Introduction to Statistics

MADIHA IQBAL

Biostatistician
[email protected]
 Objectives
• To give best understanding of basic terms
and sampling techniques
 History of Statistics
• The word statistics seems to have been
derived from the Latin word status or the
Italian word statist.
• Both these words mean a political state.
The word statist (Milton in the sense of a
statesman, i.e. a person well-versed in the
affairs of the state)
 Modern Era
• Modern concept of statistics was
illustrated by Sir R. A. Fisher (1890-1962),
• J. Neyman (1894-1983),
• E. S. Pearson (1895-1981) and many
others.
 Defination of statistics
Statistics is defined as.
• According to the croxton and cowdown
"statistics" may be defined as the science
of collection, presentation, analysis and
interpretation of numerical data"
• "Numerical facts systematically arranged"
• Statistical method to summarize the data
and use statistical procedure to reach
certain conclusion
 In simple meanings
Statistics is defined as.
• Numerical facts systematically arranged
• As a discipline
• Plural of Statistic
 Why we use Statistics

1. Clearly define the goals

2. Make a plan of what data to collect
3. Apply appropriate statistical method
4. Interpret the information
 Why we use Statistics
To give the answer of our daily lives problem
• How many males and females are present
here???
• What is the mortality rate of Pakistan
• What is proportion of male to female
students who failed every year in metric
• What is risk to develop the Lung Cancer
for those who are smokers
 Bio-Statistics
“Bio” means life.
“Statistics” Science of facts and figures
Also
It is the branch of statistics that concerns
with the applications of statistical methods
to medical and biological data.
 How Bio-Statistics Works
Planning

Collection
Interpretation of data

STATISTICS
Organization
Conclusion

Summarization
Analysis
 Types of Bio-Statistics
Statistics
Descriptive Statistics Inferential Statistics

Graphical Tabular Number

different types of Testing Hypothesis

 Data Presentation/Describing

Qualitative data catagorical Qualitative data numarical

data data

Tabulation Graphic Numaric Tabulation Graphic Numaric

al al al al
One way Bar chart Percentage Frequency Histogram
table Pie chart % distribution Polegone
Two way Cluster chart stem &leaf Box-wisker
table Imp point
N way Q1,
table Cente Q2,Q3,
Measure of dispersion,
r P10 ,
(Variation distribution
Mean, Skew,
range )
median mode Kurtosis
G.M
Varience
H.M
S.D.
T.M
Range,
Q.D, M.D
 DESCRIPTIVE STATISTICS
" deals with collection and presentation of the
data" OR
" deals with concept and method concerned with
summarization and description of important
aspect of numerical data"
(Represented with graph, chart and tables)
INFRENCIAL STATISTICS
“Deals with analysis and generalization the
conclusion” (Applying statistics)
Important Concepts
 Variables
(defined as a characteristics that varies
with an individuaL or an object)

 Confounding Variables
(that effects both the independent and
dependent variables and thus confuse the
result)
 Constant Variables
(Any value that does not but remain fixed
called constant)
 Variables & Constants
• Age Variables
• Weight
• Height
• Severity of Pain
• Salary
• Value of “g”
Constants
• Value of “pai”
QUALITATIVE QUANTITATIVE

Types of Variable
Catagorical Discrete

Attribute Countinuous
 Types of Variables
 QUANTITATIVE VARIABLES
(used to describe or measure the factors that are assumed to cause
or at least to influence the problem) OR (Can be xpressed as
non-numarically) example ; age, height, weight
 QUALITATIVE VARIABLES
(that gets modified under the influence of some other variable) OR
(Can be xpressed as non-numarically) example; sex, status,
education
 DISCRETE VARIABLES
(veriables which represent in count data) example ; person in
room number of death
 COUNTINUOUS VARIABLES
(it can takes on any value fractional or integral with the give
interval) example; age of children , height of the students
Please can you give some
examples of variables???
 Types of Variables
 Independent Variables
(used to describe or measure the factors
that are assumed to cause or at least to
influence the problem)
OR "which is responsibel for
change"
Dependent Variables
(that gets modified under the influence of
some other variable)
OR "The response of an indepndent
variable is called"
 For Example
Cause Effect/ outcome
Independent variable Dependent variable

Other Factors
(Confounding Variables)
 For Example
A relationship is shown between the low level of mother’s education and
malnutrition in under 5’s. However, family income may be related to the
mother’s education as well as to malnutrition

Mother’s Education Malnutrition

Independent variable Dependent variable

Family Income
(Confounding Variables)
 Examples
• In a study of a prevalence of a disease in
different age groups, the presence of the
disease may be referred to as a dependent
variable, whereas age is an independent
variable.
• In a study of an association between birth weight
of a child, gestation period(weeks) and smoking
status are possible factors which may influenced
the birth weight of a child. Birth weight is
dependent variable whereas smoking status and
gestation period are independent variables.
 Example
• In the study of early sitting, smiling and walking
of a child, the factors such as age, gender, birth
weight, type of feeding, education of mother and
father, birth order, number of siblings, etc. are
independent variables.

• In a study of mental disorders among elderly

population; gender, age, family type, education
level, income, family history, etc. may be taken
as independent variables.
 Measurement Scale

Nominal Ordinal Interval Ratio

 Measurement Scale
"Defined as the proces of assigning
numbers to the characteristics is called "

NOMINAL: (scale classified the object or event

in different catagories)
ORDINAL: (used when we put the value/object
from lowest to highest or most to least)
INTERVAL: ( processing a constant interval
size but not a true zero point OR is used for expressing
temperature)
RATIO: (is used to express height, weight, B.P
and Hb level )
 Data & It’s Nature
What is data??
Collection of observation in a
scientific way
 Qualitative Data
 Quantitative Data
--- Discrete data
--- Continuous data
 Measurement Scale
Nominal Scale: Male, Female.
Qualitative
Ordinal Scale: Severity of pain.

Interval Scale: Temperature.

Ratio Scales: A characteristic Quantitative

that has Zero as
absence.
Characteristics Level of Measurements
Nominal Ordinal Interval Ratio

Distinctiveness * * * *
Ordering * * *

Equal Interval * *
Absolute Zero *
Data Collection
 Data and it’s Types
What is data???

 Primary Data (Originally collected

data)

 Secondary Data (gone some

statistical procedures)
Such as Collection, classification
Tabulation and presentation.
 Why do we collect data
 Monitor progress
 Get information
 Evaluate performance and effectively use
of resource
 Identify gaps, between knowledge and
practice
 Make decisions
 Utilize in planning
 Commonly Used Data Collection
Procedure
• Experiments
• Survey
– Census (Costly and time consuming, large no.
of enumerators and supervisory staff)
• In-depth sample survey
– Save time and money
 Steps In Data Collection Methods
• Identification of the objectives
• Identification of time frame
• Selection of data instrument
• Development of data collection methods
• Selection of proper sampling design
• Responsibility of training staff
• Pre-testing
 Collection of Primary data

Direct Personal Investigation

(High Response rates but costly)
Indirect Personal Investigation
 (someone may have problem to
answer you directly)
Collection Through Questionnaires and
electronic methods
Collection Through Enumerators
Collection Through Local Sources
 Collection Of Secondary Data

Officials: Publication of Statistical Division,

Ministry of Finance, Federal and
Provincial Bureaus of Statistics
Population sciences, and
agriculture sciences
Semi-Official
Research Organizations
Private Sectors: NGOs
 Data Instruments
• Questionnaire
• Electronic devices
– Such as recorder, audio, video, etc…
 Questionnaire
A questionnaire is a document designed
for the purpose of seeking specific
information from the respondent.
 Types of questions
• Open Questions
» Difficult to analyze

• Closed Questions
» Pr-assigned choices
 Questions should be
• Simple (it is best to avoid the statements)
• Avoid asking more than one question
– Do you have temperature and pain
• Avoid from technical terms
• Watch out ambiguity
– Old people should be encouraged to
exercise; but how old is old
 Questions should be
• Avoid from leading questions
– How often do you exercise: respondent
should be exercised

• Level of language
 The most important note

1. Information collected in a
questionnaire should be based on and
limited to the objective of the study
2. It should be pre-tested for large and
expensive studies
Data collection methods
Sampling Methodology
Population & Sample
 Population & Sample
• Parameter: A calculated value from
Population

• Statistic: A calculated value from sample

 Purpose of Sampling
• statistical inference is to obtain
information about a population from
information contained in a sample.
•The sample results provide only
estimates of the values of the
population characteristics.
•With proper sampling methods, the
sample results will provide “good”
estimates of the population
characteristics.
 Advantages of Sampling
o Sampling Saves Money And Time
o useful when sampling unit are
sensitive
o for Infinite Population
o Smaller Non-Response
 Errors in sampling
• Sampling Error
– Sample size

• Non Sampling Error

– Failure to measure the units in selected sample
– Observational or defective measurements techniques
– Errors introduced in editing, coding and tabulating the
result
 Types of sampling

1. Probability Sampling
– Subjects have equal chance of selection
2. Non-probability sampling
– Subjects of un-equal chance of selection
 Probability Sampling

Probability – Sample is selected based on

a random process (which means it is
based on chance).
 Non-Probability Sampling

Non-probability– Sample is selected using

some nonrandom process (not based on
chance).