PracticeVideosBlogFlashcardsEspañol
All Practice Exams

100+ Free NZ Scholarship Statistics Practice Questions

Prepare for the New Zealand Scholarship Statistics Assessment exam with instant access — no signup required.

✓ No registration✓ No credit card✓ No hidden fees✓ Start practicing immediately
About 3% of the Level 3 Statistics cohort is awarded Scholarship Pass Rate
100+ Questions
100% Free

Loading practice questions...

Same family resources

Explore More New Zealand NCEA & University Entrance

Continue into nearby exams from the same family. Each card keeps practice questions, study guides, flashcards, videos, and articles in one place.

NCEA Level 1 ChemBio — New Zealand

Practice Questions

NCEA Level 1 Agricultural and Horticultural Science — New Zealand

Practice Questions

NCEA Level 1 Chinese — New Zealand

Practice Questions

NCEA Level 1 Commerce — New Zealand

Practice Questions

NCEA Level 1 Dance — New Zealand

Practice Questions

NCEA Level 1 Design and Visual Communication — New Zealand

Practice Questions
2026 Statistics

Key Facts: NZ Scholarship Statistics Exam

Prepare for NZ Scholarship Statistics (Standard 93201) with 100 practice questions covering resampling inference, experimental design, time series modeling, probability algebra, and critical evaluation of statistical claims.

Sample NZ Scholarship Statistics Practice Questions

Try these sample questions to test your NZ Scholarship Statistics exam readiness. Each question includes a detailed explanation. Start the interactive quiz above for the full 100+ question experience with AI tutoring.

1In a bootstrap confidence interval investigation for a population median, how is a single bootstrap sample generated from an original representative sample of size n?
A.By randomly sampling n observations from the original sample with replacement
B.By randomly sampling n observations from the original sample without replacement
C.By taking repeated random samples of size n directly from the target population
D.By selecting n observations from the normal distribution centered at the sample median
Explanation: Bootstrapping mimics the original population sampling process by resampling with replacement from the observed sample dataset. Drawing n items with replacement allows individual values to appear zero, one, or multiple times, constructing a resample of identical size n that reflects sample variability.
2A 95% bootstrap confidence interval for the difference between the median hourly earnings of male and female graduates (Male - Female) is calculated as ($4.50, $8.20). Which is the most valid statistical interpretation?
A.We can be 95% confident that the true population median hourly wage for male graduates is between $4.50 and $8.20 higher than for female graduates
B.There is a 95% probability that any randomly selected male graduate earns between $4.50 and $8.20 more per hour than a female graduate
C.In 95% of individual bootstrap resamples, male median hourly earnings exceeded female median hourly earnings by $4.50 to $8.20
D.Because the interval does not contain zero, we can conclude that 95% of male graduates earn higher hourly wages than all female graduates
Explanation: A bootstrap confidence interval estimates the plausible range for the unknown population parameter (here, the difference in population medians). Because the entire interval ($4.50 to $8.20) is positive, we are 95% confident that male graduates have a higher population median wage by at least $4.50 and up to $8.20.
3If a statistician quadruples the sample size (n) from 100 to 400 while maintaining identical sample distribution characteristics, what is the expected impact on the width of the 95% bootstrap confidence interval?
A.The confidence interval width will be approximately halved
B.The confidence interval width will be reduced by a factor of four
C.The confidence interval width will double due to increased data range
D.The confidence interval width will remain unchanged because bootstrap proportions stay fixed
Explanation: The standard error of sample statistics (and thus the width of confidence intervals) varies inversely with the square root of the sample size (1/√n). Quadrupling n reduces standard error and interval width by a factor of √4 = 2, making it roughly half as wide.
4When constructing a 95% bootstrap confidence interval for a highly skewed population using the percentile method, why is the percentile method preferred over a symmetric formula (Sample Mean ± 1.96 × SE)?
A.The bootstrap percentile method naturally reflects skewness by allowing asymmetric distance from the point estimate to the lower and upper bounds
B.The symmetric formula is mathematically invalid for sample sizes under 10,000 regardless of skewness
C.The percentile method forces the bootstrap distribution to transform into a standard normal curve
D.Symmetric intervals always overestimate the population median by exactly two standard errors
Explanation: The bootstrap percentile method takes the 2.5th and 97.5th percentiles directly from the empirical bootstrap distribution. When data is skewed, the bootstrap distribution is also skewed, so the percentile interval naturally places asymmetric bounds around the point estimate, capturing skewness far better than symmetric Gaussian approximations.
5A researcher investigates regional income disparities in New Zealand by collecting a stratified random sample across urban and rural regions. When generating bootstrap resamples, why must the resampling procedure be stratified within each region?
A.To preserve the original sampling design and fixed stratum sample sizes in every bootstrap resample
B.To ensure that rural incomes are artificially weighted to match urban income medians
C.To eliminate sampling variability between different bootstrap iterations entirely
D.Because simple unstratified resampling automatically converts continuous data into categorical strata
Explanation: The golden rule of bootstrapping is that the resampling scheme must mirror the data collection structure. If the original study fixed the sample sizes within urban and rural strata, each bootstrap iteration must resample with replacement within those individual strata to maintain the design properties.
6A 95% bootstrap confidence interval for the difference in mean test scores between two teaching methods (Method A - Method B) is (-3.2, +4.8). What is the appropriate conclusion regarding the population means?
A.There is insufficient evidence to claim a difference between the population mean scores of Method A and Method B because zero is included in the interval
B.Method B is conclusively proven to produce superior average performance compared to Method A
C.Method A is guaranteed to perform better than Method B in at least 50% of secondary classrooms
D.The sample size was invalid because negative test score differences are mathematically impossible
Explanation: When a confidence interval for a difference between two parameters spans across zero (containing both negative and positive values), a difference of zero is a plausible value. Thus, we cannot make a definitive claim that one population mean is higher than the other at the 95% confidence level.
7A 95% bootstrap confidence interval for the ratio of median house prices (Auckland / Wellington) is calculated as (1.18, 1.42). How should this interval be interpreted in context?
A.We are 95% confident that the population median house price in Auckland is between 18% and 42% higher than in Wellington
B.We are 95% confident that Auckland house prices increase by $1.18 to $1.42 for every dollar spent in Wellington
C.95% of houses in Auckland are priced between 1.18 and 1.42 times higher than the Wellington median house price
D.The ratio proves that Wellington median house prices are between 18% and 42% higher than Auckland prices
Explanation: A ratio of medians of 1.18 corresponds to a 18% increase (1.18 - 1 = 0.18), while 1.42 corresponds to a 42% increase. Therefore, we are 95% confident that Auckland's population median house price is between 18% and 42% greater than Wellington's population median.
8An online poll regarding local transport funding collects responses from 5,000 self-selected website visitors. A student computes a 95% bootstrap confidence interval from this dataset. Why is this confidence interval misleading?
A.Bootstrap resampling reflects only sampling variability, not selection bias from voluntary response sampling
B.Bootstrap confidence intervals cannot be calculated for sample sizes exceeding 1,000 observations
C.Voluntary response samples eliminate random error, rendering confidence intervals completely zero in width
D.Online poll data must be analyzed using randomization tests rather than bootstrap resamples
Explanation: Bootstrapping quantifies precision (sampling variability) based on the supplied data. It assumes the sample is representative of the target population. If the original sample is biased due to self-selection, bootstrapping produces a very narrow interval centered on a biased value.
9Suppose an original sample of 50 student test scores contains a severe recording error where one student's score of 75 was recorded as 750. What effect will this error have on a bootstrap distribution for the sample mean?
A.The bootstrap distribution will be heavily shifted to the right and exhibit extreme bimodal spikes depending on how many times the outlier is resampled
B.The bootstrap resampling process will automatically detect and eliminate the score of 750 as an outlier
C.The bootstrap distribution will remain unchanged because bootstrap medians filter out arithmetic errors
D.The confidence interval width will shrink to zero because the score of 750 dominates all resamples
Explanation: Because resampling draws from the existing sample with replacement, the extreme value (750) will appear 0, 1, 2, or more times in various resamples. Each inclusion heavily inflates the resample mean, pulling the bootstrap distribution upward and creating discrete multi-modal clusters.
10In a paired study assessing the impact of a memory training course, 40 participants take a memory test before and after the course. How should bootstrap resampling be performed to construct a confidence interval for the mean improvement?
A.Resample the pairs (Before, After) together for each participant, compute individual differences, and calculate the mean difference for each resample
B.Resample all 40 Before scores independently of all 40 After scores and subtract the resampled means
C.Combine all 80 scores into a single pool, draw two random samples of size 40 with replacement, and calculate differences
D.Resample only the 40 After scores while keeping the 40 Before scores completely fixed at their original values
Explanation: Paired data exhibits within-subject dependence. To maintain the paired structure, the unit of resampling must be the participant (or pair of measurements). Resampling intact pairs preserves the correlation between pre- and post-test scores.

About the NZ Scholarship Statistics Exam

NZ Scholarship Statistics (Standard 93201) evaluates elite Year 13 students on their capacity for complex statistical synthesis, abstract reasoning, and critical evaluation. Candidates integrate knowledge from NCEA Level 3 Statistics standards—including formal inference via bootstrapping, experimental randomization tests, additive time series decomposition, bivariate regression diagnostics, probability combinations, and report evaluation. This 100-question practice set provides rigorous preparation with comprehensive explanations and option feedback.

Assessment

One 3-hour end-of-year examination on printed paper for Performance Standard 93201, with four extended questions set in real-life contexts across statistical investigation, statistical literacy and probability, emphasising data analysis and interpretation. A formulae and tables booklet and separate answer booklet are provided and an approved calculator is required. The multiple-choice questions in this bank are an English-language study adaptation, not a simulation of the official paper.

Time Limit

3 hours (180 minutes)

Passing Score

Cut score set annually by NZQA after marking; Scholarship and Outstanding Scholarship are awarded on rank order

Exam Fee

No charge for domestic candidates; NZ$102.20 per Scholarship subject for international fee-paying students (NZQA fee schedule, 1 January 2026) (New Zealand Qualifications Authority (NZQA))

NZ Scholarship Statistics Exam Content Outline

25%

Domain 1: Statistical Inference & Experimental Design

Bootstrap confidence interval construction and interpretation, re-randomization tests, random allocation versus random sampling, and experimental control.

25%

Domain 2: Time Series & Bivariate Data Modeling

Additive models ($Y = T + S + I$), seasonal adjustments, non-linear regression models, residual diagnostic plots, and influential points.

25%

Domain 3: Probability Theory & Risk Evaluation

Conditional probability, Bayes' theorem, event independence, expected value and variance algebra for linear combinations of random variables.

15%

Domain 4: Probability Distributions & Statistical Modeling

Normal, Binomial, Poisson, Triangular, and Rectangular distribution selection, parameter estimation, assumption verification, and contextual application.

10%

Domain 5: Critical Evaluation of Statistical Reports & Media Claims

Margin of error evaluation ($1/\sqrt{n}$), non-sampling errors, survey design flaws, graphical integrity, and causality critique.

How to Pass the NZ Scholarship Statistics Exam

What You Need to Know

  • Passing score: Cut score set annually by NZQA after marking; Scholarship and Outstanding Scholarship are awarded on rank order
  • Assessment: One 3-hour end-of-year examination on printed paper for Performance Standard 93201, with four extended questions set in real-life contexts across statistical investigation, statistical literacy and probability, emphasising data analysis and interpretation. A formulae and tables booklet and separate answer booklet are provided and an approved calculator is required. The multiple-choice questions in this bank are an English-language study adaptation, not a simulation of the official paper.
  • Time limit: 3 hours (180 minutes)
  • Exam fee: No charge for domestic candidates; NZ$102.20 per Scholarship subject for international fee-paying students (NZQA fee schedule, 1 January 2026)

Keys to Passing

  • Complete 500+ practice questions
  • Score 80%+ consistently before scheduling
  • Focus on highest-weighted sections
  • Use our AI tutor for tough concepts

NZ Scholarship Statistics Study Tips from Top Performers

1Always justify statistical choices by referencing contextual details, model assumptions, and sample limitations.
2Distinguish clearly between population inference (requires random sampling) and causal inference (requires random allocation).
3When evaluating time series, examine residual plots and seasonal indices before making extrapolations.
4Master variance algebra for independent variables: Var(aX - bY) = a^2 Var(X) + b^2 Var(Y) (note the positive sign for variances).
5In critical report evaluation, scrutinize non-sampling errors (e.g. non-response, question wording) which are often far more damaging than sampling error.

Frequently Asked Questions

What is NZ Scholarship Statistics (Standard 93201)?

NZ Scholarship Statistics is a prestigious national assessment for top secondary students in New Zealand, testing high-level statistical synthesis, critical thinking, and communication beyond standard NCEA Level 3.

What topics are covered in the Scholarship Statistics examination?

The examination synthesizes concepts from across the Level 8 NZ Curriculum, including bootstrap inference, experimental re-randomization, time series decomposition, bivariate non-linear modeling, probability rules, expected values, probability distributions, and report evaluation.

How is the Scholarship exam structured?

The external exam is a 3-hour session featuring extended multi-part scenario questions where students analyze real or realistic statistical investigations and write comprehensive explanations.

What is the passing score or award criteria for Scholarship Statistics?

Scholarship is awarded to roughly the top 3% of candidates sitting Level 3 Statistics nationally. Papers are marked out of 40 (typically 5 questions worth 8 marks each), with cut scores generally around 22-24 marks for Scholarship and 30+ for Outstanding Scholarship.

How does this 100-question multiple-choice set help with Scholarship preparation?

While the official NZQA exam uses written response booklets, this 100-question practice set tests the precise theoretical nuances, mathematical concepts, model assumptions, and diagnostic skills required to achieve top marks on the written exam.