,TABLE OF CONTENT
Part I: General Concepts & Techniques
Chapter 1: Measurement
Chapter 2: Types of Studies
Chapter 3: Frequency Distributions
Chapter 4: Summary Statistics
Chapter 5: Probability Concepts
Chapter 6: Binomial Probability Distributions
Chapter 7: Normal Probability Distributions
Chapter 8: Introduction to Statistical Inference
Chapter 9: Basics of Hypothesis Testing
Chapter 10: Basics of Confidence Intervals
Part II: Quantitative Response Variable
Chapter 11: Inference About a Mean
Chapter 12: Comparing Independent Means
Chapter 13: Comparing Several Means (One-Way ANOVA)
Chapter 14: Correlation and Regression
Chapter 15: Multiple Linear Regression
Part III: Categorical Response Variable
Chapter 16: Inference About a Proportion
Chapter 17: Comparing Two Proportions
Chapter 18: Cross-Tabulated Counts
Chapter 19: Stratified Two-by-Two Tables
,Chapter 1: Measurement – Test Bank (20 MCQs)
1. A public health researcher collects data on 150 individuals’ blood pressure
readings to study hypertension prevalence. The researcher wants to summarize
the typical blood pressure in this population. Which measure of central tendency
is most appropriate?
A. Mode
B. Median
C. Mean
D. Standard deviation
Answer: C
Rationale: Blood pressure is a continuous variable measured on a ratio scale. The
mean is appropriate for symmetric, continuous data, reflecting the arithmetic
average. Mode is better for nominal data; median for skewed distributions.
Keywords: Central tendency, mean, ratio scale, continuous variable
2. A dataset includes the marital status of participants: single, married, divorced,
widowed. Which scale of measurement does this variable represent?
A. Nominal
B. Ordinal
C. Interval
D. Ratio
Answer: A
Rationale: Marital status categories are names without inherent order. Nominal
scales classify data into distinct categories; ordinal scales imply order; interval
and ratio scales are numeric.
Keywords: Nominal scale, categorical data, qualitative variable
,3. During data entry, a researcher notices some participants’ ages were
recorded incorrectly as negative numbers. This is an example of:
A. Random error
B. Systematic error
C. Precision
D. Reliability
Answer: B
Rationale: A consistent error in data entry (negative ages) is a systematic error
that biases results. Random errors fluctuate around the true value.
Keywords: Systematic error, data quality, measurement error
4. A dataset contains exam scores for 50 students. The researcher wants to
determine how variable the scores are around the mean. Which measure should
be calculated?
A. Mode
B. Standard deviation
C. Median
D. Frequency table
Answer: B
Rationale: Standard deviation quantifies the spread of continuous, interval/ratio
data around the mean, allowing assessment of variability.
Keywords: Variability, standard deviation, spread, continuous data
5. A researcher enters patient data into a spreadsheet with numeric codes for
“Yes” (1) and “No” (0) responses to smoking status. This process is known as:
A. Data cleaning
B. Coding
C. Imputation
,D. Stratification
Answer: B
Rationale: Converting categorical responses into numeric codes for analysis is
coding, a standard step in data organization.
Keywords: Data coding, categorical data, data preparation
6. In a dataset, blood type (A, B, AB, O) is recorded. Which summary statistic is
most appropriate to describe the central tendency?
A. Mean
B. Median
C. Mode
D. Standard deviation
Answer: C
Rationale: Blood type is nominal; the mode identifies the most frequently
occurring category. Mean and median are inappropriate for non-numeric data.
Keywords: Mode, nominal variable, categorical summary
7. A survey collects satisfaction ratings on a 5-point Likert scale. Which of the
following is correct regarding the scale type?
A. Nominal
B. Ordinal
C. Interval
D. Ratio
Answer: B
Rationale: Likert scales have an inherent order but unequal intervals between
points. They are ordinal variables; arithmetic mean is only sometimes
appropriate.
Keywords: Ordinal scale, Likert scale, ranking
,8. When a public health dataset has missing values, which method ensures the
highest data quality without introducing bias?
A. Deleting all incomplete cases
B. Random replacement of missing values
C. Using validated imputation methods
D. Ignoring the missing data
Answer: C
Rationale: Proper imputation maintains data integrity and reduces bias,
preserving representativeness. Random replacement or deletion can distort
results.
Keywords: Data quality, missing data, imputation, bias
9. The weight of newborns is measured using a precise electronic scale. Over
repeated measurements, the results vary slightly due to device sensitivity. This
variability represents:
A. Accuracy
B. Random error
C. Systematic error
D. Reliability
Answer: B
Rationale: Small fluctuations around the true weight are random errors; they do
not bias results but affect precision. Accuracy refers to closeness to the true
value.
Keywords: Random error, precision, variability
10. A researcher calculates the interquartile range (IQR) for systolic blood
pressure. What does this measure represent?
, A. Central tendency
B. Variability around the mean
C. Spread of the middle 50% of data
D. Maximum minus minimum
Answer: C
Rationale: IQR captures the range between the 25th and 75th percentiles,
describing dispersion for skewed or non-normal distributions.
Keywords: Interquartile range, variability, percentiles, data spread
11. Which of the following is a ratio scale variable?
A. Pain rating on 0–10 scale
B. Body mass index
C. Blood type
D. Education level
Answer: B
Rationale: BMI has a true zero, allows ratio comparisons (e.g., BMI of 30 is twice
15), and arithmetic operations are meaningful. Pain rating is ordinal; blood type
is nominal; education is often ordinal.
Keywords: Ratio scale, continuous variable, meaningful zero
12. A researcher notes that some lab measurements differ by 5–10% from
standard values consistently. This type of error primarily affects:
A. Precision
B. Accuracy
C. Random variability
D. Completeness
Answer: B
Rationale: Systematic deviations from the true value reduce accuracy, even if
Part I: General Concepts & Techniques
Chapter 1: Measurement
Chapter 2: Types of Studies
Chapter 3: Frequency Distributions
Chapter 4: Summary Statistics
Chapter 5: Probability Concepts
Chapter 6: Binomial Probability Distributions
Chapter 7: Normal Probability Distributions
Chapter 8: Introduction to Statistical Inference
Chapter 9: Basics of Hypothesis Testing
Chapter 10: Basics of Confidence Intervals
Part II: Quantitative Response Variable
Chapter 11: Inference About a Mean
Chapter 12: Comparing Independent Means
Chapter 13: Comparing Several Means (One-Way ANOVA)
Chapter 14: Correlation and Regression
Chapter 15: Multiple Linear Regression
Part III: Categorical Response Variable
Chapter 16: Inference About a Proportion
Chapter 17: Comparing Two Proportions
Chapter 18: Cross-Tabulated Counts
Chapter 19: Stratified Two-by-Two Tables
,Chapter 1: Measurement – Test Bank (20 MCQs)
1. A public health researcher collects data on 150 individuals’ blood pressure
readings to study hypertension prevalence. The researcher wants to summarize
the typical blood pressure in this population. Which measure of central tendency
is most appropriate?
A. Mode
B. Median
C. Mean
D. Standard deviation
Answer: C
Rationale: Blood pressure is a continuous variable measured on a ratio scale. The
mean is appropriate for symmetric, continuous data, reflecting the arithmetic
average. Mode is better for nominal data; median for skewed distributions.
Keywords: Central tendency, mean, ratio scale, continuous variable
2. A dataset includes the marital status of participants: single, married, divorced,
widowed. Which scale of measurement does this variable represent?
A. Nominal
B. Ordinal
C. Interval
D. Ratio
Answer: A
Rationale: Marital status categories are names without inherent order. Nominal
scales classify data into distinct categories; ordinal scales imply order; interval
and ratio scales are numeric.
Keywords: Nominal scale, categorical data, qualitative variable
,3. During data entry, a researcher notices some participants’ ages were
recorded incorrectly as negative numbers. This is an example of:
A. Random error
B. Systematic error
C. Precision
D. Reliability
Answer: B
Rationale: A consistent error in data entry (negative ages) is a systematic error
that biases results. Random errors fluctuate around the true value.
Keywords: Systematic error, data quality, measurement error
4. A dataset contains exam scores for 50 students. The researcher wants to
determine how variable the scores are around the mean. Which measure should
be calculated?
A. Mode
B. Standard deviation
C. Median
D. Frequency table
Answer: B
Rationale: Standard deviation quantifies the spread of continuous, interval/ratio
data around the mean, allowing assessment of variability.
Keywords: Variability, standard deviation, spread, continuous data
5. A researcher enters patient data into a spreadsheet with numeric codes for
“Yes” (1) and “No” (0) responses to smoking status. This process is known as:
A. Data cleaning
B. Coding
C. Imputation
,D. Stratification
Answer: B
Rationale: Converting categorical responses into numeric codes for analysis is
coding, a standard step in data organization.
Keywords: Data coding, categorical data, data preparation
6. In a dataset, blood type (A, B, AB, O) is recorded. Which summary statistic is
most appropriate to describe the central tendency?
A. Mean
B. Median
C. Mode
D. Standard deviation
Answer: C
Rationale: Blood type is nominal; the mode identifies the most frequently
occurring category. Mean and median are inappropriate for non-numeric data.
Keywords: Mode, nominal variable, categorical summary
7. A survey collects satisfaction ratings on a 5-point Likert scale. Which of the
following is correct regarding the scale type?
A. Nominal
B. Ordinal
C. Interval
D. Ratio
Answer: B
Rationale: Likert scales have an inherent order but unequal intervals between
points. They are ordinal variables; arithmetic mean is only sometimes
appropriate.
Keywords: Ordinal scale, Likert scale, ranking
,8. When a public health dataset has missing values, which method ensures the
highest data quality without introducing bias?
A. Deleting all incomplete cases
B. Random replacement of missing values
C. Using validated imputation methods
D. Ignoring the missing data
Answer: C
Rationale: Proper imputation maintains data integrity and reduces bias,
preserving representativeness. Random replacement or deletion can distort
results.
Keywords: Data quality, missing data, imputation, bias
9. The weight of newborns is measured using a precise electronic scale. Over
repeated measurements, the results vary slightly due to device sensitivity. This
variability represents:
A. Accuracy
B. Random error
C. Systematic error
D. Reliability
Answer: B
Rationale: Small fluctuations around the true weight are random errors; they do
not bias results but affect precision. Accuracy refers to closeness to the true
value.
Keywords: Random error, precision, variability
10. A researcher calculates the interquartile range (IQR) for systolic blood
pressure. What does this measure represent?
, A. Central tendency
B. Variability around the mean
C. Spread of the middle 50% of data
D. Maximum minus minimum
Answer: C
Rationale: IQR captures the range between the 25th and 75th percentiles,
describing dispersion for skewed or non-normal distributions.
Keywords: Interquartile range, variability, percentiles, data spread
11. Which of the following is a ratio scale variable?
A. Pain rating on 0–10 scale
B. Body mass index
C. Blood type
D. Education level
Answer: B
Rationale: BMI has a true zero, allows ratio comparisons (e.g., BMI of 30 is twice
15), and arithmetic operations are meaningful. Pain rating is ordinal; blood type
is nominal; education is often ordinal.
Keywords: Ratio scale, continuous variable, meaningful zero
12. A researcher notes that some lab measurements differ by 5–10% from
standard values consistently. This type of error primarily affects:
A. Precision
B. Accuracy
C. Random variability
D. Completeness
Answer: B
Rationale: Systematic deviations from the true value reduce accuracy, even if
