Fundamentals Biostatistics

Lecture 1: Introduction
Definitions, Data, Variables, Population, Sample and Sampling
Strategies
Definitions and Scope of Biostatistics, Data, and
Variables

2
Statistics
▪ Statistics is the science whereby inferences are made about specific random
phenomena on the basis of relatively limited sample data.

▪ It is concerned with the collection, organization, summarization, and analysis of data;

and the drawing of inferences about a body of data when only part of the data is
observed.

▪ The field of statistics has two main areas: mathematical statistics and applied statistics:
o Mathematical statistics concerns the development of new methods of statistical inference
and requires detailed knowledge of abstract mathematics for its implementation.

o Applied statistics involves applying the methods of mathematical statistics to specific

subject areas, such as health, economics, psychology, and business.

3
Biostatistics

▪ Biostatistics is the branch of applied statistics that applies statistical methods to

medical and biological problems.

▪ Examples of applications of biostatistics

o Design and analysis of clinical trials.

o Sampling and survey design.

o Data analysis and presentation.

o Determining disease prevalence and incidence.

o Identifying risk factors and studying disease prognosis.

o Designing population-based interventions

o Evaluating public health policies.

4
Data
Definition
Data are observations that are collected, measured, or recorded during a
research study, experiment, survey, or any other data collection process. Data
can take various forms and may be categorized based on its nature and
characteristics.

Types of Data:
o Quantitative Data: Numerical data that represents quantities and can be
measured. Examples include height, weight, temperature.
o Qualitative Data: Categorical or non-numerical data that describes qualities
or characteristics. Examples include blood type, gender, or opinion.

5
Variable

Definition

It is a characteristic that takes on different values in different persons,

places, or things.

Examples:

Heart rate, height, weights, age, gender, education level, blood type,
disease stage, smoking status, ....

6
 Types of Variables

Quantitative Qualitative

Quantitative Variables Qualitative Variables

It can be measured in the usual Characteristics that are not
sense. capable of being measured. They
are only classified into categories.

For example:
For example:
Smoking status, disease status
Height, weight, time, ....
(present absent), blood type,
disease stage (stage I, II, III ,...),
severity of disease (mild,
moderate, severe)

7
 Types of Quantitative Variables

Discrete Continuous

A discrete variable A continuous variable

is characterized by gaps or can assume any value within a
interruptions in the values that specified relevant interval of values
it can assume. assumed by the variable.
For example: For example:
▪The number of patients ▪Height, weight, time, skull
admitted to a hospital per day. circumference, temperature.
▪The number of car accidents No matter how close together the
per day in a city observed heights of two people,
we can find another person whose
height falls somewhere in
between.

8
 Types of qualitative variables

Nominal Ordinal

As the name implies it consist of Whenever qualitative observation

“naming” or classifies into can be ranked or ordered according
various mutually exclusive to some criterion.
categories
For example: For example:
- Gender (Male – female) - Blood pressure (high-good-low)
-Health status (Sick – well) -Grades (Excellent - V.good - good -
fail)
-Marital status (Married - single
- divorced )

9
Measurement scales (1 of 3)

Definition: Measurement scales, also known as data scales or levels of

measurement, categorize variables into different types based on the nature
and characteristics of the data. The scale of measurement influences the types
of statistical analyses that can be applied to the data. There are four main
types of measurement scales:
1. Nominal Scale

2. Ordinal Scale

3. Interval Scale

4. Ratio Scale

Note: measuement scales have implicatications on the choice of statistical methods

to be used to analyze data

10
Measurement scales (2 of 3)
1. Nominal Scale:
◦ Description: The simplest level of measurement that categorizes data into distinct
categories or groups without any inherent order or ranking.
◦ Examples: Categories like gender, color, or marital status.
◦ Properties: No numerical values are assigned, and there is no meaningful order
among categories.

2. Ordinal Scale:
◦ Description: Orders and ranks data into distinct categories, but the intervals
between categories are not consistent or meaningful.
◦ Examples: Educational levels (e.g., high school, bachelor's, master's), survey
responses (e.g., strongly agree, agree, neutral, disagree, strongly disagree).
◦ Properties: Relative order is meaningful, but the differences between ranks are not
consistent.

Note: Both nominal and ordinal scales are used to describe measurement level of
qualitative variables

11
 Measurement scales (3 of 3)
3. Interval Scale:
◦ Description: Orders and ranks data with consistent intervals between consecutive
points, but there is no true zero point.
◦ Examples: Temperature measured in Celsius or Fahrenheit, IQ scores.
◦ Properties: Has a meaningful order, consistent intervals, but the absence of a true
zero point means that ratios are not meaningful.

4. Ratio Scale:
◦ Description: Orders and ranks data with consistent intervals between consecutive
points and has a true zero point, allowing for meaningful ratios.
◦ Examples: Height, weight, income, age.
◦ Properties: Has a meaningful order, consistent intervals, and a true zero point,
making ratios meaningful.

Note: Both interval and ratio scales are used to describe the measurement level of
quantitative variables

12
Population, Sample and Sampling
Strategies

13
Population and Sample

▪ Population consists of all individuals (or objects) that are studied.

▪ Sample is a subset of the population.

14
Sampling vs census (1 of 3)
Definition:
◦ Sampling: Involves selecting a subset of individuals from a larger population
to represent and make inferences about that population.
◦ Census: Encompasses the complete enumeration or collection of data from
every individual in a population.

Inclusion:
◦ Sampling: Only a portion (sample) of the population is included in the study.
◦ Census: Every individual in the entire population is included.

Scope:
◦ Sampling: Provides information about the population through the study of a
smaller, carefully chosen subset.
◦ Census: Aims to collect data from the entire population, leaving no one out.

15
Sampling vs census (2 of 3)
Time and Resource Requirements:
◦ Sampling: Typically requires less time and resources compared to a census,
especially for large populations.
◦ Census: Can be resource-intensive and time-consuming, especially for large
populations.

Precision and Accuracy:

◦ Sampling: Provides estimates and inferences about the population;
precision depends on the sample size and methodology.
◦ Census: Assumes complete accuracy for the entire population but can be
logistically challenging.

Feasibility:
◦ Sampling: More feasible for large populations where a complete
enumeration is impractical.
◦ Census: More feasible for small populations or when resources allow for
complete data collection.
16
Sampling vs census (3 of 3)
Variability and Error:
◦ Sampling: Introduces variability due to the inherent randomness of the
sampling process; may involve sampling error.
◦ Census: Assumes no sampling error; potential errors may arise from non-
response or data collection issues.

Statistical Inference:
◦ Sampling: Statistical methods are used to infer characteristics of the
population from the sample.
◦ Census: Directly describes the entire population without the need for
statistical inference.

Representativeness:
◦ Sampling: Requires careful design to ensure that the sample is
representative of the population.
◦ Census: Presumed to be representative as it includes every individual.

17
Random sample and representativeness

▪ Random sample is when every individual in the population has a

chance to be selected in the sample.

▪ For a sample to be representative, it should be selected at random

and of adequate size

18
Probabilistic/Random Sampling Strategies

Probabilistic sampling techniques involve random selection, where

every individual in the population has a known, non-zero chance of
being included in the sample. These methods provide a solid
foundation for statistical inference and help ensure that the sample is
representative of the entire population. The common probabilistic
sampling techniques are:
1. Simple Random Sampling (SRS)
2. Stratified Random Sampling
3. Systematic Sampling
4. Cluster Sampling
5. Multi-Stage Sampling

19
Simple Random Sampling (SRS):

◦ Description: Involves randomly selecting individuals from the entire

population.
◦ Procedure: Each member of the population has an equal chance of
being chosen. This can be achieved using random number
generators or a random process.
◦ Advantages: Simple, easy to understand, and ensures every
individual has an equal probability of inclusion.
◦ Limitations: May not account for specific characteristics or
stratification within the population.
◦ Example: Assign each member of a population a unique number
and use a random number generator to select a specified number
of individuals.

20
Stratified Random Sampling:
◦ Description: Divides the population into subgroups or strata based
on certain characteristics, and then random samples are taken from
each stratum.
◦ Procedure: Ensures representation from different subgroups,
reducing sampling bias.
◦ Advantages: Improved precision and representation compared to
simple random sampling.
◦ Limitations: Requires accurate information about the population's
characteristics for effective stratification.
◦ Example: Divide a population of students into strata based on grade
levels (e.g., freshman, sophomore, junior, senior) and then
randomly select a proportionate number of students from each
stratum.

21
Systematic Sampling:

◦ Description: Selects every nth individual from a list after a random start.

◦ Procedure: Establish a random starting point, choose a constant interval,

and select individuals at that interval.

◦ Advantages: Simplicity, especially when a complete list of the population is

available.

◦ Limitations: Susceptible to periodicity if there's a pattern in the list that

aligns with the chosen interval.

◦ Example: Select every 10th patient from a list of patients visiting a clinic
after randomly choosing a starting point between 1 and 10.

22
Cluster Sampling

◦ Description: Divides the population into clusters and randomly

selects entire clusters. All individuals within the selected clusters
are included in the sample.
◦ Procedure: Randomly select clusters, and then include all
individuals within those clusters.
◦ Advantages: Cost-effective for large populations, especially when
it's difficult to obtain a complete list of the population.
◦ Limitations: Potential for greater variability within clusters; not as
precise as some other methods.
◦ Example: Divide a city into clusters based on neighborhoods,
randomly select a few neighborhoods, and then survey all
households within the selected neighborhoods.

23
Multi-Stage Sampling:

◦ Description: Combines two or more sampling methods in stages.

◦ Procedure: Involves multiple levels of sampling, such as cluster
sampling followed by simple random sampling within selected
clusters.
◦ Advantages: Can be more practical and cost-effective for large and
diverse populations.
◦ Limitations: Complexity increases with each stage, requiring careful
planning and execution.
◦ Example: In a study of schools in a country, first, randomly select
several districts, then within each district, randomly select schools,
and finally, within each school, randomly select students.

24