Biostatistics Practice Questions

Q: A study is conducted to compare the mean hemoglobin (Hb) levels between two independent groups. Which statistical test is most appropriate?

Unpaired t-test. ***Unpaired t-test*** - This test is specifically used to compare the **means** of a continuous outcome variable (like hemoglobin level) between **two independent, unrelated groups**. - It is based on the assumption that the data is normally distributed and variances are equal (though modifications exist if variances are unequal, known as Welch's t-test). *Paired t-test* - The paired t-test is used when the data comes from **dependent** or **related groups**, such as measuring the same individuals before and after an intervention (pre-post study). - Since the question specifies two **independent** groups, this test is incorrect. *Chi-square test* - This test is used to analyze the association or difference between **two or more categorical variables** (e.g., comparing proportions or frequencies in nominal data). - It is unsuitable for comparing the **mean** of a **continuous variable** like Hb levels. *ANOVA* - Analysis of Variance (ANOVA) is used to compare the **means** of a continuous variable among **three or more independent groups**. - Since the study involves only **two groups**, the unpaired t-test is the simpler and more conventional choice, although ANOVA yields the same result when reduced to two groups.

Q: In a village of 100 children, 10 children have a past history of measles (i.e., they are not at risk now), 20 new cases of measles were reported this year. What is the incidence of measles in this population for the year?

22.22 %. ***22.22 %*** - Incidence is calculated as the ratio of **new cases** (20) to the **population at risk** (susceptible population) over the specified period. - The **population at risk** is the total population (100) minus those who are already immune (10), making the denominator 90. Incidence = (20/90) × 100 = **22.22 %**. *20 %* - This result is obtained by incorrectly using the **total population** (100) as the denominator (20/100 × 100), ignoring the already immune group. - Using the total population in the denominator leads to an underestimate of the true **attack rate** or incidence among the susceptible group. *10 %* - This figure represents the proportion of children who had suffered from measles in the past (10/100), reflecting a form of **past prevalence**, not incidence. - Incidence focuses exclusively on the **new cases** that developed within the year. *30 %* - This percentage represents the **cumulative prevalence** at the end of the year, including both old (10) and new (20) cases, divided by the total population (30/100). - Incidence requires the denominator to be the **population at risk** (those who could develop the disease), not the total population.

Q: The given image shows:

Spaghetti plot. ***Spaghetti plot*** - A **spaghetti plot** displays multiple overlapping individual trajectory lines, with each line representing one subject's data over time or across conditions, creating a visual that resembles **strands of spaghetti**. - It is commonly used in **longitudinal studies** to show individual patient responses or changes, allowing visualization of both individual patterns and overall trends. *Kaplan Meier plot* - A Kaplan-Meier plot is a **survival curve** that shows the probability of survival over time with characteristic **step-wise decreases** at each event occurrence. - It is specifically used for **time-to-event analysis** and does not show multiple individual trajectory lines like a spaghetti plot. *Funnel plot* - A funnel plot is used in **meta-analyses** to assess **publication bias** by plotting study effect sizes against their precision (standard error). - It creates a **symmetrical funnel shape** when no bias is present and does not show individual subject trajectories over time. *Forest plot* - A forest plot is used in **meta-analyses** to display individual study results and **pooled effect estimates** with confidence intervals. - It features **squares or diamonds** representing effect sizes with **horizontal lines** for confidence intervals, not overlapping trajectory lines.

Q: The given image shows:

Funnel plot. ***Funnel plot*** - This graph displays individual study results (trial size vs. effect size), forming a **funnel-like shape** when no publication bias is present. - The observed asymmetry in the plot, where many small trials with a larger effect size are on one side, suggests potential **publication bias**, indicating this is a funnel plot. *Kaplan Meier plot* - A Kaplan-Meier plot is used to estimate the **survival function** from lifetime data. - It displays the probability of an event (e.g., death, disease remission) over time, characterized by a **stepped curve**. *Spaghetti plot* - A spaghetti plot is used to visualize longitudinal data, showing **multiple individual trajectories** over time. - Each line in a spaghetti plot represents data from a **single subject** or entity across different time points. *Forest plot* - A forest plot graphically presents the results of individual studies included in a **meta-analysis** and the overall pooled estimate. - It typically shows **effect size** and **confidence intervals** for each study, often represented by squares and horizontal lines.

Q: The given image shows:

Funnel plot. **Funnel plot** - A **funnel plot** displays the **effect sizes** from individual studies against a measure of their precision (e.g., sample size or standard error), typically in the context of a meta-analysis. - The characteristic **funnel shape** (or inverted triangle) arises because smaller studies (lower precision) have wider confidence intervals and thus show more variability, while larger studies (higher precision) cluster more closely around the true effect size. *Kaplan Meier plot* - A **Kaplan-Meier plot** is used to estimate and display **survival probabilities** over time, particularly in clinical trials or observational studies. - It shows a stepwise-decreasing curve as events (e.g., death, disease recurrence) occur, which is not what is depicted in the image. *Spaghetti plot* - A **spaghetti plot** displays the **individual trajectories** of multiple subjects over time, often used to visualize longitudinal data. - Each line in a spaghetti plot represents a single subject's path, which is distinct from the scattered points representing individual studies on the given image. *Forest plot* - A **forest plot** is a graphical display used in **meta-analyses** to present the results of individual studies and their combined effect. - It typically shows the effect estimate and confidence interval for each study as horizontal lines, with a diamond representing the overall pooled estimate, which is different from the funnel shape seen here.

Q: A new diagnostic test is introduced. Based on the image provided showing performance curves of both tests, what characteristics does the new test demonstrate compared to the old one?

Higher sensitivity and higher specificity. ***Correct Option: Higher sensitivity and higher specificity*** The blue curves (representing the new test) show **better separation between 'healthy' and 'diseased' populations** compared to the red curves (old test). This indicates improved discriminatory power. - The **reduced overlap** between the blue curves means fewer cases fall into the ambiguous zone - **Smaller FN (False Negative) region** → Higher sensitivity (better detection of diseased individuals) - **Smaller FP (False Positive) region** → Higher specificity (better identification of healthy individuals) - The **higher peak and clearer separation** demonstrate that the new test performs better in both correctly identifying disease when present AND correctly ruling out disease when absent *Incorrect: Higher sensitivity and lower specificity* - This would occur if the cutoff point was shifted to reduce false negatives at the cost of increasing false positives - The image shows BOTH FN and FP regions are smaller, not a trade-off *Incorrect: Lower sensitivity and higher specificity* - This would occur if the cutoff point was shifted to reduce false positives at the cost of increasing false negatives - Again, the image shows improvement in both parameters simultaneously *Incorrect: Lower sensitivity and lower specificity* - This represents a worse test with more overlap and larger FN/FP regions - The blue curves clearly show IMPROVEMENT, not deterioration, compared to the red curves

Q: Which type of statistical graph is shown in the image below?

Frequency polygon. ***Frequency polygon*** - A frequency polygon is constructed by plotting a **point at the midpoint of each class interval** at a height corresponding to its frequency, and then **connecting these points with straight lines**. - The image clearly shows points plotted and connected by straight lines, representing the frequency distribution, which is characteristic of a frequency polygon. *Histogram* - A histogram uses **contiguous bars to represent the frequency distribution** of continuous data, where the width of the bar represents the class interval and the height represents the frequency. - The image does not display bars, but rather a line graph connecting points, which differentiates it from a histogram. *Sector diagram* - A sector diagram, also known as a **pie chart**, divides a circle into sectors that represent proportions of a whole. - The image is a two-dimensional graph with x and y axes, not a circular representation of proportions. *Scatter diagram* - A scatter diagram displays **individual data points** for two variables to show their relationship, without connecting them with lines in a continuous manner like a frequency polygon. - While it shows points, they are connected to form a shape, indicating a distribution over intervals rather than individual data points of two distinct variables.

Q: The following scatter plot of 4 different samples shows the correlation between weight and height in the samples. All 4 samples have the same coefficient of correlation of 0.6 taken together, what will be the net correlation coefficient?

More than 0.6. ***More than 0.6*** - When multiple distinct samples (each with r = 0.6) are combined, the **overall correlation coefficient typically increases** beyond the individual correlations - This occurs because the **between-group variance** adds to the total variance when groups form separate clusters along the regression line - The combined dataset captures both the within-group correlation (0.6) and the systematic separation of groups, resulting in a stronger overall linear relationship - This is a well-recognized phenomenon in biostatistics: **pooled correlation > individual correlations when groups are spatially separated** *Less than 0.6* - This would occur only if the groups showed contradictory trends or if combining them obscured the linear relationship - Not applicable here since all groups show the same positive correlation and align along a common trend *Equal to 0.6* - This is incorrect because **the correlation coefficient of pooled data ≠ mean of individual correlations** - The pooled correlation is calculated from all combined data points, which includes both within-group and between-group variance - Mathematical property: when distinct clusters exist, pooling increases the correlation coefficient *Cannot be calculated* - The correlation coefficient **can be calculated** by pooling all data points and applying the standard Pearson correlation formula - Sufficient information exists to compute the combined correlation from the aggregated dataset

Q: What does the given image show?

Scatter diagram. ***Scatter diagram*** - A scatter diagram displays individual data points using **Cartesian coordinates**, where each point represents the values of two different variables, which are 'Weight' and 'Height' in this image. - It is used to observe the **relationship or correlation** between two quantitative variables. *Histogram* - A **histogram** is used to represent the **distribution of a single numerical variable**, divided into bins, and shows the frequency of data points falling into each bin. - This image clearly plots individual points for two variables, not frequency distributions within bins.

Q: What does the following diagram show?

***A line graph displaying average daily temperatures*** - The image presents a **line graph** with "Average daily temperatures" as its title. - The Y-axis is labeled "Temperature (°F)," indicating that the graph tracks temperature over a period, likely days.

Question 1

A study is conducted to compare the mean hemoglobin (Hb) levels between two independent groups. Which statistical test is most appropriate?

Accepted Answer

Unpaired t-test

Answer

Paired t-test

Answer

Chi-square test

Answer

ANOVA

Question 2

In a village of 100 children, 10 children have a past history of measles (i.e., they are not at risk now), 20 new cases of measles were reported this year. What is the incidence of measles in this population for the year?

Accepted Answer

22.22 %

Answer

10 %

Answer

30 %

Answer

20 %

Question 3

The given image shows:

Accepted Answer

Spaghetti plot

Answer

Kaplan Meier plot

Answer

Funnel plot

Answer

Forest plot

Question 4

The given image shows:

Accepted Answer

Funnel plot

Answer

Kaplan Meier plot

Answer

Spaghetti plot

Answer

Forest plot

Question 5

The given image shows:

Accepted Answer

Funnel plot

Answer

Kaplan Meier plot

Answer

Spaghetti plot

Answer

Forest plot

Question 6

A new diagnostic test is introduced. Based on the image provided showing performance curves of both tests, what characteristics does the new test demonstrate compared to the old one?

Accepted Answer

Higher sensitivity and higher specificity

Answer

Higher sensitivity and lower specificity

Answer

Lower sensitivity and higher specificity

Answer

Lower sensitivity and lower specificity

Question 7

Which type of statistical graph is shown in the image below?

Accepted Answer

Frequency polygon

Answer

Histogram

Answer

Sector diagram

Answer

Scatter diagram

Question 8

The following scatter plot of 4 different samples shows the correlation between weight and height in the samples. All 4 samples have the same coefficient of correlation of 0.6 taken together, what will be the net correlation coefficient?

Accepted Answer

More than 0.6

Answer

Less than 0.6

Answer

Equal to 0.6

Answer

Cannot be calculated

Question 9

What does the given image show?

Accepted Answer

Scatter diagram

Answer

Histogram

Answer

Bar chart

Answer

Pie chart

Question 10

What does the following diagram show?

Biostatistics — MCQs

Biostatistics — MCQs

On this page

Practice by Chapter

Want unlimited practice?