BIOSTATISTICS AND RESEARCH METHODOLOGY- 2mark

1. Report Writing in Research Study – A structured document presenting research

findings, including introduction, methodology, results, discussion, and
conclusion. It ensures clarity, reproducibility, and credibility of research.

2. Confidence Interval (CI) – A range of values, derived from sample data, that is
likely to contain the population parameter with a certain probability (e.g., 95%
CI). It provides an estimate of precision.

3. Chi-Square Test – A statistical test used to determine if there is a significant

association between categorical variables in a contingency table. It assesses
independence or goodness of fit.

4. Probability – A measure of how likely an event is to occur, ranging from 0

(impossible) to 1 (certain). Formula: P(A) = Favorable outcomes / Total
outcomes.

5. Standard Error of Mean (SEM) – A measure of how much the sample mean is
expected to vary from the true population mean. Formula: SEM = SD / √n (where
SD = standard deviation, n = sample size).

6. 2² and 2³ Designs –

o 2² Design: A factorial experiment with two factors, each having two levels.

o 2³ Design: A factorial experiment with three factors, each having two

levels. These designs help study interactions between factors.

7. Applications of Nonparametric Tests –

o Used when data do not meet normality assumptions.

o Applied in small sample sizes.

o Used for ordinal or categorical data.

o Examples: Wilcoxon test, Kruskal-Wallis test.

8. Degrees of Freedom (df) – The number of independent values that can vary in a
statistical calculation. Formula for df in t-test: df = n - 1 (where n = sample size).

9. Difference Between Standard Deviation (SD) and Standard Error of Mean

(SEM) –

o SD: Measures data dispersion within a dataset.

 o SEM: Estimates accuracy of the sample mean in relation to the
population mean.

10. Define Scatter Plots – A graphical representation of the relationship between

two numerical variables, showing how they correlate. Points are plotted on an X-
Y axis to visualize patterns or trends.

1. Multiple Regression – A statistical technique used to predict the value of a

dependent variable based on multiple independent variables. It helps in
understanding the relationship between variables and making predictions.

2. Pharmaceutical Examples for Data Analysis Using SPSS –

o Analyzing drug dissolution profiles.

o Studying patient compliance data.

o Identifying trends in adverse drug reactions.

o Conducting stability studies for pharmaceuticals.

3. Factorial Design – A statistical experimental design that studies multiple factors

simultaneously, each at different levels, to evaluate their individual and
interaction effects. Example: 2³ factorial design in drug formulation.

4. Power of a Study – The probability that a test correctly rejects a false null
hypothesis (1 - β). A higher power (typically ≥ 80%) means a lower chance of Type
II error (false negative).

5. Two Methods of Sample Size Calculation in Research Study –

o Cohen’s Method: Based on effect size, significance level, and power.

o Formula-Based Approach: Uses statistical formulas incorporating

variance and expected difference.

6. Degrees of Freedom (df) – The number of independent values that can vary in a
statistical calculation. Example: In a t-test, df = n - 1 (where n is sample size).

7. Standard Error of Mean (SEM) and Its Significance –

o SEM = SD / √n, where SD = standard deviation, n = sample size.

o Significance: Indicates how accurately a sample mean represents the

population mean. A lower SEM means more precise estimation.

8. One-Tailed and Two-Tailed Tests –

 o One-Tailed Test: Tests for an effect in only one direction (greater or
smaller).

o Two-Tailed Test: Tests for an effect in both directions (higher or lower

than expected).

9. Pharmaceutical Examples for Optimization Techniques –

o Formulation optimization: Using Response Surface Methodology (RSM)

for drug formulation.

o Process optimization: Optimizing tablet compression parameters.

10. Wilcoxon Rank Sum Test – A nonparametric test used to compare two
independent groups when data are not normally distributed. It evaluates
differences in medians rather than means.

1. Biostatistics – The application of statistical methods to biological and health

sciences, used for analyzing medical and pharmaceutical data, clinical trials,
and epidemiological studies.

2. Median Calculation – Arrange the data in ascending order:

26, 33, 35, 42, 44, 66, 68, 72, 80, 85

o Median = (44 + 66) / 2 = 55

3. SEM (Standard Error of the Mean) – Measures the accuracy of a sample mean in
estimating the population mean. Formula:

SEM=SDnSEM = \frac{SD}{\sqrt{n}}

where SD = standard deviation, n = sample size.

4. Need for Research –

o To discover new knowledge.

o To solve existing problems.

o To improve decision-making.

o To enhance efficiency in various fields.

5. Cohort Study – A type of observational study where a group (cohort) is followed

over time to assess the relationship between risk factors and disease outcomes.
Example: Studying smokers vs. non-smokers to observe lung cancer incidence.

6. Types of Hypothesis in Regression Modelling –

 o Null Hypothesis (H₀): No relationship exists between independent and
dependent variables.

o Alternative Hypothesis (H₁): A significant relationship exists between

variables.

7. R Online Statistical Software – A free, open-source software used for statistical

computing, data analysis, visualization, and machine learning. It supports
libraries like ggplot2, dplyr, and caret.

8. Mode Calculation – The mode is the value with the highest frequency. From the
given data:

o X: 0, 1, 2, 3, 4, 5

o F: 42, 55, 32, 22, 15, 6

o The highest frequency is 55 at X = 1, so Mode = 1.

9. Optimization Techniques in Response Surface Methodology (RSM) –

o Central Composite Design (CCD)

o Box-Behnken Design

o Taguchi Method

o Full Factorial Design

o Sequential Quadratic Programming

10. Applications of Factorial Design –

• Drug formulation and development.

• Process optimization in pharmaceuticals.

• Studying drug-excipient interactions.

• Optimizing clinical trial designs.

1. Probability – The measure of the likelihood of an event occurring, ranging from 0

(impossible) to 1 (certain). It is calculated as:

P(A)=Favorable OutcomesTotal OutcomesP(A) = \frac{\text{Favorable

Outcomes}}{\text{Total Outcomes}}

2. Null Hypothesis (H₀) and Alternative Hypothesis (H₁) –

 o H₀ (Null Hypothesis): Assumes no effect or relationship exists. Example:
"There is no difference in drug efficacy."

o H₁ (Alternative Hypothesis): Assumes a significant effect or relationship.

Example: "The drug improves patient recovery rates."

3. Wilcoxon Rank Sum Test – A non-parametric test used to compare two

independent groups when the data is not normally distributed. It ranks values
from both groups and tests if their distributions differ.

4. Response Surface Plot – A 3D graphical representation of how response

variables change with variations in two or more independent variables in
optimization studies. Used in Response Surface Methodology (RSM).

5. F-Statistics – Used in ANOVA (Analysis of Variance) to compare variances

among groups. It is calculated as:

F=Variance between groupsVariance within groupsF = \frac{\text{Variance between

groups}}{\text{Variance within groups}}

A high F-value indicates significant differences between groups.

6. Excel – A spreadsheet software used for data analysis, statistical calculations,

visualization, and automation with built-in formulas like AVERAGE, STDEV,
T.TEST, REGRESSION in Data Analysis ToolPak.

7. Range Calculation – The range is the difference between the maximum and
minimum values in a dataset.
Given data: 188, 178, 173, 164, 172, 183, 184, 185, 211, 217, 232, 240

o Range = Max – Min = 240 – 164 = 76

8. Central Composite Design (CCD) – A design of experiments (DOE) technique

used in Response Surface Methodology (RSM) to model and optimize
responses. It consists of:

o Factorial points

o Axial points

o Center points

9. Advantage of Factorial Design –

o Identifies interactions between factors.

o More efficient than one-factor-at-a-time experiments.

o Reduces the number of trials while maximizing information gain.

 o Useful for process optimization in pharmaceuticals and manufacturing.

1. Standard Error of Mean (SEM) – Measures the variability of sample means

around the population mean. It is calculated as:

SEM=σnSEM = \frac{\sigma}{\sqrt{n}}

where σ is the standard deviation and n is the sample size. A smaller SEM indicates
more precise estimates.

2. Central Composite Design (CCD) – A Response Surface Methodology (RSM)

technique used for optimization. It includes:

o Factorial points

o Axial points

o Center points
It helps model quadratic effects in experiments.

3. Semi-logarithmic Plots – Graphs where one axis (x or y) is on a logarithmic

scale and the other is on a linear scale. Used to represent exponential
relationships, such as drug concentration vs. time in pharmacokinetics.

4. Histogram – A bar chart representing the frequency distribution of numerical

data. The x-axis represents intervals (bins), and the y-axis represents frequency.
Used to visualize data distribution.

5. Types of Correlation –

o Positive Correlation (both variables increase together)

o Negative Correlation (one variable increases, the other decreases)

o No Correlation (no relationship between variables)

o Linear & Non-linear Correlation

6. Classification of Observational and Experimental Studies –

o Observational Studies (No intervention, only observation):

▪ Cross-sectional

▪ Case-control

▪ Cohort studies

o Experimental Studies (Controlled intervention applied):

 ▪ Randomized controlled trials (RCTs)

▪ Non-randomized trials

7. Coefficient of Variation (CV) – Measures relative variability as a percentage.

CV=Standard DeviationMean×100CV = \frac{\text{Standard Deviation}}{\text{Mean}}

\times 100

Used for comparing variability across different datasets.

8. Type I and Type II Errors in Hypothesis Testing –

o Type I Error (False Positive): Rejecting H₀ when it's true (α-error).

o Type II Error (False Negative): Failing to reject H₀ when it's false (β-error).

9. Degrees of Freedom (df) – Number of independent values in a dataset that can

vary while estimating parameters.
Formula for a single sample t-test:

df=n−1df = n - 1

10. Relationship Between Sample Size and Power of the Study –

• Larger sample size → Higher power (more ability to detect true effects).

• Power depends on effect size, significance level (α), and variability.

• Higher power reduces Type II errors (β).

1. Latin Square Technique

• A statistical design used to control two sources of variability in an experiment

(e.g., row and column effects).

• The arrangement ensures that each treatment appears only once per row and
per column.

• Used in agriculture, pharmaceuticals, and industrial experiments.

2. Prospective vs. Retrospective Studies

• Prospective Study: Data is collected forward in time from a defined starting

point. Example: Clinical trials.

• Retrospective Study: Data is collected from past records. Example: Case-

control studies in epidemiology.
3. Applications of SAS (Statistical Analysis System)

• Data management and preprocessing.

• Statistical analysis (e.g., regression, hypothesis testing).

• Machine learning and predictive modeling.

• Clinical trials and pharmaceutical research.

4. Power of a Study

• The probability of correctly rejecting the null hypothesis (H₀) when it is false.

• Higher power reduces Type II errors (β-error).

• Formula: Power=1−β\text{Power} = 1 - \beta

5. Report Writing in Research Study

• Abstract: Summary of findings.

• Introduction: Background and objectives.

• Methodology: Data collection and analysis methods.

• Results & Discussion: Findings and interpretation.

• Conclusion: Summary and recommendations.

• References: Cited sources.

6. Formulas for Given Tests

a) Kruskal-Wallis Test (Non-parametric ANOVA alternative):

H=12N(N+1)∑Ri2ni−3(N+1)H = \frac{12}{N(N+1)} \sum \frac{R_i^2}{n_i} - 3(N+1)
where R_i = sum of ranks for group i, n_i = sample size of group i, N = total sample size.

b) Friedman Test (Non-parametric test for repeated measures):

χF2=12nk(k+1)∑Rj2−3n(k+1)\chi^2_F = \frac{12}{nk(k+1)} \sum R_j^2 - 3n(k+1)
where R_j = sum of ranks for each treatment, n = number of subjects, k = number of
treatments.

7. Sample vs. Population Parameter

• Sample: A subset of the population (e.g., 100 patients in a city).

• Population Parameter: A characteristic of the entire population (e.g., all

patients in a country).

• Sample statistics (mean, variance) estimate population parameters.

8. Histogram and Its Advantage

 • Histogram is a graphical representation of the frequency distribution of
numerical data.

• Advantage: Helps visualize data distribution (normal, skewed, bimodal, etc.).

Typical Histogram Example:

| █

| ███

| ██████

| ██████████

|██████████████

|_________________

10 20 30 40

9. Steps in Formulating an Optimal Design

• Define objective and constraints.

• Select design type (e.g., full factorial, CCD).

• Identify key variables and responses.

• Perform statistical analysis (e.g., ANOVA, regression).

• Validate and optimize the model.

10. Nominal vs. Ordinal Data

• Nominal Data: Categorical with no order (e.g., Gender: Male/Female).

• Ordinal Data: Categorical with a meaningful order (e.g., Pain level: Mild,
Moderate, Severe).

1. Applications of Nonparametric Tests

• Used when data does not follow a normal distribution.

• Suitable for ordinal or categorical data.

• Examples:

o Mann-Whitney U Test: Compares two independent groups.

o Wilcoxon Signed-Rank Test: Compares paired data.

o Kruskal-Wallis Test: Alternative to one-way ANOVA.

2. Define Sample Size

 • The number of observations or units selected from a population for analysis.

• Larger sample size = higher accuracy and lower standard error.

3. When is Median More Important than Mean?

• When data has outliers (e.g., income distribution).

• When the data is skewed (e.g., house prices).

• Median better represents central tendency in non-symmetric distributions.

4. Features of Normal Distribution Pattern

• Symmetrical bell-shaped curve.

• Mean = Median = Mode.

• 68-95-99.7 Rule:

o 68% of data within 1 standard deviation.

o 95% within 2 SDs.

o 99.7% within 3 SDs.

5. Optimization Techniques

• Gradient Descent

• Genetic Algorithms

• Simulated Annealing

• Response Surface Methodology (RSM)

• Linear & Nonlinear Programming

6. Define Standard Error of Mean (SEM)

• Measures the accuracy of a sample mean compared to the population mean.

• Formula: SEM=σnSEM = \frac{\sigma}{\sqrt{n}} where σ = standard deviation, n =

sample size.

7. One-Tailed vs. Two-Tailed Tests

• One-tailed test: Tests a specific direction (e.g., mean is greater than X).

• Two-tailed test: Tests for any difference (e.g., mean is different from X).

• One-tailed has more power, but two-tailed is more conservative.

8. What is Regression?
 • A statistical method to determine the relationship between independent and
dependent variables.

• Types:

o Linear Regression (Y = mX + C).

o Multiple Regression (more than one independent variable).

o Logistic Regression (for categorical outcomes).

9. Descriptive vs. Inferential Statistics

• Descriptive Statistics: Summarizes data (mean, median, SD, charts).

• Inferential Statistics: Makes predictions (hypothesis testing, confidence

intervals).

10. Define Mann-Whitney U Test

• A nonparametric test for comparing two independent groups.

• Alternative to the t-test when data is not normally distributed.

• Used in small sample sizes or ordinal data.