Bakhtyar Ali Shah

Objectives of the lecture

 At the end of the lecture the students
should be able to define;
 Statistics
 Population
 Inclusion and exclusion criteria
 Data
 variables

2
DEFINITIONS
 STATISTICS (political science??)

 Aggregate of facts. Eg: patients attending a

clinic
 Characteristics of a data set. Eg: mean, SD.

 Applied Mathematics dealing with

collecting, classifying, analyzing,

summarizing and interpreting data.

3
DEFINITIONS
 Limitation of statistics
1. Statistical laws are held to be true on
the average
2. Can analyze only the collective matters
and not individual events
3. Results are ascertained by samples so
affected by sampling errors

4
 3.TARGET POPN (Syn:
Reference; Universe; 4.EXTERNAL
Source Popn ;) POPN

1.Study (syn: Actual;

Examples: KMC
Accessible) POPN
student for addiction

2.SAMPLE Immigrant population

of Pakistan in England,
2)sampling error or and extrapolating the
random error results to Pakistan5
 General Population

Inclusion criteria
Age of LESS then 60 years is included

Included Population

Exclusion criteria
Diabetic patients

Study Population

Sample
6
Parameter: is a characteristic of
measurement summarized by descriptive
statistics for population. UNKNOWN
Eg. Average income of household in Pakistan.
Hb% level of People of Pakistan.
Statistic: is the same characteristic / set of
measurement summarized by descriptive statistics
for sample. ESTIMATED BY SAMPLE

7
DATA (Datum)
Collection of information in numerical form

SOURCES
 PRIMARY DATA:

 Collected directly from the field of inquiry

 Collected for a specific purpose

 Information provided is detailed, complete and to the

point
 SECONDARY DATA

 Date Collected for another purpose by another agency

 Contain unnecessary information

 Cost effective 8
 VARIABLES
Variable is characteristics of a person,
object or phenomena that is amenable to
change and is measurable.
 Qualitative (or categorical or attribute)
data
can be separated into different categories
that are distinguished by some nonnumeric
characteristics or quality. Gender, occupation,
• Quantitative data
numbers representing counts or measurements

9
VARIABLES (cont’d)
Qualitative data
NOMINAL & ORDINAL

1 Nominal
Characterized by data that consist of names, labels, or
categories only. The data cannot be arranged in an
ordering scheme (such as low to high)
Example: occupation. Government service, business,
teacher
survey responses yes, no 10
 VARIABLES (cont’d)
 2 Ordinal
involves data that may be arranged in some
order, but differences between data values either
cannot be determined or are meaningless

Example: Severity of cough: Mild, Moderate,

Severe. Economic status: poor, middle or upper
class
11
 VARIABLES (cont’d)
2 Quantitative data
Continuous
(numerical) data result from infinitely possible
values that correspond to some continuous scale that
covers a range of values without gaps, interruptions, or
jumps

2 3

12
 VARIABLES (cont’d)
2 Quantitative data

Discrete
data result when the number of
possible values is either a finite number or a
‘countable’ number of possible values
0, 1, 2, 3, . . .

13
 VARIABLES
(Epidemiological classification)

A dependent variable is one which

changes as a result of the independent
variable being changed. (disease)
(outcome)
An independent variable is collected /
manipulated by the researcher.
(exposure) (risk factors)
14

VARIABLES (cont’d)
Example:
Scientists will manipulate the vitamin C intake in a
group of lets say 100 people who are over the age
of 40 years.
50 people will be given a daily high dose of vitamin
C and 50 people will be given a placebo pill over a
period of 25 years.
The goal is to see if the high vitamin C dosage
affects the people's life span
Independent variable of your daily vitamin C
intake can determine the dependent variable of
your life span 15
 Examples of research question for type of data.

 Are birth weights of new born of mothers exposed

to smoking lower then those, whose mothers are not
exposed?
 Is there any relationship b/w mothers nutritional
status and low birth weight babies?
 Does the average duration of stay of a new born
male baby in ICU more then new born female baby?
 Is there any effect of education of mother on
prevalence of diarrhea among children under 5
years?

16
 EXAMPLE
A researcher was interested in studying the
proportion of adult smokers in Peshawar. A
random sample of size 10000 adults was selected.
It was observed that 42% of the adults were
smokers.
 What was the researchers population?
 What was the researchers sample?
 What was the measurement of interest to the researcher?
 What type of measurement was researcher dealing with?
 What was the parameter of interest?
 What was the estimate for parameter in the population?

17
 SOLVED
A researcher was interested in studying the
proportion of adult smokers in Peshawar. A random
sample of size 10000 adults was selected. It was
observed that 42% of the adults were smokers.
 What was the researchers population? Adult population of
Peshawar
 What was the researchers sample? 10000 adults of Peshawar
 What was the measurement of interest to the researcher?
smokers
 What type of measurement was researcher dealing with?
qualitative
 What was the parameter of interest? Proportion of adult smokers
 What was the estimate for parameter in the population? 42%

18
 TUTORIAL
A researcher was interested in studying the average
income of household in Pakistan. A random sample of
size 1000 household was selected. The average income
of these household was Rs 3000 per month.
 What is your population?
 What is your sample?
 What is your measurement of interest?
 What type of measurement are you dealing with?
 What is the parameter of interest?
 What is the estimate for parameter in the population?

19
 TUTORIAL SOLVED
A researcher was interested in studying the average income
of household in Pakistan. A random sample of size 1000
household was selected. The average income of these
household was Rs 3000 per month.
 What is your population? household in Pakistan
 What is your sample? 1000 household
 What is your measurement of interest? income of household
 What type of measurement are you dealing with? Continuous
 What is the parameter of interest? average income of household in
Pakistan
 What is the estimate for parameter in the population? Rs 3000 per
month
 If you are asked to select a random sample of 1000 household from the
population. what would be your strategy to select a representative
sample? Explain. 20
HOME WORK

An epidemiological study was carried out in

Peshawar to see whether obesity is associated with
hypertension in young adults. The investigators
decided that it was not feasible to take a sample
from all the young adults in peshawar.
It was felt by the investigators that commercial
fitness centres may provide a good source of young
adults.

21
A random sample of young adults was studied from
several fitness centres, randomly selected from a
detailed list of fitness centres in peshawar and the
body weight and blood pressure of the selected
subjects was measured.
Answer the following questions :
(a) What is the target population for this study ?
(b) What is the study population in this study ?
(c) Does the sample represent the study population ?
(d) Does the study population represent the target
population ?
22
 o u
k y
a n
T h

23