: the cost of lessons and licence tests (several hundred dollars) is harder for families in lower-income areas such as Northland and Gisborne; testing centres can be far away in remote North Island areas; access to a car and a supervising full-licence driver for practice varies.
  6. Some young people in rural areas may also drive without progressing past a learner or restricted licence.
  7. Caution: the table is a national sample of licence holders only, while the map shows a percentage of all under-26s (including those with no licence), so the two resources do not measure exactly the same thing.
Full working
  1. P(fullcity) = 2964 ≈ 0.453 | Resource 2A
  2. P(fullrural) = 3596 ≈ 0.365 | Resource 2A
  3. South Island: mostly > 34%, none < 25%map
  4. North Island: many regions < 25% (Northland, Gisborne, Hawke's Bay)map
  5. Conclusion: South Island under-26s are more likely to hold a full licence; the city versus rural pattern explains the North Island but not the rural South Island, so cost and access are likely factors too
Question THREE

Question THREE

a
Achieved

Auckland: 1 700 000 (2018) to 2 300 000 (2048). Christchurch: 360 000 to 460 000. Which has the larger percentage increase? Justify.

Answer

Auckland: about 35.3% increase, compared with about 27.8% for Christchurch. Auckland has the larger percentage increase.

Step by step
  1. Percentage increase = increase ÷ starting value × 100.
  2. Auckland grows by 600 000 from 1 700 000.
  3. Christchurch grows by 100 000 from 360 000.
  4. Compare the two percentages, not the raw increases.
Full working
  1. Auckland: 2 300 000 − 1 700 0001 700 000 × 100 = 600 0001 700 000 × 100
  2. ≈ 35.3%
  3. Christchurch: 460 000 − 360 000360 000 × 100 = 100 000360 000 × 100
  4. ≈ 27.8%
  5. 35.3% > 27.8%, so Auckland has the larger percentage increase
b
Merit

Using the two pie charts, describe and explain two similarities or differences in how people travel to work in Auckland and Christchurch.

Answer

Difference 1: car travel is far more common in Christchurch (80.7%) than in Auckland (55.6%). Difference 2: public transport is used much more in Auckland (24.3%) than in Christchurch (6.8%). (A similarity: cycling is small and almost the same in both, 1.9% and 2.1%.)

Step by step
  1. Difference 1: in both cities the car is the most common way to get to work, but in Christchurch about 4 in 5 people use a car compared with just over half in Auckland.
  2. This suggests Christchurch is easier to drive around (a smaller, flatter city with less congestion) so people choose the car.
  3. Difference 2: about a quarter of Auckland workers use public transport, which is about 3.6 times the Christchurch rate.
  4. This shows Auckland has a larger bus, train and ferry network, and heavy traffic and expensive parking push people onto public transport.
  5. Auckland also has more walking (17.2% compared with 10%), probably because more people live and work close together in the central city.
  6. Similarity: cycling is about 2% in both cities, so very few people choose to cycle to work in either city.
Full working
  1. Car: Auckland 55.6%, Christchurch 80.7%difference 25.1 percentage points
  2. Public transport: Auckland 24.3%, Christchurch 6.8%24.3 ÷ 6.8 ≈ 3.6 times
  3. Walking: Auckland 17.2%, Christchurch 10%
  4. Cycling: Auckland 1.9%, Christchurch 2.1%similar
  5. Christchurch relies much more on cars; Auckland makes much more use of public transport
c
Merit

Using the box plots of car travel times, discuss whether North Island workers tend to take longer to get to work than South Island workers.

Answer

Agree, with some caution. The North Island median (48 min) is 10.5 minutes higher than the South Island median (37.5 min), and the whole North Island box (31.5 to 69.5 min) is shifted to the right of the South Island box (18 to 55 min). The spreads are similar (IQR 38 and 37 min). DBM ÷ OVS = 10.5 ÷ 51.5 ≈ 0.20, which only just reaches the one-fifth guide, so the evidence that North Island travel times tend to be longer is borderline. With the stricter one-third guide often used for samples of 30 to 40, you would say instead that the evidence is not strong enough; either call is reasonable if it is justified.

Step by step
  1. Centre: the North Island median is 48 minutes and the South Island median is 37.5 minutes, so the typical North Island trip is longer.
  2. Shift: the North Island lower quartile, upper quartile, minimum and maximum are all higher than the South Island values.
  3. Spread: the IQRs are almost the same, so the two groups vary by a similar amount.
  4. Overlap: the boxes overlap a lot, and each median lies inside the other box.
  5. Distance between medians is 10.5 min and the overall visible spread (from the South Island LQ to the North Island UQ) is 51.5 min.
  6. 10.5 is just over one fifth of 51.5, so we can tentatively make the claim, but a different sample of 40 could change it.
  7. Note the survey only includes people who drive to work.
Full working
  1. Medians: NI 48, SI 37.5DBM = 48 − 37.5 = 10.5
  2. Quartiles: NI 31.5 to 69.5, SI 18 to 55
  3. IQR: NI = 69.5 − 31.5 = 38, SI = 55 − 18 = 37
  4. OVS = 69.5 − 18 = 51.5
  5. DBMOVS = 10.551.5 ≈ 0.204 > 0.2
  6. Agree (tentatively): North Island car commuters tend to take longer to get to work
d
Achieved

Discuss one factor that could change the survey results and how it would affect travel times in Auckland or Christchurch.

Answer

The day and time the trip was made. If people answered about trips in school holidays or outside rush hour, Auckland times would come out shorter than usual because traffic is much lighter; in normal peak-hour traffic Auckland commutes are much longer. Christchurch, with less congestion, would change less.

Step by step
  1. Travel time by car depends heavily on traffic.
  2. Auckland is known for heavy peak-hour congestion, so the time of the trip matters a lot there.
  3. School holidays, public holidays or working flexible hours all reduce traffic and shorten times.
  4. Other acceptable factors: roadworks or a crash on a main motorway (longer times); weather such as heavy rain (longer times); where in the city the person lives (outer suburbs mean longer trips); people guessing or rounding their average time in an online survey.
Full working
  1. Factor: time of day or time of year of the commute
  2. Peak traffic → longer Auckland times; holidays or off-peak → shorter
  3. Christchurch has less congestion, so its times would change less
e
Excellence

Given median travel times of 38 min (Auckland) and 20 min (Christchurch), compare these with the island box plots, link to the features of the graph, and say why you may or may not trust the survey.

Answer

Christchurch fits the survey well: its 20 min median sits just above the South Island LQ (18 min), and the 12 Christchurch points in the survey mostly lie between about 4 and 23 min (median about 18 min), on the left of the South Island distribution. Auckland's 38 min is below the North Island median of 48 min, but the 19 Auckland points in the survey cluster around 41 to 48 min (median about 47 min), so the survey's Auckland car drivers report longer times than 38 min. Both cities are below their island medians, so the long upper whiskers (NI up to 84 min, SI up to 64 min) come mostly from people outside the two cities. The survey is partly trustworthy: participants were randomly chosen from registrations and the city shares look representative, but the samples are small, it is an online self-reported estimate, and it covers car drivers only.

Step by step
  1. South Island: Christchurch's median of 20 min is close to the SI lower quartile of 18 min, so a typical Christchurch commute is shorter than about three quarters of the South Island sample.
  2. In the dot plot almost every Christchurch point is below the SI median of 37.5 min; the longer SI times (40 to 64 min) are from other South Island places, probably rural people driving long distances.
  3. North Island: Auckland's 38 min median sits between the NI lower quartile (31.5) and median (48), so on the published figure Auckland is quicker than the North Island sample overall.
  4. However, the Auckland points in the survey are mostly between 40 and 48 min, higher than 38. One reason is that the 38 min figure covers all methods, including walking (17.2%), while the survey only asked car drivers.
  5. The very long and very short NI times (14 to 37 min and 68 to 84 min) come mostly from other North Island places, so the NI box is very wide.
  6. Reasons to trust: participants were randomly selected from NZTA car registrations; there are 40 in each island; Auckland makes up 19 of 40 NI people (about 48%) and Christchurch 12 of 40 SI people (30%), roughly in line with their share of each island's population.
  7. Reasons not to trust: 40 is a small sample per island; the survey was online, so only people who chose to respond are included (non-response bias); people estimated their own average time, which leads to guessing and rounding; car registration addresses may not be where the driver lives; and we do not know when the survey was done or whether the 38 and 20 minute figures come from the same year.
Full working
  1. Christchurch median 20 min vs SI: LQ 18, median 37.5, mean 36.12520 is about the SI lower quartile
  2. Christchurch survey points: 12 values, about 4 to 42 min, median ≈ 18 minread from dot plot
  3. Auckland median 38 min vs NI: LQ 31.5, median 48, mean 49.72538 is between LQ and median
  4. Auckland survey points: 19 values, about 25 to 73 min, median ≈ 47 minread from dot plot
  5. Auckland share of NI sample = 1940 ≈ 48%; Christchurch share of SI sample = 1240 = 30%
  6. Conclusion: Christchurch survey data agree with the 20 min median; Auckland survey car times are longer than 38 min; results are only partly reliable because of the small, online, self-reported, car-only sample

AS 91946 past papers · All NCEA Level 1 Maths

Worked by Vertex Academy teachers. NZQA's marking schedule for 2026 is not out yet; these are our answers, and the grade on each part is our estimate.