“150”, Aptitude Test Questions and Answers for Statistician II -National Insurance Corporation (NIC).
ABSTRACT
This preparation package provides 150 multiple-choice aptitude test questions and
answers for Statistician II – National Insurance Corporation (NIC), Tanzania.
The questions are designed to assess candidates’ understanding of statistics,
data analysis, probability, sampling, statistical reporting, insurance data,
customer-flow analysis, performance measurement, data quality, and
evidence-based decision-making. Particular emphasis is placed on realistic
NIC-related scenarios, closely related answer choices, analytical reasoning,
interpretation of statistical information, and practical application rather
than simple memorization. Each question is accompanied by a clear answer and
detailed rationale to help candidates understand the underlying statistical
concepts and strengthen their readiness for the Public Service Online Aptitude Test.
Prepared
by: Statistician II
An
author based in Dar-es-salaam.
0628729934.
Date:
August 17, 2026
Dear
applicants,
This
collection of questions and answers has been prepared to help all of you to understand
the key areas tested during the interview. The goal is to provide a useful, and
practical study guide so you can all perform confidently and fairly in the
selection process. I wish you the best of luck, and may this resource support
you in achieving success!
Warm
regards,
Statistician
II
For
Personal Use by Applicants Preparing for Statistician II National Insurance Corporation (NIC).
ALL
QUESTIONSARE COMPILED TOGETHER.
1. An NIC branch
recorded the following numbers of customers served over five consecutive
months: 1,200, 1,350, 1,500, 1,650 and 1,800. Management states that customer
flow increased by 10% per month throughout the period. Which assessment is
statistically most appropriate?
A.
The statement is correct because each month gained 150 customers
B. The statement is correct because the average monthly increase was 10%
C. The statement is inaccurate because the percentage increase was not constant
D. The statement is inaccurate because customer flow should be measured
annually
Answer:
C
Rationale: The number of
customers increased by 150 each month, but the percentage increase was not
constant because the same absolute increase represents different percentages of
each month's previous value. For example, the increase from 1,200 to 1,350 is
12.5%, whereas the increase from 1,650 to 1,800 is approximately 9.1%.
Therefore, the data demonstrate a constant absolute increase but not a
constant percentage increase. A Statistician should distinguish between these
two types of growth before describing a trend to management.
2. An insurer wants
to estimate the proportion of policyholders who are satisfied with claims-handling
services. The policyholders are distributed across motor, property, health and
other insurance classes, with substantially different numbers in each class.
Which sampling approach would best ensure that each class is appropriately
represented?
A.
Select policyholders randomly from the complete customer register
B. Select equal numbers of policyholders from every regional branch
C. Select customers who recently completed claims at selected branches
D. Divide policyholders by insurance class and randomly sample within each
class
Answer:
D
Rationale: Stratified random
sampling is most appropriate because the population naturally consists of
distinct insurance classes and their representation may differ substantially.
Dividing the population into strata such as motor, property, health and other
classes before randomly selecting respondents helps ensure that each important
group is represented. A simple random sample could underrepresent smaller
classes, while convenience-based approaches introduce selection bias. The
objective is not merely to obtain a large sample but to obtain one that
adequately reflects the structure of the population.
3. An NIC
statistical report shows that the mean monthly claim amount increased from TZS
1.8 million to TZS 2.4 million, while the median increased from TZS 1.2 million
to TZS 1.3 million. Which interpretation is most defensible?
A.
Large claims probably increased their influence on the average
B. Typical claim amounts probably increased by exactly TZS 600,000
C. Claim amounts probably became more evenly distributed overall
D. The median indicates that every claim increased substantially
Answer:
A
Rationale: The mean
increased by TZS 600,000 while the median increased by only TZS 100,000. Such a
pattern can occur when relatively large claims become more frequent or larger,
pulling the mean upward more strongly than the median. The mean is sensitive to
extreme observations, whereas the median is more resistant to them. Therefore,
the available information suggests that high-value claims may have had a
greater influence on the average, although further analysis would be required
to establish the exact cause.
4. An NIC analyst
calculates a correlation coefficient of −0.82 between customer waiting
time and customer satisfaction across several branches. Which conclusion is
most appropriate?
A.
Longer waiting times directly cause lower satisfaction in every branch
B. Branches with longer waiting times tend to have lower satisfaction
C. A reduction in waiting time will necessarily increase satisfaction
D. Customer satisfaction explains exactly 82% of waiting-time differences
Answer:
B
Rationale: A correlation
coefficient of −0.82 indicates a strong negative linear association between the
two variables: branches with higher waiting times tend, on average, to have
lower satisfaction scores. However, correlation alone does not establish
causation, so it would be inappropriate to conclude that reducing waiting time
will necessarily cause satisfaction to increase. Also, −0.82 is a correlation
coefficient, not a statement that 82% of variation is explained. The
appropriate interpretation therefore focuses on the observed association while
avoiding causal claims.
5. An NIC branch
reports 4,800 customers in January and 5,520 customers in February. Management
wants the percentage change in customer flow from January to February. Which
value is correct?
A.
12.0%
B. 13.0%
C. 15.0%
D. 17.0%
Answer:
C
Rationale: The increase in
customers is 5,520 − 4,800 = 720. The percentage increase is therefore 720÷4,800×100=15%.
The denominator must be the original January figure because percentage change
measures the change relative to the starting value. This distinction is
important in performance reporting because using the February figure as the
denominator would produce a different and inappropriate percentage for
describing the increase from January to February.
6. A dataset
containing monthly claim payments has one extremely large claim that is several
times larger than all other claims. If the Statistician wants a measure of
central tendency that is least affected by this unusually large observation,
which should be preferred?
A.
Arithmetic mean
B. Weighted mean
C. Geometric mean
D. Median
Answer:
D
Rationale: The median is
generally less affected by extreme observations than the mean because it
depends on the middle position of ordered observations rather than the
magnitude of every observation. A very large insurance claim can substantially
increase the arithmetic mean and potentially distort the apparent typical claim
amount. The median therefore provides a more robust description of the central
claim amount when the distribution is strongly skewed or contains influential
outliers.
7. An analyst
receives monthly customer-flow data from 20 NIC branches. Two branches report
exceptionally high customer numbers compared with all others. Before preparing
the performance report, what should the analyst do first?
A.
Remove the two observations to prevent distortion
B. Verify the observations against the original records
C. Replace the observations using the overall branch average
D. Report the values as errors because they differ substantially
Answer:
B
Rationale: An unusually
large observation should not automatically be deleted. It may represent a
genuine difference in branch activity, but it may also result from a recording,
coding or transmission error. The appropriate first step is therefore to verify
the observations against source records and investigate their plausibility.
Removing or replacing data without evidence can introduce analyst bias and
produce an inaccurate management report. A Statistician should distinguish
between a genuine outlier and a data-quality problem before deciding how to
treat it.
8. An NIC survey
receives responses from 900 customers out of 3,000 customers contacted. If the
900 respondents are systematically different from those who did not respond,
which problem poses the greatest threat to the validity of the satisfaction
results?
A.
Sampling variation
B. Measurement rounding
C. Non-response bias
D. Random classification error
Answer:
C
Rationale: Non-response bias
occurs when people who respond differ systematically from those who do not
respond in ways related to the subject being measured. For example, customers
with particularly positive or negative experiences may be more likely to
complete a survey than indifferent customers. A response rate of 30% alone does
not prove that the results are biased, but systematic differences between
respondents and non-respondents create a serious representativeness problem.
The Statistician should therefore assess non-response patterns rather than
relying only on the number of completed questionnaires.
9. An NIC
performance report shows that total premium income increased by 20%, while the
number of active policies increased by only 5%. Assuming the figures are
measured consistently, which interpretation is most reasonable?
A.
Average premium income per policy increased
B. Total customer numbers increased by exactly 15%
C. Claims expenditure must have declined by 15%
D. Policy cancellations necessarily increased during the period
Answer:
A
Rationale: If total premium
income rises by 20% while the number of active policies rises by only 5%,
premium income per active policy must have increased, assuming the comparison
is based on the same definitions and period. This does not necessarily mean
that every individual policy became more expensive; the change could reflect
differences in product mix, policy values, pricing or other factors. The other
alternatives introduce conclusions that cannot be established from the
information provided.
10. A Statistician
compares the average customer waiting time between two branches. Branch A has a
mean of 18 minutes and a standard deviation of 3 minutes, while Branch B has a
mean of 18 minutes and a standard deviation of 11 minutes. What does the
difference in standard deviation primarily indicate?
A.
Branch A serves more customers than Branch B
B. Branch B has a higher average waiting time
C. Waiting times vary more widely in Branch B
D. Branch B necessarily has lower customer satisfaction
Answer:
C
Rationale: Both branches
have the same mean waiting time of 18 minutes, but Branch B has a much larger
standard deviation. This indicates that waiting times in Branch B are more
dispersed around the mean, meaning customers experience a wider range of
waiting times. Standard deviation measures variability rather than the level of
the average. It does not by itself establish customer satisfaction, customer
volume or the exact cause of the greater variation.
11. An NIC analyst
wants to determine whether the average monthly customer flow at a branch has
changed from the historical benchmark of 2,000 customers. A sample of recent
monthly observations is available. Which statistical approach is most
appropriate for formally assessing whether the observed difference could
reasonably be due to sampling variation?
A.
Calculate only the percentage increase
B. Conduct a hypothesis test concerning the population mean
C. Compare the largest observation with the historical benchmark
D. Rank the monthly observations from smallest to largest
Answer:
B
Rationale: The question
concerns whether an observed sample mean differs significantly from a specified
historical population benchmark. A hypothesis test concerning the population
mean provides a formal framework for evaluating whether the observed difference
is sufficiently large relative to sampling variability. Merely calculating a
percentage increase describes the difference but does not assess statistical
evidence. Ranking observations or examining only the maximum similarly fails to
address the inferential question.
12. An NIC
statistical bulletin reports that the number of claims rose from 10,000 to
12,000, while the number of active policies increased from 400,000 to 600,000
during the same period. What happened to the claims frequency, measured as
claims divided by active policies?
A.
It increased from 2.5% to 3.0%
B. It increased from 2.0% to 2.5%
C. It decreased from 3.0% to 2.5%
D. It decreased from 2.5% to 2.0%
Answer:
D
Rationale: Initially, claims
frequency was 10,000÷400,000=2.5%. In the later period, it was 12,000÷600,000=2.0%.
Therefore, although the absolute number of claims increased by 2,000, claims
occurred at a lower rate relative to the number of active policies. This
illustrates why a Statistician should use rates or frequencies when comparing
populations of different sizes rather than relying solely on absolute counts.
13. An NIC manager
asks for a monthly report showing the number of customers, percentage change
from the previous month, average waiting time and satisfaction score for each
branch. What is the main statistical advantage of presenting these measures
together?
A.
It eliminates the need to validate branch data
B. It allows different dimensions of performance to be compared
C. It guarantees that all branches have equal operating conditions
D. It converts qualitative customer opinions into financial values
Answer:
B
Rationale: Combining
customer volume, growth, waiting time and satisfaction provides management with
a multidimensional view of branch performance. A branch may have high customer
numbers but poor waiting times, or low traffic but strong satisfaction.
Examining several indicators together helps management identify patterns that
would not be visible from a single measure. However, the approach does not
eliminate the need for data validation or guarantee that branches operate under
identical conditions.
14. An NIC survey
asks customers to rate service satisfaction on a scale from 1 to 5. The
Statistician wants to compare the distribution of responses across branches
without assuming that the numerical differences between categories are exactly
equal. Which summary would be most defensible?
A.
Median satisfaction score
B. Arithmetic mean satisfaction score
C. Variance of satisfaction scores
D. Standard deviation of satisfaction scores
Answer:
A
Rationale: A 1-to-5
satisfaction scale is commonly treated as ordinal when the analyst cannot
assume that the distance between adjacent categories is equal. The median is
appropriate for ordinal data because it identifies the central ordered category
without requiring equal intervals between categories. Although means and
standard deviations are sometimes reported for rating scales, doing so
implicitly treats the categories as having meaningful equal numerical
distances. For a cautious statistical interpretation, the median is therefore the
more defensible summary.
15. An NIC branch has the capacity to serve
50 customers per hour and actually serves an average of 40 customers per hour
during a particular period. What is the branch's average service-capacity
utilization?
D. 90% C. 80% B. 70% A.
60%
Answer: C
Rationale: Service-capacity utilization is
calculated as actual average service volume divided by available service
capacity, multiplied by 100. Thus, 40÷50×100=80%. The branch is therefore operating at 80% of its available
average service capacity during the specified period. This measure provides a
standardized indication of how much of the branch's available service capacity
is being utilized.
16. A dataset
contains monthly premium income for several years. The Statistician wants to
compare the general movement over time while reducing the influence of
short-term monthly fluctuations. Which technique is most appropriate?
A.
Moving average
B. Frequency polygon
C. Cross-sectional sampling
D. Simple randomization
Answer:
A
Rationale: A moving average
smooths short-term fluctuations by averaging observations over successive time
windows, making the underlying trend easier to identify. This is particularly
useful for monthly business data where seasonal or random fluctuations can
obscure longer-term movements. A frequency polygon is primarily a graphical
representation of a distribution, while sampling and randomization are methods
related to data selection rather than trend smoothing.
17. An NIC analyst
finds that the average monthly number of customer complaints fell from 800 to
600 after a new service procedure was introduced. However, during the same
period, the number of customers served fell from 20,000 to 10,000. Which
measure would provide a fairer comparison of complaint occurrence?
A.
Total complaints only
B. Complaint rate relative to customers served
C. Percentage reduction in staff numbers
D. Difference between monthly complaint totals
Answer:
B
Rationale: Comparing only
the number of complaints can be misleading when the size of the customer
population changes substantially. The appropriate approach is to calculate
complaints relative to the number of customers served. Before the change, the
complaint rate was 800/20,000 = 4%, while after the change it was 600/10,000 =
6%. Thus, although total complaints declined, complaints represented a larger
proportion of customers after the new procedure. This demonstrates the
importance of using standardized rates when the underlying population changes.
18. An NIC analyst obtains
a sample mean claim amount of TZS 2.5 million and constructs a 95% confidence
interval of TZS 2.2 million to TZS 2.8 million. Which interpretation is
statistically most appropriate?
A.
95% of individual claims fall between TZS 2.2 and 2.8 million
B. There is a 95% probability that every future claim falls in this range
C. The sample mean has a 95% probability of remaining exactly TZS 2.5 million
D. The interval provides a 95% confidence procedure for the population mean
Answer:
D
Rationale: A 95% confidence
interval provides a statistical procedure for estimating the population mean,
with the stated confidence level referring to the long-run performance of the
interval construction method. It does not mean that 95% of individual claims
must lie inside the interval, because individual observations may vary much
more widely. Nor does it mean that future claims will necessarily fall within
the interval. The correct interpretation therefore relates the interval to the
unknown population mean rather than individual claims.
19. An NIC manager
receives a report stating that Branch X generated 30% more premium income than
Branch Y. Branch X, however, has twice the number of active policies. Which
additional measure would be most useful for assessing the difference in
performance?
A.
Premium income per active policy
B. Total number of branch employees
C. Number of complaints without a denominator
D. Number of months each branch has operated
Answer:
A
Rationale: Because Branch X
has twice as many active policies, comparing total premium income alone does
not adequately account for the different sizes of the two branches. Premium
income per active policy standardizes the comparison and can indicate whether
Branch X generates more premium per policy rather than simply having more
policies. Other measures may provide useful contextual information, but they do
not directly address the effect of the difference in policy volume.
20. A survey of NIC
customers is conducted only among people who visit branches during working
hours. The Statistician concludes that the results represent all NIC customers.
What is the principal methodological concern?
A.
The sample may exclude customers using other service channels
B. The sample size is automatically too large for statistical analysis
C. Branch customers cannot provide reliable satisfaction information
D. Working-hour surveys always produce normally distributed responses
Answer:
A
Rationale: Customers who use
digital platforms, agents, telephone services or visit branches outside the
selected period may have different experiences from those included in the
survey. Restricting data collection to branch visitors during working hours can
therefore create coverage or selection bias. The problem is not necessarily the
sample size but whether the sampling frame adequately represents the target
population. A Statistician should identify which segments of the population are
systematically excluded before generalizing the findings.
21. An NIC report
shows the following annual claim frequencies: 2.0%, 2.2%, 2.4%, 2.3% and 2.5%.
Which statement best describes the pattern?
A.
Claim frequency increased continuously every year
B. Claim frequency decreased continuously after the first year
C. Claim frequency generally increased despite a temporary decline
D. Claim frequency remained constant throughout the five-year period
Answer:
C
Rationale: The sequence
rises from 2.0% to 2.2%, then to 2.4%, falls slightly to 2.3%, and rises again
to 2.5%. Therefore, the overall direction is upward, but the increase was not
continuous because there was a temporary decline between the third and fourth
years. A Statistician preparing a trend report should preserve this nuance
rather than describing the series as either perfectly increasing or completely
constant.
22. A Statistician is
asked to prepare a monthly performance report for management. Which approach
would provide the strongest basis for decision-making?
A.
Present raw figures without interpretation to avoid bias
B. Present only percentages because they are easier to compare
C. Present the highest-performing branches to emphasize success
D. Present validated indicators with trends, comparisons and relevant
interpretation
Answer:
D
Rationale: A management
report should transform validated data into useful information for
decision-making. Presenting indicators together with trends, meaningful
comparisons and appropriate interpretation allows managers to understand what
has changed, where performance differs and what issues may require attention.
Raw figures alone may obscure important patterns, while percentages alone may
hide differences in scale. Selecting only successful branches would also
introduce a serious reporting bias.
23. An NIC dataset
contains 10,000 customer records. A particular field, “customer age,” is
missing for 18% of the records. Before selecting an imputation method, what
should the Statistician investigate first?
A.
Whether the missing values occur systematically across groups
B. Whether the dataset can be reduced to 1,000 records
C. Whether the missing values should automatically be coded as zero
D. Whether the average age should replace every missing observation
Answer:
A
Rationale: Before deciding
how to handle missing data, the Statistician should examine the pattern and
mechanism of missingness. If missing ages occur disproportionately in
particular branches, customer groups or data-collection periods, simply
replacing them with an overall average could distort the analysis. Coding
missing age as zero is generally inappropriate because zero is not a meaningful
missing-value representation for this variable. Understanding why and where
data are missing is therefore an essential first step before selecting an
appropriate treatment.
24. An NIC analyst
wants to determine whether two categorical variables, such as branch region
and complaint category, are statistically associated. Which method is
most appropriate?
A.
Paired t-test
B. Chi-square test of independence
C. Pearson correlation of the category labels
D. One-sample test of a population mean
Answer:
B
Rationale: Both branch
region and complaint category are categorical variables, so a chi-square test
of independence is appropriate for assessing whether the distribution of
complaint categories differs systematically across regions. A t-test is
designed primarily for comparing means of quantitative variables, while Pearson
correlation requires appropriately measured numerical variables. Assigning
numerical codes to categories does not automatically make correlation an appropriate
method because the numerical labels do not necessarily represent meaningful
quantitative distances.
25. NIC records show
that the number of customers served increased substantially during a month, but
the customer satisfaction score also declined. A manager concludes that the
increase in customer traffic caused dissatisfaction. What should the
Statistician do before supporting this conclusion?
A.
Accept the conclusion because both indicators changed simultaneously
B. Reject the conclusion because customer traffic cannot affect satisfaction
C. Examine other relevant factors and the relationship between the variables
D. Replace the satisfaction figures with the previous month's average
Answer:
C
Rationale: Simultaneous
changes in two variables do not by themselves establish causation. Higher
customer traffic may contribute to longer waiting times, but satisfaction could
also be affected by staffing levels, service interruptions, claims delays,
system problems, seasonal factors or other variables. The Statistician should
therefore investigate the relationship between customer volume and satisfaction
while controlling for or examining relevant factors before making a causal
statement. Replacing the observed satisfaction data would improperly alter the
evidence rather than analyze it.
26. An NIC analyst
compares two branches using the following figures: Branch A processed 8,000
policies with 240 complaints, while Branch B processed 5,000 policies with 200
complaints. Which conclusion is statistically most appropriate?
A.
Branch A has a higher complaint rate than Branch B
B. Branch B has a higher complaint rate than Branch A
C. Both branches have the same complaint rate
D. Complaint rates cannot be compared using these figures
Answer:
B
Rationale: Complaint rates
must be calculated relative to the number of policies processed. Branch A has
240/8,000=3%, while Branch B has 200/5,000=4%. Therefore, Branch B has the
higher complaint rate despite having fewer total complaints. This illustrates
why absolute complaint counts can produce misleading comparisons when branches
differ in the volume of business handled.
27. An NIC claims department finds that 40%
of submitted claims are motor claims. Among motor claims, 8% are fraudulent,
while among non-motor claims, 3% are fraudulent. Which statement is most
appropriate?
A. Fraudulent claims are
equally common in both categories
B. Motor claims have a higher conditional fraud rate
C. Non-motor claims have a higher conditional fraud rate
D. Overall fraud cannot be assessed from conditional rates
Answer: B
Rationale: The fraud rate among motor claims is
8%, compared with 3% among non-motor claims. Therefore, conditional on the
claim being in a particular category, motor claims have the higher observed
fraud rate. The 40% motor-claim proportion would be needed for calculating the
overall probability that a randomly selected claim is fraudulent, but it does
not change the comparison between the two conditional fraud rates.
28. NIC receives
monthly data from branches in different formats. Some branches record dates as
DD/MM/YYYY, while others use MM/DD/YYYY. What is the most important statistical
data-management concern before combining the records?
A.
Increasing the number of observations
B. Converting all values into percentages
C. Standardizing the date format before analysis
D. Removing branches with different recording systems
Answer:
C
Rationale: Inconsistent date
formats can lead to incorrect interpretation of observations, particularly
where day and month values are both 12 or below and the software cannot
determine which convention was intended. Before combining datasets, the
Statistician should establish and apply a consistent date format and verify the
conversion. Increasing the sample size does not solve the problem, and deleting
branches would unnecessarily reduce the dataset and potentially introduce bias.
29. An NIC analyst
calculates an average monthly customer flow of 2,400 customers from a sample of
branches. A second analyst obtains an average of 2,650 after including several
very large urban branches. What is the most likely explanation?
A.
The second mean must be statistically incorrect
B. The sample composition can influence the calculated mean
C. Urban branches should always be excluded from analysis
D. A larger sample automatically produces a lower mean
Answer:
B
Rationale: The composition
of a sample directly affects descriptive statistics. If the second sample
contains several branches with substantially higher customer volumes, those
observations can raise the overall mean. A larger sample does not automatically
produce either a higher or lower mean; its effect depends on the values added.
The appropriate response is to examine the sampling design and characteristics
of the included branches rather than assuming that one calculation is
incorrect.
30. An NIC branch
records customer arrivals during six one-hour periods as follows: 35, 42, 51,
48, 39 and 45. Which measure would be most appropriate for describing the
typical hourly customer flow if no major outliers are present?
A.
Mean hourly customer flow
B. Maximum hourly customer flow
C. Range of hourly customer flow
D. Difference between consecutive observations
Answer:
A
Rationale: When observations
are quantitative, reasonably balanced and free from major outliers, the
arithmetic mean provides a useful measure of typical customer flow. The maximum
identifies only the busiest period, while the range measures dispersion rather
than central tendency. The differences between consecutive observations can describe
changes but do not summarize the typical level of customer arrivals.
📘 Get the Full Aptitude Test Questions PDF through your Gmail (Questions 1–150)
You’ve just accessed the first 30 questions. The full set of 150 expertly prepared aptitude test questions for Statistician II -National Insurance Corporation (NIC), Is available, pay, and get access.
To get access to the full PDF, please make a payment of Tsh 10,000 to the LIPA numbers below:
After payment, please send a text message to notify us of your payment:
⚠️ Important Notice
- The PDF will be watermarked with your name and phone number and protected for personal use only.
- Redistribution, sharing, screenshotting, or copying the contents is strictly prohibited. When you share unlawfully, your name and phone number are visible and easy to trace as you leaked a document to other third parties.
- Legal action may be taken against the misuse of this material.
Thank you for supporting quality content. Best of luck in your interview preparation!
.){kind=link}
0 Comments
PLACE YOUR COMMENT HERE
WARNING: DO NOT USE ABUSIVE LANGUAGE BECAUSE IT IS AGAINST THE LAW.
THE COMMENTS OF OUR READERS IS NOT OUR RESPONSIBILITY.