1Introduction

Statistics plays a vital role in pharmaceutical sciences by providing methods for collecting, organizing, analyzing, and interpreting data. It supports evidence -based decision-making in areas such as drug development, clinical research, quality control, ph armacovigilance, and patient care. Among the various branches of statistics, descriptive statistics serves as the foundation for understanding and summarizing data in a clear and meaningful manner.

Pharmaceutical and biological data often exhibit variabili ty due to differences among individuals and experimental conditions. Descriptive statistical techniques help researchers summarize datasets, identify patterns, assess variability, and understand data distributions before applying advanced analytical methods. The effective use of these techniques improves the accuracy and reliability of research findings and healthcare decisions.

In pharmaceutical practice, data may be classified as nominal, ordinal, interval, or ratio, depending on their characteristics and measurement scales. Understanding these data types is essential for selecting appropriate statistical methods. Common sources of pharmaceutical data include clinical trials, pharmacokinetic studies, pharmacovigilance reports, and quality assurance investigations.

Key descriptive measures such as mean, median, mode, range, variance, and standard deviation are used to describe central tendency and dispersion, while skewness helps assess the shape and distribution of data. With the growing use of computational tools, Python libraries such as NumPy and Pandas have become valuable resources for efficient data analysis and interpretation. Therefore, a sound understanding of descriptive statistics and its application using Python is essential for modern pharmaceutical research and practice.

1.1Descriptive Statistics

Descriptive statistics are methods used to summarize, organize, and present data meaningfully. In pharmaceutical sciences, they help researchers understand clinical outcomes, drug performance, laborato ry measurements, and patient responses before applying advanced statistical tests.

1.1.1Types of data in pharmaceutical sciences

1.1.1.1 Qualitative (Categorical) Data Qualitative data represent attributes or categories that cannot be measured numerically. They are further classified as: Nominal Data: Nominal data, also known as categorical data, is a type of data used in statistics to label variables without providing any quantitative value. The key characteristic of nominal data is that it categorizes data into distinct groups that do not have a specific order or ranking.

This means that while the data can be divided into categories, these categories cannot be meaningfully arranged in a hierarchical order. Some examples of nominal data are: Ordinal Data: Ordinal data is a form of categorical data that has a meanin gful order among its categories. But, it lacks any numerical value or a fixed interval that can separate them from each other.

In simpler terms, ordinal data represents variables that can be ranked or ordered, but the precise difference between the ranks is not known. This form of data is frequently used with surveys and questionnaires to collect responses that involve subjective judgments or preferences. A few examples of such cases are: Applications of Ordinal Data agreement levels, such as: "Strongly Disagree, Disagree, Neutral, Agree, Strongly Agree". proficiency levels (e.g., Beginner, Intermediate, Advanced). "Mild," "Moderate," "Severe").

• Gender: Male, Female, Non-binary, Other
• Marital Status: Single, Married, Divorced, Widowed
• Nationality: American, Canadian, Indian, Chinese, Australian, etc.
• Blood Type: A, B, AB, O
• Capturing the preferences or the ranking
• Educational Assessments
• Survey and Questionnaires (Likert Scales): Widely used to gauge opinions or
• Education and Performance Evaluation: Grading systems (e.g., A, B, C, D, F) or
• Healthcare and Medicine: Assessing patient conditions or pain intensity (e.g., "None,"

1.1.2Quantitative (Numerical) Data

Quantitative data result from counting or measurement and are represented numerically: Discrete Data: Discrete data refers to specify and distinct values or observations that can be counted. Unlike continuous data, which can take any value within a range, discrete data can only take particular values. In pharmaceutical science, discrete data are commonly obtained when counting patients, tablets, adverse drug reactions, microbial colonies, or prescription errors.

Examples: 1. Number of students in a classroom: Students can be in dividually counted, but you can’t have a fraction of a student. 2. Number of attendees at a conference: Attendees are countable entities; there isn’t a scenario where half a person attends. 3.

Types of vehicles in a parking lot: Vehicles are distinctly categoriz ed, for instance, you can’t have a vehicle that’s simultaneously a car and a motorcycle. Continuous Data: A continuous data is a type of measurement where values can fall anywhere along a range, including fractions and decimals, with no gaps between possible scores. Between any two points on a continuous scale, there are infinite possible values.

Measurable values on an uninterrupted scale. Examples: Table: 1.1.3 .Types Of Data Type of Data Definition Nature of Data Can Be Ordered? Mathematical Operations Possible?

Pharmaceutical Example Nominal Data Data classified into categories or names without any order Qualitative No No meaningful calculations Blood group (A, B, AB, O), gender, dosage form (tablet, capsule, syrup) Ordinal Data Categorical data with a meaningful order or ranking Qualitative Yes Limited calculations Pain severity (mild, moderate, severe), patient satisfaction scale Discrete Data Numerical data that can take only specific whole- number values Quantitative Yes Arithmetic operations possible Number of tablets dispensed, number of patients in a trial Continuous Data Numerical data that can take any value within a range, including Quantitative Yes Full arithmetic operations possible Body weight, blood pressure, drug concentration in plasma decimals

• Weight, height, or length
• The temperature in Kelvin (since zero Kelvin means zero heat)
• Length of time/duration (e.g. seconds, minutes, hours)

2Sources of data in pharmacy

Data in pharmacy research and practice are collected from different sources to support drug discovery, formulation development, clinical research, pharmacovigilance, hospital pharmacy, and public health studies. Pharmaceutical data can originate from sever al sources depending on the purpose of the study or healthcare activity. These sources are broadly classified into Primary Sources and Secondary Sources.

2.1Clinical trials

Clinical trials are systematic research studies conducted on human participants to evaluate the safety, efficacy, dosage, and side effects of drugs, vaccines, medical devices, or treatment methods before they are approved for public use. Clinical trials are carefully designed, reviewed and completed, and need to be approved before they can start. People of all ages can take part in clinical trials, including children.

2.1.1Objectives of Clinical Trials

• To determine whether a drug is safe
• To evaluate therapeutic effectiveness
• To identify adverse effects
• To compare new treatment with existing therapy
• To establish appropriate dosage regimen

2.1.2Importance of Clinical Trials

There are 4 phases of biomedical clinical trials: evaluate a safe dosage range and identify side effects. a larger group of human subjects to monitor for any adverse effects. countries, and are often the step right before a new treatment is approved. in a wide population over a longer timeframe. Table: 2.1.3.Phases of Clinical trials

• Ensure patient safety
• Provide scientific evidence for drug approval
• Improve treatment outcomes
• Support evidence-based medicine
• Phase I studies usually test new drugs for the first time in a small group of people to
• Phase II studies test treatments that have been found to be safe in phase I but now need
• Phase III studies are conducted on larger populations and in different regions and
• Phase IV studies take place after country approval and there is a need for further testing

2.2Pharmacovigilance

Pharmacovigilance is the science and activities related to the detection, assessment, understanding, and prevention of adverse effects or any other drug -related problems. It plays an important role in ensuring the safety and effective use of medicines duri ng both clinical trials and post -marketing periods. Pharmacovigilance helps healthcare professionals and regulatory authorities identify adverse drug reactions (ADRs), monitor drug safety, and take appropriate actions to minimize risks associated with medicines.

2.2.1Characteristics of Pharmacovigilance

o Drug Safety Monitoring o Detection of Adverse Drug Reactions (ADRs) o Risk Assessment o Prevention of Drug-Related Problems o Post-Marketing Surveillance o Collection and Analysis of Data o Improves Patient Safety o Regulatory Decision Support o Multidisciplinary Approach o Continuous and Systematic Process

2.3Quality control

Quality control (QC) is a part of pharmaceutical quality assurance that focuses on testing and verifying the quality, safety, purity, and effectiveness of pharmaceutical products. It involves a series of procedures and analytical tests performed on raw materials, in -process materials, and finished products to ensure that they meet predefined standards and specifications. Quality control helps in mainta ining consistency in pharmaceutical manufacturing and ensures that medicines are safe for patient use It involves testing and evaluating raw materials, manufacturing processes, and finished products according to pharmacopeial standards.

2.3.1Objectives of Quality Control

• To ensure the quality and safety of medicines
• To maintain batch-to-batch consistency
• To detect contamination or defects
• To ensure compliance with pharmacopeial standards
• To provide effective and stable pharmaceutical products

2.3.2Important Quality Control Tests

Tablets Capsules Syrups and Solutions Sterile Products

• Weight variation test
• Hardness test
• Friability test
• Disintegration test
• Dissolution test
• Uniformity of weight
• Dissolution test
• Moisture content
• pH determination
• Viscosity
• Clarity test
• Sterility test
• Pyrogen test
• Particulate matter test

2.3.3Characteristics of Quality Control in Pharmacy

1Product Testing - QC involves testing raw materials, intermediate products, and finished

products. Examples:

• Tablet hardness test
• Assay test
• Sterility testing

2Standardization- Products are evaluated according to official standards such as

• IP (Indian Pharmacopoeia)
• BP (British Pharmacopoeia)
• USP (United States Pharmacopeia)

3Accuracy and Precision -Analytical methods used in QC must provide accurate and

reproducible results.

4Detection of Defects

QC helps identify: 5.Documentation-All QC procedures and test results are properly recorded and maintained 6.Compliance with GMP-Quality control follows Good Manufacturing Practices (GMP) to ensure safe production 7.Stability Monitoring-QC checks the stability of drugs during storage and determines shelf life. 8. Use of Analytical Techniques-Different analytical instruments are used for testing. For eg., HPLC,UV Spectrophotometer ,Dissolution apparatus,pH meter

• Contamination
• Impurities
• Incorrect labeling
• Manufacturing errors

2.4Pharmacokinetic (PK) Studies

Pharmacokinetic (PK) studies are scientific investigations that evaluate how a drug moves through the body over a period of time. These studies focus on the processes of absorption, distribution, metabolism, and excretion (ADME) of drugs. Pharmacokinetic studies help researchers understand the concentration of a drug in the bloodstream and tissues a fter administration and determine how long the drug remains active in the body.

Key Processes in PK Studies (ADME) site. Factors such as chemical properties, formulation, and route of administration infl uence absorption speed and completeness. rotein binding, solubility, and tissue permeability affect distribution. tes. Enzymes like cytochrome P450 play a major role, and some drugs (prodrugs) bec

• Absorption: How a drug enters systemic circulation from its administration
• Distribution: The movement of the drug from the bloodstream to tissues and organs. P
• Metabolism: Chemical transformation of the drug, primarily in the liver, into metaboli

ome active only after metabolism.

but also through bile, lungs, or sweat. Excretion rates influence drug half- life and accumulation risk.

• Excretion: Removal of the drug and its metabolites, mainly via the kidneys,

3Measures of Central Tendency

Measures of central tendency are statistical methods used to determine the central or average value of a dataset. They help summarize large amounts of data into a single representative value. The three main measures are: 3.1 Mean: Mean is the average of all the values given in a set of data.

There are three types of mean such as Arithmetic mean, Geometric mean and Harmonic mean. 3.1.1 Arithmetic Mean: Arithmetic mean of a set of observation is defined as the sum of all observations divided by the total number of observations and it is denoted by xˉ. Direct method and short cut method are used for calculating arithmetic mean. Shortcut method is preferred when the number of observations is large.

Mean xˉ= Sum of all values Total number of values 3.1.1.1 Ungrouped Data (Individual series): For ungrouped data, the mean is found by adding all individual values and dividing by the total number of observations. For example, if x1, x2, x3,…xn be the values of n observations then arithmetic mean is given as, xˉ= x1+x2+x3+…xn n By Direct Method, the mean is calculated as follows: xˉ= ∑xi n Where, n is total number of observations, ∑xi is sum of all observations By Short cut method the mean is calculated as follows: xˉ=a+ ∑d n Where, xˉ = Mean, a= Assumed mean, d= Deviation(x-a) and n= number of observations. Example 1: Find the arithmetic mean for the daily study hours of 5 students: 5, 7, 8, 4, and 6 by using direct method.

Solution: Here, total number of students (n) =5 Total study hours of 5 students (∑x) = 5+7+8+4+6 = 30 Thus, Arithmetic Mean xˉ= ∑x n =

• Mean - Sum of all observations divided by the total number of observations.
• Median- The middle or central value in an ordered set.
• Mode - The most frequently occurring value in a data set.

5  =6

The arithmetic mean of the given individual series is 6 hours. Example 2: Calculate the arithmetic mean of marks obtained by 5 students out of 100 by using short cut method

Students 1 2 3 4 5
Marks(x) 35  40  60 75 90

Solution: Here, consider assumed mean ‘a’ = 60 from the above table. Then calculate the deviation from assumed mean d = x – a.

Students 1 2 3 4 5
Marks(x) 35  40  60 75 90
d = x-a -25 -20 0 15 30 ∑d  = 0

Here, number of students (n) = 5, ∑d = 0 and a = 60 xˉ=a+ ∑d n

= 60+
5 = 60+0 = 60

Thus the arithmetic mean of mark is 6 3.1.1.2 Ungrouped frequency distribution (Discrete series): For a discrete frequency distribution with values x1, x2, x3,…xn and frequencies f1, f2, f3,…fn then, the mean is calculated using the formula is, xˉ=

f1x1+f2x2+f3x3+…fnxn

f1+f2+f3+…fn By Direct Method, the mean is calculated as follows: xˉ= ∑fixi ∑fi By Short cut method the mean is calculated as follows: xˉ=a+ ∑fd ∑f where, d = x-a Example: Find the mean of the following data for discrete series by using direct method and shortcut method. Solution: By direct method, the mean is calculated as follows: Here frequency (f) is multiply with the variable (x).

x 5 10 15 20 25
f 5 2 2 3 4 ∑f =16
x 5 10 15 20 25
f 5 2 2 3 4
fx 25 20 30 60 100 ∑fx = 235

Here, ∑f =16 and ∑fx = 235 xˉ = ∑fx ∑f =

16   = 14.68

Therefore, the mean is 14.68 3.1.1.3 Grouped frequency distribution (Continuous series): Similarly for a continuous distribution, we take the midpoints of each class. If m1, m2, m3….mn are the midpoints of the class with frequencies f1, f2, f3,…fn then xˉ =

f1m1+f2m2+f3m3+⋯fnmn

f1+f2+f3+⋯fn By Direct Method, the mean is calculated as follows: xˉ= ∑fimi ∑fi By Short cut method the mean is calculated as follows: xˉ=a+ ∑fd ∑f where, d = x-a Example: Calculate the mean of the following data using Direct Method and Short -Cut Method:

Age(in years) 20-30 30-40 40-50 50-60 60-70
No. of workers 8 15 12 9 6

Solution: By direct method, the mean is calculated as follows: Hence Mid -point of ‘x’ is calculated and this mid-point is multiplied by ‘f’ to calculate ‘fm’.

Age (x) 20-30 30-40 40-50 50-60 60-70
Workers (f) 8 15 12 9 6 ∑f = 50
Mid-point (m) 25 35 45 55 65
fm 200 525 540 495 390 ∑fm=2150

Here, ∑f = 50 and ∑fm=2150 xˉ= ∑fm ∑f =

50  = 43

Therefore, the mean is 43. By Short cut method the mean is calculated as follows: Here, consider assumed mean ‘a’ =45. Here, ∑f = 50 and ∑fd = -100 xˉ = a+ ∑fd ∑f = 45 +

-100
= 45-2 = 43

Therefore, the mean is 43.

3.1.2Geometric Mean

Geometric mean of a set of n observations is defined as the nth root of their product. Geometric Mean (G.M) =(a₁ × a₂ × … × aₙ)1/n (i. e) In geometric mean, we first multiply the given number altogether and then take the nth root of the given product. Example : Find the geometric mean of 2, 6, 9, 5, 12.

Solution: Given data values: 2, 6, 9, 5, and 12 We know that the formula to find the geometric mean is GM = (a₁ × a₂ × … × aₙ)1/n Now, substitute the values in the formula, GM = (2 × 6 × 9 × 5 × 12)1/5 = (6480)1/5

Age (x) 20-30 30-40 40-50 50-60 60-70
Workers (f) 8 15 12 9 6 ∑f = 50
Mid-point (m) 25 35 45 55 65
d = m - a -20 -10 0 10 20
fd -160 -150 0 90 120 ∑fd = -100

Thus, the 5th root of 6480 is 5.785. Therefore, the geometric mean of 2, 6, 9, 5 and 12 is 5.79 .

3.1.3Harmonic Mean

Harmonic mean is defined as the reciprocal of the arithmetic mean of the reciprocals of the given items, Harmonic Mean = 𝑛

𝑥1
+ 1
𝑥2

+ .... + 1 𝑥𝑛 Example: Find the harmonic mean for data 2, 5, 7, and 9. Solution: Finding the reciprocal of 2, 5, 7, 9. (i.e) ½ = 0.5, ⅕ = 0.2, 1/7 = 0.14 1/9 = 0.11 Harmonic Mean = 𝑛

𝑥1
+ 1
𝑥2

+ .... + 1 𝑥𝑛 Here n=4 and calculate the average of the reciprocal values.

Average = (0.5 + 0.2 + 0.143 + 0.11)/4
= 0.953/4

Take the reciprocal of the average value. Harmonic Mean = 1/ Average

= 4/0.953
= 4.19

Hence, the harmonic mean for the data 2, 5, 7, 9 is 4.19.

3.2Median

Median is the positional average and is defined as the size of the item which lies at the centre when all items are arranged in either ascending or descending order of their magnitude. If the number of observations or data in the series is odd the median is calculated as follows. Median= (

n+1
2 )

th term If the number of observations or data in the series is even the median is calculated as follows. Median= (n

2)

th term+(n

2+1)

th term If the distribution is continuous then median could be defined as, Median=L+ ( N 2-CF f ) ×h Where L=Lower limit of the median class CF= Cumulative frequency of the class preceding to the median class f = frequency of the median class h = interval of the median class Example 1: Find the median of the given data set 60, 70, 10, 30, and 50. Solution: Order the given data in ascending order as: 10, 30, 50, 60 and 70. Here n = 5, the number of observations is odd the median is calculated as, Median= (

n+1
2 )

th term = (

5+1
2 )

th term = =3rdterm

=50

Thus the median is 50. Example 2: Find the median for the data 8, 5, 7, 10, 15, 21. Solution: Arranging the given data in ascending order, we get, 5, 7, 8, 10, 15, 21.

Here, the number of observations is 6, which is even. Median= (n

2)

th term+(n

2+1)

th term =

(6
2)

th term+(6

2+1)

th term = 3rdterm+4thterm Here, 3rdterm = 8 and 4thterm = 10 Therefore, median =

8+10

2 = 9 Hence, the median of the given data is 9. Example 3: Find the median number of hours studied per week . The following table shows the distribution of the number of hours spent studying per week by a group of students: Hours Studied (Per week) 0 - 5 5 - 10 10 - 15 15 - 20 20 - 25

Frequency 8 15 25 12 10

Solution: For finding the Median we have to build a table with cumulative frequency as, Now using the formula, Median=L+ ( N 2-CF f ) ×h Where , N = ∑f =70, L=10, f=25, h=5, cf=23 Median=10+ (

2 -23
25 ) ×5
= 10+ (
35-23
25 ) ×5
= 10+2.4 = 12.4

Thus, the median number of hours per week is 12.4 hours. 3.3 Mode: The mode is that value in a series whi ch occurs with the highest frequency. Mode represents the value occurring most often in the data. 3.3.1 Individual observations: In case of individual observations count the repetitions of each and every item. 3.3.2 Discrete series: The varible takes values x1, x2, x3….xn With frequencies f1, f2, f3,…fn respectively. The observation with highest frequency is the mode of distribution. 3.3.3 Continuous frequency distribution: The class with maximum frequency is called modal class and value of mode belongs to this class.

Mode is determined by the following formula, Mode = l+ (

f1-f0
2f1-f0-f2

) ×c where, l = Lower limit of the modal class f1 = frequency of the modal class f0 = frequency of the previous class f2 = frequency of the next class C = class-interval of the modal class Example 1: Find the mode of following values 120, 121, 123, 122, 125, 124, 120, 121, 122,

123, 125, 124, 122, 120, 122.

Solution: Arranging the given values in the form of frequency table. x 0 - 5 5 - 10 10 - 15 15 - 20 20 - 25

f 8 15 25 12 10

CF 0 + 8 = 8 8 + 15 = 23 23 + 25 = 48 48 + 12 = 60 60 + 10 = 70

x 120 121 122 123 124 125
f 3 2 4 2 2 2

Here maximum frequency is 4. Hence, corresponding value of variant is 122. Thus, mode value = 122.

Example 2: Find the mode of following values

x 100 120 140 160 180 200
f 15 18 25 20 17 14

Solution: The maximum frequency is 25, which corresponds to the value is 140. Thus, mode value = 140. Example 3: Compute the mode from the following data.

x 2-6 6-10 10-14 14-18 18-22 22-26 26-30
f 1 9 21 47 52 36 19

Solution: Here the highest frequency is 52. Hence, mode of frequency is 52 and modal class

is 18-22.

(i.e) l=18, f1=52, f0=47, f2 = 36, and c=4, we can calculate the mode as Mode = l+ (

f1-f0
2f1-f0-f2

) ×c

= 18+ (
52-47
2×52-47-36) ×4
= 18.95

4. A measures of dispersion: A measure of dispersion is a statistical metric that quantifies how spread out or scattered data points are around an average. 4.1. Absolute Measures Of Dispersion: These quantify dispersion in the original units of the data: Mean Deviation: The average of absolute deviations from a central measure.

Range: The difference between the highest and lowest values. Variance: The average of the squared deviations from the mean. Standard Deviation: The square root of the variance, showing dispersion in the same units as the data. 4.2.

Relative Measures Of Dispersion: We use relative measures of dispersion to measure the two qua ntities that have different units to get a better idea of the scattering of the data. Here are some of the relative measures of dispersion: Coefficient of Range: It is defined as the ratio of the difference between the highest and lowest value in a data set to the sum of the highest and lowest value. Coefficient of Variation: It is defined as the ratio of the standard deviation to the mean of the data set.

We use percentages to express the coefficient of variation. Coefficient of Mean Deviation: It is defi ned as the ratio of the mean deviation to the value of the central point of the data set. 4.2.1 Mean Deviation (M.D): Individual observation: It is defined as the arithmetic mean of the absolute deviation of the individual observation from either arithmetic mean or median of the given set of observations. i.e. M.D= ∑|x-x| n OR M.D= ∑|x-median| n Grouped data: Mean deviation of grouped data is given as M.D= ∑ f|x-x| ∑ f OR 𝑀. 𝐷 = ∑ 𝑓|𝑥−𝑚𝑒𝑑𝑖𝑎𝑛| ∑ 𝑓 Relative Measure: Relative measure with respect to mean deviation is the coefficient of mean deviation and it is given as Coefficient of M.D = M.D Median OR Coefficient of M.D = M.D Mean Example 1: Find the mean deviation and coefficient of M.D mean from of following data:

2, 4, 6, 8, 10.

Solution: First calculate mean because we have to find mean deviation from mean. Mean x = ∑ x n =

2+4+6+8+10

=

5 =6

Now prepare table for calculation of |x-x |

x 2 4 6 8 10
|x-x | 4 2 0 2 4

M.D= ∑|x-x| n =

5 =2.4

Coefficient of M.D = M.D Mean =

2.4
6  = 0.4

Example 2: Calculate mean deviation from mean and median for the internal marks of 5 students out of 10 marks. Marks: 2, 6, 8, 4, 4 Solution: Mean x = ∑ x n =

5 = 4.8

Median= (

n+1
2 )

th term = (

5+1
2 )

th term = 3rdterm Arranged in ascending order: 2, 4, 4, 6, 8 Thus the median is 4.

x 2 6 8 5 4
|x-x | 2.8 1.2 3.2 0.2 0.8

|x-median| 2 2 4 1 0 Thus, ∑|x-x| =8.2 , ∑|x-median| = 9 . M.D = ∑|x-x| n =

8.2
5 = 1.64

M.D= ∑|x-median| n =

5= 1.8

4.2.2Range

Range is the difference between the highest value and lowest value of the given set of observation (i.e.) Range R= H-L Where, R is the range of observations, H is the highest value of observation and L is lowest value of the observation. Coefficient of Range: Coefficient of range is another meas ure of variation which is the ratio between the difference of the extreme values and the sum of the extreme values (i.e) Coefficient of Range = H-L H+L Example 1: Find out the range for the following observations 20, 42, 13, 71, 54, 93, 15, 16 . Solution: Given, Largest Value of Observation = 93 and Smallest Value of Observation =

13.

Thus, the range of the data set is,

Range = 93 - 13

Range = 80 Example 2: Find Range and coefficient of range.

Students 1 2 3 4 5 6 7
marks 50 60 65 75 85 45 68

Solution: Here, H=85, L=45 Range = H – L

= 85 – 45 = 40

Coefficient of Range = H-L H+L =

85-45
85+45 =
130  = 0.3076

Example: Find out the range for the following frequency distribution table for the marks scored by class 10 students.

Marks Intervals  10-20 20-30 30-40

Number of Students 8 25 9 Solution: Given, Largest Value: Take the Higher Limit of the Highest Class = 40 Smallest Value: Take the Lower Limit of the Lowest Class = 10

Range = 40 - 10

Range = 30 Thus, the range of the data set is 30. 4.2.3. Variance: The variance is a measure of how far a set of data are dispersed out from their mean or average value. It is denoted as ‘𝜎2’.

To calculate variance, you square the difference of each data point from the mean to ensure negative and positive deviations don't cancel each other out. 𝜎2 = ∑(𝑥𝑖 − x)2 𝑁 𝜎2 is population variance, x̄ is the mean of the population dataset, N is the total number of observations Example: A company tracks the ages of 5 students: 30, 36, 40, 44, 50. Find the variance. Solution: To f ind the Mean x̄ = ∑ x 𝑁 =

30 + 36 + 40 + 44 + 50

=

5  = 40

Subtract the Mean and Square the Result

(30 − 40)2 = (−10)2 = 100
(36 − 40)2 = (−4)2= 16
(40 − 40)2 = (0)2    = 0
(44 − 40)2 =(4)2     = 16
(50 − 40)2 = (10)2  = 100

Sum the Squared Differences

100 + 16 + 0 + 16 + 100 = 232

Divide by the Number of Data Points N 𝜎2 =

5  = 46.4

4.2.4. Standard Deviation: Standard deviation is widely used measure of variation. It is also called as root mean square deviation and denoted Greek letter (sigma). 4.2.4.1.

Individual Series: Standard deviation is calculated by using mean and assumed mean value for individual series 𝜎 = √∑(𝑥−𝑥 )2 𝑛 where 𝑥 =mean σ = √∑𝑑2 𝑛 − ( ∑d 𝑛 ) where d=x-A and A=Assumed mean. 4.2.4.2 Discrete Series: Standard deviation is calculated by using mean and assumed mean value for discrete series σ = √∑f(x-x )2 ∑ f where 𝑥 =mean σ = √∑𝑓𝑑2 ∑ f − ( ∑𝑓d ∑ f ) where d=x-A and A=Assumed mean. 4.2.4.3 Continuous Series: Standard deviation is calculated by using mean and assumed mean value for continuous series σ = √∑fd2 ∑ f - ( ∑fd ∑ f) ×c where d= x-x c σ = √∑fd2 ∑ f - ( ∑fd ∑ f) ×c where d= x-A c Example 1: Calculate mean, Standard deviation and coefficient of variation for the following data 2, 3, 5, 4, 8, 1 Solution: Mean x = ∑ x n =

6 = 3.8
x 2 3 5 4 8 1
x-x -1.8 -0.8 1.2 0.2 4.2 -2.8
(x-x )2 3.24 0.64 1.44 0.04 17.64 7.84

σ =√∑(x-x )2 n

= √30.84
6  =√5.14 = 2.267

Coefficient of variation = σ

x  ×100

=

2.267
3.8 ×100 = 59.65

Example 2: Calculate coefficient of standard deviation from following data

15, 16, 18, 17, 13, 14.

Solution: Mean x = ∑ x n =

6 = 15.5

Let A=18 and preparing the following table

x 15 16 18 17 13 14

d = x-A -3 -2 0 -1 -5 -4

d2 9 4 0 1 25 16

Here, ∑ d = 15, ∑ d2= 55 σ = √∑d2 n - ( ∑d n )

=√55

6 - (

6 )
=√9.16-6.25  =   √2.91  = 1.705

Coefficient of standard deviation = S.D Mean =

1.705
15.5   = 0.11

Example 3: Calculate the standard deviation of following data

x 1 2 3 4 5
f 3 5 2 4 2

Solution: x f fx x-x

d2

(x-x )2= fd2

1 3 3 -1.81 3.276 9.828
2 5 10 -0.81 0.656 3.280
3 2 6 0.19 0.036 0.072
4 4 16 1.19 1.416 5.664
5 2 10 2.19 4.796 9.592
Total 16 45   28.436

Mean x = ∑ fx ∑ f =

16= 2.81

σ = √∑f(x-x )2 ∑ f = √∑fd2 ∑ f

=√28.436
16  =√1.777  =1.33

Example 4: Calculate mean and coefficient of variation

x 0-10 10-20 20-30 30-40 40-50
f 3 4 2 3 5

Solution: x f Mid-point m x-x d= x-A c

A=25,C=10

fd fd2

0-10 3 5 -1.81 -2 -6 12
10-20 4 15 -0.81 -1 -4 4
20-30 2 25 0.19 0 0 0
30-40 3 35 1.19 1 3 3
40-50 5 45 2.19 2 10 20
Total 17    3 39

Mean x =A+ ∑ fd ∑ f ×C

=25+
17 ×10
=25+1.76=26.76

σ = √∑d2 n - ( ∑d n ) ×C

= √39
17 - (
17)
×10
=√2.294-0.030  ×10
= √2.264  ×10
=1.504 ×10 =15.04

Coefficient of standard deviation = S.D

Mean ×100

=

15.04
26.76  ×100 = 0.562×100 =56.20

5Skewness and Understanding Distribution Shape in Biological Measurements

5.1Skewness and Distribution Shape

Skewness is a key statistical measure that shows how data is spread out in a dataset. It tells us if the data points are skewed to the left (negative skew) or to the right (positive skew) in relation to the mean. It is important because it helps us to understand the shape of the data distribution which is important for accurate data analysis and helps in identifying outliers and finding the be st statistical methods to use for analysis.

In this article, we will see skewness, different types of skewness and its core concepts.

5.1.1Types of Skewness

1Positive Skewness (Right Skew)

In a positively skewed distribution, the right tail is longer than the left which means most data points are on the left with a few large values pulling the distribution to the right. Relationship: Mean > Median > Mode Examples: Income distribution, exam scores and stock market returns.

2Negative Skewness (Left Skew)

In a negatively skewed distribution, the left tail is longer which means most data points are on the right with a few smaller values pulling the distribution to the left. Relationship: Mean < Median < Mode Examples: Test scores on easy exams, age at retirement and gestational age at birth.

3Zero Skewness (Symmetrical Distribution)

Zero skewness shows a perfectly symmetrical distribution where the mean, median and mode are equal. In a symmetrical distribution, the data points are evenly distributed around the central point. Relationship: Mean = Median = Mode Example: A perfectly balanced dataset with equal frequencies of all values.

Table: 5.1.1.Types of skewness

5.1.2Measurement of Skewness

Skewness is a statistical measure used to determine the degree and direction of asymmetry in a distribution. It indicates whether the data are symmetrically distributed or skewed toward the right or left side. Measurement of skewness is important in biological and pharmaceutical research because it helps resea rchers understand the shape of data distribution and select appropriate statistical methods for analysis.

Several methods are used to measure skewness. One of the simplest methods is based on the relationship among mean, median, and mode. Karl Pearson’s Coefficient of Skewness Skewness= Mean−Mode Standard Deviation When the mode is not clearly defined, the following formula is used: Skewness= 3(Mean−Mode ) Standard Deviation Interpretation Another commonly used method is Bowley’s Coefficient of Skewness , which is based on quartiles.

Skewness= Q3+Q1−2Q2

• Skewness = 0 → Symmetrical distribution
• Skewness > 0 → Positively skewed distribution
• Skewness < 0 → Negatively skewed distribution

Q3−Q1

Where: Measurement of skewness is widely used in biological measurements such as drug concentration, body weight, enzyme activity, blood glucose level, and hospital stay duration. Understanding skewness helps in identifying outliers, checking data normality, and improving the accuracy of statistical interpretation in pharmaceutical and biomedical research. Understanding skewness and distribution shape is essential in pharmaceutical and biological research because it improves data interpretation, supports accurate statistical analysis, and enhances evidence-based decision-making.

• Q1 = First quartile
• Q2 = Median
• Q3 = Third quartile

5.1.3Importance in Biological Measurements

Skewness plays a significant role in biological and pharmaceutical measurements because it helps in understanding the pattern and distribution of collected data. In biological sciences, measurements such as blood pressure, blood glucose level, enzyme activity, plasma drug concentration, body weight, cholesterol level, and hospital stay duration often do not follow a perfectly symmetrical or normal distribution. The study of skewness helps researchers determine whether the data are evenly distributed or shifted toward higher or lower values.

One of the major importances of skewness is its role in selecting appropriate statistical methods for data analysis. Many parametric statistical tests, including the t-test, correlation, regression, and analysis of varia nce (ANOVA), assume that the data are normally distributed. When biological data show positive or negative skewness, these assumptions may not be valid.

In such cases, researchers may apply data transformation methods or use non -parametric statistical techniques to improve the accuracy and reliability of the results. Biological and pharmaceutical data are often skewed due to: Overall, the measurement and interpretation of skewness provide important insights into biological variability and data behaviour. It supports accurate statistical analysis, improves research reliability, enhances clinical decision-making, and contributes to better understanding of biological and pharmaceutical phenomena.

• Patient variability
• Disease heterogeneity
• Outliers
• Instrument sensitivity

6Descriptive Statistical Analysis Using Python Libraries such as NumPy and Pandas

Descriptive statistical analysis is an essential step in biological, pharmaceutical, and clinical research because it helps summarize, organize, and interpret large amounts of data in a meaningful form.

6.1Python

Python is a high-level, interpreted, and general-purpose programming language widely used in scientific research, data analysis, artificial intelligence, web development, and pharmaceutical sciences. Python is known for its simple syntax, readability, and ease of learning, making it suitable fo r students, researchers, and professionals. It supports multiple programming paradigms such as procedural, object-oriented, and functional programming.

6.1.1Python Library

A Python library is a collection of prewritten modules, functions, and packages th at provide specific functionalities to perform different tasks without writing code from scratch. Libraries simplify programming by offering reusable code for mathematical operations, data analysis, visualization, machine learning, web development, and scientific computing. 6.1.2.Common Python Libraries Used in Pharmaceutical and Biological Research

1NumPy

NumPy (Numerical Python) is one of the fundamental Python libraries used for numerical and mathematical computations. It provides support for large multidimensional arrays and matrices along with a collection of mathematical functions to perform fast calcu lations. NumPy is widely used in pharmaceutical and biomedical research for statistical analysis, pharmacokinetic calculations, and scientific computing.

Main Features Common Functions

• Performs mathematical operations efficiently
• Supports arrays and matrices
• Provides statistical and algebraic functions
• Faster computation compared to normal Python lists

import numpy as np
data = [10, 20, 30, 40]
print(np.mean(data))
print(np.median(data))
print(np.std(data))

Pharmaceutical Applications

• Drug concentration analysis
• Statistical calculations
• Pharmacokinetic studies
• Biological data processing

2Pandas

Pandas are a powerful Python library used for data manipulation and data analysis. It provides easy-to-use data structures such as Series and Data Frame for handling structured data. Pandas are commonly used in clinical research, epidemiology, pharmacovigilance, and healthcare data analysis.

Main Features Common Functions

• Organizes data into tables
• Handles missing values
• Performs sorting and filtering
• Generates statistical summaries

Import pandas as pd

data = { 'Patient': ['A', 'B', 'C'],

'BP': [120, 130, 125]

} df = pd.DataFrame(data)

print(df)
print(df.describe())

Pharmaceutical Applications

• Clinical trial data analysis
• Patient record management
• Laboratory data handling
• Pharmacovigilance databases

3Matplotlib

Matplotlib is a visualization library used for creating graphs, charts, histograms, scatter plots, and other scientific figures. It helps researchers understand patterns, trends, and relationships in pharmaceutical and biological data. Main Features Common Functions

• Creates high-quality graphs
• Supports line charts, bar charts, histograms, and pie charts
• Customizes labels, colors, and titles

import matplotlib.pyplot as plt

x = [1, 2, 3, 4]

y = [10, 20, 25, 30]
plt.plot(x, y)
plt.xlabel("Time")
plt.ylabel("Drug Concentration")
plt.title("Drug Concentration vs Time")
plt.show()

Pharmaceutical Applications

• Pharmacokinetic curves
• Clinical trial visualization
• Statistical graph plotting
• Quality control charts

4SciPy

SciPy (Scientific Python) is a library used for advanced scientific and technical computing. It is built on NumPy and provides additional modules for optimization, statistics, signal processing, and integration. Main Features Common Functions

• Advanced statistical analysis
• Scientific computation tools
• Optimization techniques
• Signal and image processing

From Scipy import stats
data = [12, 15, 18, 20, 22]
print(stats.tmean(data))

Pharmaceutical Applications

• Statistical hypothesis testing
• Bioinformatics analysis
• Signal processing in medical devices
• Experimental data analysis

5Scikit-learn

Scikit-learn is a machine learning library used for predictive analysis and data modeling. It

provides algorithms for classification, regression, clustering, and data mining. Main Features

• Machine learning algorithms

 Predictive modeling

Common Functions

• Data classification and clustering
• Model evaluation tools

from sklearn.linear_model import LinearRegression

Pharmaceutical Applications

• Disease prediction
• Drug discovery
• Patient risk analysis

 Predictive healthcare models

6.1.3Importance of Python Libraries in Pharmaceutical Sciences

Python libraries simplify scientific programming and improve efficiency in pharmaceutical and biomedical research. They help researchers perform statistical analysis, visualize experimental

results, manage clinical datasets, develop predictive models, and interpret biological

measurements accurately. These libraries save time, reduce computational errors, and support evidence-based healthcare research. Role of NumPy and Pandas in Descriptive Statistics NumPy is a Python library mainly used for numerical computations and handling multidimensional arrays.

It provides mathematical and statistical functions that allow researchers to calculate measures such as mean, median, standard deviation, variance, minimum, and maximum values efficiently. Pandas is a data analysis library built on top of NumPy. It provides d ata structures such as Series and DataFrame that simplify the organization, manipulation, and analysis of tabular data.

Pandas is particularly useful in pharmaceutical and biological research because it allows easy handling of patient records, clinical trial data, laboratory results, and pharmacokinetic datasets. Common Descriptive Statistical Measures Descriptive statistics help summarize the central tendency, variability, and distribution of data.

1Mean

The mean represents the average value of a dataset and is calculated by dividing the sum of all observations by the total number of observations. Mean= ∑X N In pharmaceutical research, the mean may represent the average blood glucose level, average drug concentration, or average patient age. Interpretation of Mean Useful for normally distributed data

• A higher mean indicates larger average values.
• Sensitive to extreme values (outliers).

2Median

The median is the middle value in an ordered dataset and represents the central point of the distribution. Interpretation of Median

• Less affected by extreme values.
• Preferred for skewed biological data.
• Useful in income distribution, hospital stay duration, and pharmacokinetic data.

3Mode

The mode is the most frequently occurring value in a dataset. Interpretation of Mode

• Useful for categorical or nominal data.
• Indicates the most common observation.

4Standard Deviation

Standard deviation measures the spread or dispersion of data around the mean. SD(σ) = √∑(𝑿−𝑿̅)𝟐 𝑵 Interpretation of Standard Deviation

• Small SD → data are closely clustered around the mean.
• Large SD → greater variability in observations.
• Important in assessing consistency of drug response

5Variance

Variance measures the squared deviation from the mean. Variance (𝜎2) = ∑(𝑿−𝑿̅)𝟐 𝑵 Interpretation of Variance Example of Descriptive Statistical Analysis Using Python Sample Dataset Suppose a researcher records systolic blood pressure values from patients:

• Higher variance indicates more variability.
• Helps compare variability between datasets.

import Numpy as np
import Pandas as pd
bp = [120, 125, 118, 130, 128, 122, 135, 140, 126, 124]

data = pd.DataFrame(bp, columns=['Blood_Pressure'])

print("Mean:", np.mean(bp))
print("Median:", np.median(bp))
print("Standard Deviation:", np.std(bp))
print(data.describe())

Interpretation of Results Mean Blood Pressure The mean provides the average blood pressure of the study population. If the mean is elevated, it may indicate a higher risk of hypertension within the group. Median Blood Pressure The median helps determine the central tendency when extreme values are present.

If the median differs greatly from the mean, the data may be Skewed. Standard Deviation A low standard deviation indicates that patient blood pressure values are close to the average, whereas a high standard deviation suggests greater variation among patients. Minimum and Maximum Values These values help identify the range of measurements and detect unusually lo w or high observations.

Quartiles Quartiles divide the dataset into four equal parts and help understand data spread and variability. Importance in Pharmaceutical and Biological Research Descriptive statistical analysis using Python is highly valuable in p harmaceutical sciences because it: Python libraries such as NumPy and Pandas simplify statistical calculations, reduce computational errors, and improve reproducibility of research findings. Therefore, they are widely used in Clinical researc h, Pharmacovigilance, Epidemiology, Bioinformatics, and Pharmaceutical quality control studies.

Conclusion Descriptive statistics forms the foundation of data analysis in pharmaceutical and biomedical sciences by providing systematic methods for organizing, summarizing, and interpreting data. Understanding different types of data and their sources is essential for selecting appropriate statistical techniques and ensuring the reliability of research findings. Measures of central tendency, such as mean, median, and mode, help identify the representative value of a dataset, while measures of dispersion, including range, variance, and standard deviation, describe the variability and consistency of observations.

The study of Skewness and distribution shape further enhances the interpretation of biological and pharmaceutical data by revealing patterns of asymmetry and guiding the selection of suitable statistical methods. These concepts are widely applied in clinical trials, pharmacovigilance, quality control, and pharmacokinetic studies, where accurate data analysis is crucial for evaluating drug safety, efficacy, and quality. With the increasing availability of digital tools, Python libraries such as NumPy, Pandas, Matplotlib, SciPy, and Scikit-learn have become valuable resources for performing statistical calculations, managing datasets, and visualizing results efficiently.

Their application improves the accuracy, reproducibility, and efficiency of pharmaceutical research. Overall, a strong understanding of descriptive statistics and its practical implementation using modern computational tools enables researchers, healthcare professionals, and pharmaceutical scientists to make informed decisions, draw meaningful conclusions, and contribute to evidence-based healthcare and drug development. References 1.Daniel, W.

W., & Cross, C. L. (2018).

Biostatistics: A foundation for analysis in the health sciences (11th ed.). Wiley. Page No: 29-42 2.Rosner, B.

(2015). Fundamentals of biostatistics (8th ed.). Cengage Learning.

Page No: 45–55 3. Friedman, L. M., Furberg, C.

D., DeMets, D. L., Reboussin, D. M., & Granger, C.

B. (2015). Fundamentals of clinical trials (5th ed.).

Springer. Page No: 1-10 4. Daniel, W.

W., & Cross, C. L. (2018).

Biostatistics: A foundation for analysis in the health sciences (11th ed.). Wiley. 5. McKinney, W.

(2022). Python for data analysis (3rd ed.). O'Reilly Media. 6.

Motulsky, H. (2022). Intuitive biostatistics: A nonmathematical guide to statistical thinking (5th ed.).

Oxford University Press. 7.Bolton, S., & Bon, C. (2019). Pharmaceutical statistics: Practical and clinical applications (6th ed.).

CRC Press. 8.Elzinga, C. H., Molenberghs, G., & Verbeke, G. (2021).

Statistics for pharmaceutical and biomedical sciences. Springer. 9.VanderPlas, J. (2023).

Python data science handbook (2nd ed.). O'Reilly Media. 10.McKinney, W. (2010).

Data structures for statistical computing in Python. Proceedings of the 9th Python in Science Conference, 56–61 11.Hunter, J. D.

(2007). Matplotlib: A 2D graphics environment. Computing in Science & Engineering, 9(3),

• Summarizes clinical and laboratory data efficiently
• Helps identify trends and patterns
• Assists in data cleaning and error detection
• Supports interpretation of pharmacokinetic and clinical trial data
• Improves decision-making in healthcare research

90–95

12.Harris, C. R., Millman, K. J., van der Walt, S.

J., et al. (2020). Array programming with NumPy.

Nature,

585(7825), 357–362.

13.Shargel, L., Wu-Pong, S., & Yu, A. B. C.

(2015). Applied biopharmaceutics and pharmacokinetics (7th ed.). McGraw-Hill. 14.

Manikandan S. Frequency distribution. J Phamacol Pharmacother

2011;2:54-6.

15. Kothari, C. R. 1985.

Research Methodology: Methods and Techniques. Delhi: New Age International (P) Limited. 16. Rao PS Sundar, Richard J.

Introduction to biostatistics and research methods. 4th ed. New Delhi, India: Prentice Hall of India Pvt Ltd; 2006. 17. Macbin D, Campbell MJ, Payers P, Pinol A.

Statistical Tables for the Design of Clinical Studies .Oxford: Blackwell Scientific Publications, 1996.