Biostatistics Practice Questions

Q: In a clinical trial, blood pressure was measured in a group of patients before and after treatment. Which of the following tests will be suitable as a test of significance?

Paired t-test. ### Explanation **Why Paired t-test is correct:** The scenario describes a **"before and after"** study design involving the same group of individuals. In biostatistics, when two sets of observations are made on the same subjects (paired data), and the variable being measured is **quantitative/numerical** (e.g., Blood Pressure in mmHg), the **Paired t-test** is the most appropriate parametric test. It compares the mean difference between the two sets of observations to determine if the treatment effect is statistically significant. **Why the other options are incorrect:** * **Mann-Whitney U test:** This is a non-parametric alternative to the unpaired t-test. It is used for comparing two independent groups when the data is ordinal or not normally distributed. * **Student’s t-test (Unpaired/Independent):** This is used to compare the means of two **independent** groups (e.g., comparing BP between Group A and Group B). It cannot be used for "before and after" data in the same group. * **ANOVA (Analysis of Variance):** This is used when comparing the means of **three or more** independent groups. **High-Yield Clinical Pearls for NEET-PG:** 1. **Parametric vs. Non-Parametric:** If the data in this question were non-normally distributed, the non-parametric equivalent of the Paired t-test would be the **Wilcoxon Signed-Rank Test**. 2. **Key Identifier:** Whenever you see "before and after," "pre-test and post-test," or "matched pairs" in a question involving numerical data, think **Paired t-test**. 3. **Qualitative Data:** If the study measured a qualitative change (e.g., Improved vs. Not Improved) before and after treatment, the **McNemar Test** would be used instead.

Q: The 'Design Effect' is associated with which of the following sampling techniques?

Cluster sampling. ### Explanation **Why Cluster Sampling is Correct:** The **Design Effect (Deff)** is a correction factor used to account for the loss of statistical efficiency when using **Cluster Sampling** instead of Simple Random Sampling (SRS). In cluster sampling, individuals within a cluster (e.g., a village or household) tend to be more similar to each other than to individuals in other clusters (intra-cluster correlation). This "homogeneity" reduces the amount of unique information collected, leading to a larger standard error. To compensate for this, the sample size calculated for SRS must be multiplied by the Design Effect to achieve the same power. For example, in WHO’s EPI cluster surveys for immunization, a default Design Effect of **2** is often used. **Why Other Options are Incorrect:** * **A. Stratified Sampling:** This technique usually *increases* precision by dividing the population into homogenous subgroups. The design effect is typically < 1, meaning a smaller sample size might suffice compared to SRS. * **B. Systemic Sampling:** This involves selecting subjects at fixed intervals (e.g., every 10th person). While it is a type of probability sampling, it does not inherently require a design effect correction unless clustering occurs. * **D. Simple Random Sampling (SRS):** This is the "gold standard" or baseline for comparison. By definition, the Design Effect for SRS is **1.0**. **High-Yield Pearls for NEET-PG:** * **Formula:** $Deff = \frac{\text{Variance of Cluster Sample}}{\text{Variance of Simple Random Sample}}$. * **Sample Size Calculation:** Total Sample Size = $n (SRS) \times \text{Design Effect}$. * **Cluster Sampling** is the most common method used in field health surveys (e.g., NFHS, Vaccination coverage) because it is logistically easier and more cost-effective than SRS. * **Key Concept:** As intra-cluster correlation increases, the Design Effect increases.

Q: A district has the following population data and average population served per doctor. Calculate the harmonic mean: AREA | POPULATION SERVED PER DOCTOR | POPULATION | NUMBER OF DOCTORS ---------------------------------------------------------------------- RURAL | 1000 | 50000 | 50 URBAN | 500 | 50000 | 100 TOTAL | - | 100000 | 150

667. ### Explanation **1. Why Option C (667) is Correct:** In biostatistics, the **Harmonic Mean (HM)** is the preferred measure of central tendency for **rates and ratios** (e.g., speed, population per doctor, or cases per unit time). It is defined as the reciprocal of the arithmetic mean of the reciprocals of the values. However, when dealing with groups of different sizes (as in this population data), we calculate the **Weighted Harmonic Mean**. The formula is: $$HM = \frac{\text{Total Population}}{\sum (\frac{\text{Population}}{\text{Value}})}$$ In this case, the "Value" is the population served per doctor. * **Total Population** = 100,000 * **Total Doctors** = (50,000 / 1000) + (50,000 / 500) = 50 + 100 = 150 * **Calculation:** $100,000 / 150 = 666.67$ Rounding off gives **667**. This represents the true average workload per doctor across the entire district. **2. Why Other Options are Incorrect:** * **Option A (500):** This is simply the lower value (Urban) and ignores the Rural data. * **Option B (567):** This value does not correspond to standard statistical measures for this dataset. * **Option D (750):** This is the **Arithmetic Mean** of the two rates $(1000 + 500) / 2$. The Arithmetic Mean overestimates the average when dealing with rates and should be avoided here. **3. High-Yield Clinical Pearls for NEET-PG:** * **Arithmetic Mean:** Best for normally distributed data (e.g., Height, BP). * **Geometric Mean:** Best for data following a logarithmic scale or growth rates (e.g., bacterial counts, serial dilutions, titers). * **Harmonic Mean:** Best for **rates, ratios, and speeds**. It is always the smallest of the three means ($AM > GM > HM$). * **Median:** Best for skewed data or data with extreme outliers (e.g., survival time, incubation period).

Q: Which of the following statements regarding the mode is false?

The mode cannot be calculated for any type of data.. ### Explanation **Why Option C is the Correct (False) Statement:** In biostatistics, the **Mode** is defined as the value that occurs most frequently in a dataset. The statement "The mode cannot be calculated for any type of data" is false because the mode is, in fact, the **only** measure of central tendency that can be used for all levels of data: nominal, ordinal, interval, and ratio. It is particularly uniquely suited for qualitative (nominal) data where calculating a mean or median is mathematically impossible. **Analysis of Incorrect Options:** * **Option A (True):** If every value in a dataset occurs only once (e.g., 1, 2, 3, 4, 5), there is no repeating value, and thus, the dataset has **no mode**. * **Option B (True):** A dataset can be **bimodal** (two modes) or **multimodal** (more than two modes) if multiple values share the highest frequency. * **Option D (True):** The mode is the most appropriate measure for **nominal data** (e.g., determining the most common blood group in a population or the most frequent side effect of a drug). **High-Yield Clinical Pearls for NEET-PG:** * **Relationship in Normal Distribution:** Mean = Median = Mode. * **Skewed Distributions:** * **Positively Skewed:** Mean > Median > Mode (Mode is at the peak). * **Negatively Skewed:** Mode > Median > Mean. * **Stability:** The mode is the least stable measure of central tendency as it can change significantly with small changes in the dataset. * **Empirical Formula:** $Mode = (3 \times \text{Median}) - (2 \times \text{Mean})$.

Q: Which of the following is not a measure of dispersion?

Mode. **Explanation:** In biostatistics, data is summarized using two primary types of measures: **Measures of Central Tendency** (averages) and **Measures of Dispersion** (variability). **Why Mode is the correct answer:** **Mode** is a measure of **Central Tendency**, not dispersion. It is defined as the value that occurs most frequently in a data set. While it identifies the "center" or most popular observation, it provides no information about how spread out or scattered the data points are from one another. **Why the other options are incorrect (Measures of Dispersion):** * **Range (D):** The simplest measure of dispersion, calculated as the difference between the maximum and minimum values. * **Mean Deviation (A):** The arithmetic average of the absolute deviations of observations from the mean. * **Standard Deviation (B):** The most commonly used measure of dispersion in medical research. It quantifies the amount of variation or dispersion of a set of values around the mean. **High-Yield Clinical Pearls for NEET-PG:** * **Measures of Central Tendency:** Mean (arithmetic average), Median (middle value), and Mode (most frequent). * **Measures of Dispersion:** Range, Interquartile Range (IQR), Mean Deviation, Standard Deviation, and Coefficient of Variation. * **Normal Distribution:** In a perfectly symmetrical bell-shaped curve, Mean = Median = Mode. * **Standard Deviation (SD):** In a normal distribution, 1 SD covers 68% of data, 2 SD covers 95%, and 3 SD covers 99.7%. * **Relative Dispersion:** The **Coefficient of Variation** is used to compare dispersion between two series with different units.

Q: When the mean, median, and mode of a distribution are all zero, what type of distribution is it?

Standard normal distribution. ### Explanation **1. Why the Correct Answer is Right:** In a **Normal Distribution** (Gaussian distribution), the curve is perfectly symmetrical and bell-shaped. In such a distribution, the **Mean, Median, and Mode are all equal** (Mean = Median = Mode). A **Standard Normal Distribution** is a specific type of normal distribution where the data is standardized using Z-scores. It is defined by two specific parameters: * **Mean ($\mu$) = 0** * **Standard Deviation ($\sigma$) = 1** Since it is a symmetrical distribution, the Median and Mode also coincide with the Mean at the center. Therefore, when all three measures of central tendency are zero, it must be a Standard Normal Distribution. **2. Why the Incorrect Options are Wrong:** * **B & C (Skewed Distributions):** In skewed distributions, the mean, median, and mode are pulled apart. In **Positively skewed** data, Mean > Median > Mode. In **Negatively skewed** data, Mode > Median > Mean. They cannot all be equal to zero. * **D (J-shaped Distribution):** This is an asymmetrical distribution where the frequency is at its maximum at one end of the scale. It does not follow the central symmetry required for the mean, median, and mode to coincide at zero. **3. Clinical Pearls & High-Yield Facts for NEET-PG:** * **Z-score:** Indicates how many standard deviations a value is from the mean. In a standard normal distribution, the Z-score of the mean is 0. * **Area under the curve:** In a normal distribution, **68%** of values fall within ±1 SD, **95%** within ±2 SD (specifically 1.96 SD), and **99.7%** within ±3 SD. * **Symmetry Rule:** If you know a distribution is perfectly symmetrical and unimodal, the Mean, Median, and Mode will always be identical.

Q: Which of the following variables is measured on an ordinal scale?

Severity of anemia. ### Explanation In biostatistics, data is classified into four levels of measurement: Nominal, Ordinal, Interval, and Ratio. **Why "Severity of Anemia" is correct:** The **Ordinal scale** is used for data that can be categorized and placed in a **logical rank or order**, but the mathematical distance between the ranks is not defined. Severity of anemia is typically classified as *Mild, Moderate, or Severe*. While we know "Severe" is worse than "Mild," we cannot mathematically quantify exactly how much worse it is using just these labels. Other common ordinal examples include cancer staging (Stage I-IV) or Likert scales (Satisfied to Dissatisfied). **Analysis of Incorrect Options:** * **A. Type of Anemia:** This is a **Nominal scale**. It categorizes data into groups based on names or labels (e.g., Microcytic, Macrocytic, Normocytic) without any inherent quantitative ranking or order. * **C. Hemoglobin level & D. Serum ferritin level:** These are **Ratio scales** (a type of continuous/numerical data). They have a true zero point and consistent intervals between values. For example, a hemoglobin of 14 g/dL is exactly double 7 g/dL. **High-Yield Clinical Pearls for NEET-PG:** * **NOIR Mnemonic:** **N**ominal (Name only), **O**rdinal (Order/Rank), **I**nterval (Fixed distance, no true zero), **R**atio (True zero). * **Qualitative Data:** Includes Nominal and Ordinal scales. * **Quantitative Data:** Includes Discrete (whole numbers, e.g., number of beds) and Continuous (e.g., height, weight, biochemical values). * **Key Distinction:** If you can rank the categories but cannot subtract them, it is **Ordinal**. If you can measure the exact difference between values, it is **Numerical/Ratio**.

Q: Which type of diagram best demonstrates a secular trend?

Line diagram. ### Explanation **1. Why Line Diagram is Correct:** A **secular trend** refers to the long-term changes (increases or decreases) in the occurrence of a disease or health event over a prolonged period (usually years or decades). A **Line Diagram** is the most suitable tool for this because it plots data points chronologically and connects them, effectively showing the **direction and rate of change** over time. It allows for the visualization of trends, fluctuations, and projections, making it the gold standard for time-series data in epidemiology. **2. Why Other Options are Incorrect:** * **Bar Diagram:** These are used for **discrete (qualitative)** data to compare different categories (e.g., number of cases in different cities). They do not show continuous trends over time as effectively as lines. * **Histogram:** This is used for **continuous quantitative** data to show frequency distributions (e.g., age distribution of a population). It represents a snapshot in time rather than a long-term trend. * **Pie Chart:** This is used to show the **proportional distribution** of a whole (e.g., the percentage of different causes of maternal mortality). It cannot depict changes over time. **3. High-Yield Clinical Pearls for NEET-PG:** * **Time Trends:** * *Secular:* Long-term (e.g., the decline of Polio over decades). * *Periodic:* Seasonal (e.g., Dengue in monsoons) or Cyclic (e.g., Measles epidemics every 2-3 years). * **Frequency Polygon:** Created by joining the midpoints of a histogram; also used for frequency distributions. * **Scatter Diagram:** Used to show the **correlation** between two continuous variables. * **Ogive:** A graph representing cumulative frequency.

Question 1

In a clinical trial, blood pressure was measured in a group of patients before and after treatment. Which of the following tests will be suitable as a test of significance?

Accepted Answer

Paired t-test

Answer

Mann Whitney U test

Answer

Student's t-test

Answer

ANOVA

Question 2

The 'Design Effect' is associated with which of the following sampling techniques?

Accepted Answer

Cluster sampling

Answer

Stratified sampling

Answer

Systemic sampling

Answer

Simple Random Sampling

Question 3

A district has the following population data and average population served per doctor. Calculate the harmonic mean:

AREA | POPULATION SERVED PER DOCTOR | POPULATION | NUMBER OF DOCTORS
----------------------------------------------------------------------
RURAL | 1000 | 50000 | 50
URBAN | 500 | 50000 | 100
TOTAL | - | 100000 | 150

Accepted Answer

667

Answer

500

Answer

567

Answer

750

Question 4

Which of the following statements regarding the mode is false?

Accepted Answer

The mode cannot be calculated for any type of data.

Answer

Some datasets may have no mode.

Answer

A dataset can have more than one mode.

Answer

The mode can be calculated for nominal data.

Question 5

Which of the following is not a measure of dispersion?

Accepted Answer

Mode

Answer

Mean deviation

Answer

Standard deviation

Answer

Range

Question 6

When the mean, median, and mode of a distribution are all zero, what type of distribution is it?

Accepted Answer

Standard normal distribution

Answer

Negatively skewed distribution

Answer

Positively skewed distribution

Answer

J-shaped distribution

Question 7

Which of the following variables is measured on an ordinal scale?

Accepted Answer

Severity of anemia

Answer

Type of anemia

Answer

Hemoglobin level

Answer

Serum ferritin level

Question 8

Regarding crude birth rate, all are true except?

Accepted Answer

Better measure of fertility than general fertility rate

Answer

Unaffected by age distribution

Answer

Indicator of fertility

Answer

Excludes stillbirths

Question 9

Which type of diagram best demonstrates a secular trend?

Accepted Answer

Line diagram

Answer

Bar diagram

Answer

Histogram

Answer

Pie chart

Question 10

Total cholesterol level is represented by the equation: Total cholesterol level = a + b (calorie intake) + c (physical activity) + d (body mass index). This equation is an example of which type of regression?

Accepted Answer

Multiple linear regression

Answer

Simple linear regression

Answer

Simple curvilinear regression

Answer

Multiple logistic regression

Biostatistics — MCQs

Biostatistics — MCQs

On this page

Practice by Chapter

Want unlimited practice?