“200”, Aptitude Test Questions and Answers
for Statistician Grade II – MDA & LGA.
ABSTRACT
This document contains 200 multiple-choice
questions created to help candidates prepare for the Statistician Grade II –
MDA and LGA aptitude test in Tanzania. The questions are designed to feel like
the real exam, with tricky options that test understanding, thinking ability,
and decision-making. They cover important areas like statistics, data analysis,
sampling, and real-life work situations in government. Each question includes
the correct answer and a clear explanation to help learners understand the concept
better. Overall, this material is meant to build confidence, improve skills,
and give candidates a strong chance of passing the exam.
Prepared by: Statisticians.
Compiled by Johnson Yesaya.
0628729934.
Date: March 15, 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,
For Personal Use by Applicants Preparing
for MDA and LGA Statistician Grade II at Public Service Recruitment Service.
ALL
QUESTIONS ARE COMPILED TOGETHER.
QUESTION 1
A dataset collected from multiple LGAs shows unusually low variance despite diverse populations. What is the MOST plausible concern?
A. Data may be over-dispersed across regions B. Data collection tools were
inconsistent across LGAs C. Data may have been overly standardized or
manipulated D. Sample size is too large to detect variation
Answer: C
Rationale:
Low variance in a context where populations are naturally diverse is
suspicious. It suggests the data may have been smoothed, standardized
excessively, or even manipulated to appear uniform. Large sample sizes usually
increase detection of variation, not suppress it, and inconsistent tools would
more likely increase variability rather than reduce it.
QUESTION 2
A national survey
aims to ensure that small but important subgroups are not missed during
sampling. Which approach is MOST appropriate?
A. Increasing sample size uniformly across all regions B. Applying
disproportionate sampling to key subgroups C. Using systematic sampling with a
fixed interval D. Selecting only high-population areas
Answer: B
Rationale:
Disproportionate sampling ensures that small but important subgroups are
adequately represented, even if they are not large in population size. This
improves analytical power for subgroup analysis, unlike uniform expansion which
may still underrepresent smaller groups.
QUESTION 3
A statistician
notices that increasing sample size does not improve the accuracy of estimates.
What is the MOST likely explanation?
A. The data distribution is perfectly normal B. The sampling frame excludes key
population groups C. The confidence level is too low D. The sampling method is
biased or flawed
Answer: D
Rationale:
Accuracy is affected by bias, not just sample size. If the sampling process is
flawed, increasing sample size will only increase the volume of biased data
without improving correctness.
QUESTION 4
Which situation
BEST illustrates non-sampling error?
A. Using incorrect data entry procedures during processing B. Selecting too few
respondents for a survey C. Random variation in selecting households D. Drawing
a sample that is too small for inference
Answer: A
Rationale:
Non-sampling errors arise from issues like data entry mistakes, measurement
errors, or processing inaccuracies. These errors are not related to how the
sample is selected but rather how the data is handled after collection.
QUESTION 5
A report
concludes that a program caused improved outcomes, yet no control group was
used. What is the MAIN limitation?
A. Results cannot be generalized to other populations B. The data lacks
descriptive statistics C. The sample size is insufficient for inference D. Causality
cannot be reliably established
Answer: D
Rationale:
Without a control group, it is impossible to determine whether observed changes
are due to the program or other external factors. Establishing causality
requires comparison against a baseline or control condition.
QUESTION 6
Which indicator
is MOST appropriate for measuring inequality in income distribution?
A. Mean income level across regions B. Median household expenditure C. Gini
coefficient of income distribution D. Total national income growth rate
Answer: C
Rationale:
The Gini coefficient specifically measures inequality by quantifying how income
is distributed across a population. Mean and total income do not capture
disparities, while median provides central tendency but not inequality.
QUESTION 7
Two variables
show a strong correlation in a dataset. Which conclusion is MOST appropriate?
A. One variable fully determines the other B. The variables are independent C. The
relationship must be due to random error D. There may be an association, but
causality is not established
Answer: D
Rationale:
Correlation indicates association but does not prove causation. Other variables
or hidden factors may explain the relationship, requiring cautious
interpretation.
QUESTION 8
A dataset
contains extreme outliers that significantly affect the mean. Which measure is
MOST robust?
A. Arithmetic mean B. Mode C. Median D. Range
Answer: C
Rationale:
The median is resistant to extreme values because it depends on the central
position of data rather than magnitude. Outliers can heavily distort the mean
but have minimal effect on the median.
QUESTION 9
Which method BEST
reduces interviewer bias during data collection?
A. Using standardized questionnaires and training B. Increasing the number of
interviewers C. Selecting only experienced respondents D. Limiting the number
of questions asked
Answer: A
Rationale:
Standardization and proper training ensure consistency in how questions are
asked and responses recorded, minimizing interviewer influence on responses.
QUESTION 10
A government
statistician aggregates district data into regional totals. What is the PRIMARY
drawback?
A. Increased sampling error B. Loss of detailed local information C.
Overestimation of sample size D. Introduction of measurement bias
Answer: B
Rationale:
Aggregation reduces granularity, potentially masking important local variations
and patterns that are critical for targeted policy decisions.
QUESTION 11
Which scenario
BEST reflects selection bias?
A. Randomly selecting households across regions B. Using probability sampling
techniques C. Increasing sample size proportionally D. Surveying only urban
residents for national data
Answer: D
Rationale:
Excluding rural populations creates a biased sample that does not represent the
entire population, leading to skewed conclusions.
QUESTION 12
What is the MAIN
purpose of weighting survey data?
A. To reduce sample size requirements B. To eliminate all forms of bias C. To
adjust for unequal probabilities of selection D. To simplify data analysis
procedures
Answer: C
Rationale:
Weighting corrects for unequal representation in samples, ensuring results
better reflect the population structure. It cannot eliminate all bias but
improves representativeness.
QUESTION 13
Which is the MOST
appropriate use of administrative data in statistics?
A. Replacing all survey data collection methods B. Avoiding the need for data
validation C. Eliminating the need for sampling techniques D. Complementing
survey data for efficiency and coverage
Answer: D
Rationale:
Administrative data enhances statistical systems by providing continuous and
cost-effective data, but it must complement—not replace—survey data due to
potential quality limitations.
QUESTION 14
A statistician
observes that a sampling frame excludes remote households, yet results are
generalized nationally. What is the MOST critical issue?
A. Measurement inconsistency across regions B. Sampling bias affecting
representativeness C. Data entry errors during compilation D. Overestimation
due to large sample size
Answer: B
Rationale:
Excluding remote households introduces systematic undercoverage, which directly
leads to sampling bias. This compromises representativeness because certain
population segments are omitted entirely. Unlike measurement or entry errors,
which affect accuracy at observation level, sampling bias affects the validity
of conclusions drawn about the whole population, making it the most critical
issue.
QUESTION 15
A dataset shows
high variability within samples, but the average value remains consistent
across repeated samples. What does this MOST likely indicate?
A. Presence of systematic bias in sampling B. High precision with low
reliability C. Low precision with consistent central tendency D. Errors
concentrated in specific observations
Answer: C
Rationale:
High variability indicates low precision, while consistent averages suggest the
estimator is unbiased. This reflects stable central tendency but inconsistent
individual observations.
QUESTION 16
In designing a
national survey, stratification is introduced. What is the PRIMARY advantage?
A. Reducing total sample size without conditions B. Ensuring proportional
representation of subgroups C. Eliminating all sources of sampling error D.
Simplifying data analysis procedures
Answer: B
Rationale:
Stratified sampling ensures that important subgroups are adequately
represented, especially when populations are heterogeneous. It improves
precision and representativeness. While it may reduce variance, it does not
eliminate sampling error entirely nor necessarily reduce sample size without
careful allocation.
QUESTION 17
A statistician
finds that increasing sample size does not reduce bias. What is the BEST
explanation?
A. Bias is unrelated to sample size adjustments B. Larger samples increase
measurement error C. Bias is caused by random variation only D. Sampling
distribution has shifted unpredictably
Answer: A
Rationale:
Bias is a systematic error, not random error. Increasing sample size reduces
random error (variance) but does not correct systematic flaws such as poor
sampling design or exclusion of groups. Therefore, bias remains unchanged
regardless of sample size increases.
QUESTION 18
A report presents
percentages without indicating base values. What is the MAIN risk?
A. Increased computational complexity B. Loss of statistical significance C.
Misinterpretation due to lack of context D. Inconsistency in data coding
Answer: C
Rationale:
Percentages without base values can be misleading because they hide the actual
magnitude of data. A percentage change from a small base may appear large but
be insignificant in reality. This leads to misinterpretation and poor
decision-making, especially in policy contexts.
QUESTION 19
A survey
experiences a high rate of unanswered questions from respondents. What type of
error does this represent?
A. Sampling error B. Processing error C. Measurement bias D. Non-response error
Answer: D
Rationale:
Non-response error occurs when participants fail to provide information,
leading to missing data that may bias results if the non-respondents differ
systematically from respondents.
QUESTION 20
A survey uses
convenience sampling due to time constraints. What is the MOST likely
consequence?
A. Increased statistical efficiency B. Reduced operational cost without
trade-offs C. Limited generalizability of findings D. Improved accuracy of
population estimates
Answer: C
Rationale:
Convenience sampling does not ensure representativeness, as participants are
selected based on accessibility rather than randomness. This leads to biased
samples and limits the ability to generalize findings to the broader
population, even if the data appears internally consistent.
QUESTION 21
In hypothesis
testing, failing to reject a false null hypothesis is known as:
A. Type I error B. Measurement bias C. Sampling error D. Type II error
Answer: D
Rationale:
A Type II error occurs when a false null hypothesis is not rejected. This means
the test fails to detect an actual effect. It differs from Type I error, which
involves rejecting a true null hypothesis. Understanding this distinction is
critical in statistical inference.
QUESTION 22
A dataset is
highly skewed to the right. Which measure is MOST appropriate for central
tendency?
A. Mean B. Mode C. Median D. Range
Answer: C
Rationale:
In skewed distributions, the mean is influenced by extreme values, making it
less reliable. The median, being the middle value, is resistant to outliers and
better represents the central tendency of skewed data.
QUESTION 23
A statistician
uses outdated population data for weighting survey results. What is the MAIN
implication?
A. Introduction of systematic bias B. Increased sampling variance C. Reduced
data collection cost D. Improved comparability over time
Answer: A
Rationale:
Using outdated weights misrepresents current population structure, leading to
biased estimates. This affects the validity of conclusions since results no
longer reflect the true population distribution.
QUESTION 24
Which of the
following BEST defines a parameter?
A. Summary of a sample characteristic B. Measure derived from observed data C.
Numerical value describing a population D. Estimate subject to sampling
variability
Answer: C
Rationale:
A parameter describes a population characteristic, such as population mean or
variance. Unlike statistics, which are derived from samples, parameters are
fixed but usually unknown.
QUESTION 25
A large sample
still produces inconsistent results across repeated studies. What is the MOST
likely issue?
A. Persistent variability due to design issues B. Proper stratification across
groups C. High measurement precision across samples D. Inconsistency in the
sampling frame or design
Answer: D
Rationale:
Large samples should reduce random variation. Persistent inconsistency suggests
structural issues in sampling design or frame, not random error.
QUESTION 26
Which approach
BEST improves data reliability in field surveys?
A. Increasing questionnaire length significantly B. Reducing number of
enumerators C. Standardizing data collection procedures D. Avoiding supervision
during data collection
Answer: C
Rationale:
Standardization ensures consistency in how data is collected across different
enumerators and locations. This minimizes variability due to human factors and
improves reliability.
QUESTION 27
A correlation
coefficient close to zero indicates:
A. Strong linear relationship B. Weak or no linear relationship C. Perfect
inverse relationship D. Causal relationship between variables
Answer: B
Rationale:
Correlation measures linear association. A value near zero suggests no linear
relationship, though non-linear relationships may still exist. It does not
imply causation.
QUESTION 28
Which situation
BEST justifies use of cluster sampling?
A. Homogeneous population spread evenly B. Need for maximum statistical
precision C. Requirement for detailed subgroup analysis D. Population widely
dispersed geographically
Answer: D
Rationale:
Cluster sampling is efficient when populations are geographically dispersed, as
it reduces travel and operational costs. However, it may increase sampling
error compared to stratified sampling.
QUESTION 29
A confidence
interval becomes narrower when:
A. Sample size decreases B. Confidence level increases C. Sample size increases
D. Population variance increases
Answer: C
Rationale:
Larger sample sizes reduce standard error, leading to narrower confidence
intervals. Increasing confidence level or variance widens the interval instead.
QUESTION 30
Which factor MOST
threatens validity of administrative data?
A. High frequency of data updates B. Inconsistent definitions across sources C.
Large volume of records collected D. Automated data processing systems
Answer: B
Rationale:
Different definitions across institutions lead to incomparable datasets,
undermining validity. Even large or frequently updated data becomes unreliable
if concepts are inconsistent.
📘 Get the Full Aptitude Test Questions PDF through your Gmail (Questions 1–200)
You’ve just accessed the first 30 questions. The full set of 200 expertly prepared aptitude test questions for Statistician Grade II – MDA & LGA. 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!
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.