3. Statistical Analysis
In Statistics collecting information for statistical analysis is called
collection of Data and the aggregate of the objects of study is
called the population .
There are two methods of collecting data:
1. Census Method: It is the study of whole population.
2. Sampling Method: It is the study of the sample of the
population.
4. What is Sampling?
Sampling is the act, process, or technique of selecting a suitable
sample, or a representative part of a population for the purpose
of determining parameters or characteristics of the whole
population.
•The sample is representative
of the population.
5. i. Time: Studying a sample out of whole population saves
time of studying each and every element of population.
ii. Cost: Studying a sample out of whole population saves
cost involved in studying each and every element of
population.
iii. Reliability: Results obtained from sampling are often
more reliable than results obtained from study of
whole population.
iv. Details of information: Again as the size of a sample is
small, every member of the sample can be studied
rigorously and detailed information can be obtained
about it .
Advantages of Sampling
6. TERMINOLOGIES
Population: the aggregation of the specified elements as defined for a
given survey.
Sample or Target population: the aggregation of the population from
which the sample is actually drawn.
Sample frame: a specific list that closely approximates all elements in the
population—from this the researcher selects units to create the study
sample.
Sample: a set of cases that is drawn from a larger pool and used to make
generalizations about the population.
Sample element: a case or a single unit that is selected from a population
and measured in some way.
Interface: It is the result of the sampling process. It is the data collected
from samples which can depict the characteristics of the population.
8. There are 2 methods of sampling:
Non-Probability sampling
Probability sampling
Types of Sampling
9. Non-Probability sampling
• In Non-Probability sampling an item is
included in the sample on the basis of
personal judgment of the investigator.
10. Types of Non-Probability sampling
Judgement sampling: the selection of
respondents is predetermined according to
the characteristic of interest made by the
researcher.
eg: If we want to investigate expenditure pattern
of 900 students on roll, say we study 100
students at our will. So there is no rule for
going against our own will.
11. Quota sampling: To avoid the expenses of
approaching the chosen people, in quota
sampling the investigator interviews all the
people he meets up to a certain number
called his Quota. E.g.: age, sex, working
class,etc.
12. Cluster sampling: in this case we first classify all
the members into several groups or clusters
wherein the individual members belonging to
the cluster will be chosen from. Next, we
perform a simple random sampling of size k
from the K clusters formed. Finally, all the
members from each of the k clusters will be
enumerated and be a part of the sample.
13. Convenience sampling: As the name suggests in
this method items for sample are selected
according to the convenience of the
investigator.
E.g.: If a person wants to study problems of the
higher education, he may choose a college
nearer to his residence and interview
students & teachers over there.
14. Sequential sampling: Here a number of sample
are drawn one after other.
If the result obtained from the first sample are
satisfactory then further samples are not
drawn. But if the results obtained from first
sample are not satisfactory the first sample is
rejected. Now a second sample is drawn and
so on…. similarly
15. Simple Random sampling : the technique of
obtaining the sample by giving each member
of the population an equal chance of being
included in the sample.
Example: If we want to select 100 boys out of 900.
Lottery method: We can write names of all boys on chits of paper and
select 100 chits giving names of selected boys.
Table of random numbers: Random numbers table, drawing out of a hat,
random timer, etc.
Types of Probability sampling
16. Systematic sampling : easier alternative to simple
random sampling in the sense that there will be
less pieces of paper to prepare and to be drawn.
E.g.: If we want to select 100 girls out of 900 girls.
We need to make random list of there names.
And then say we decide to choose every 9th girl
out of first 10. then girls listed 19,29,39….will be
choose for sample to be studied .
17. Multi-stage sampling : - is used when the
population is very large and coming from a wide
area.
E.g.: If we are conducting survey for teachers from
Maharashtra. Then at first stage we will divide state into
different districts and select few districts. Then at second
stage districts may be divided in talukas and select few
districts. Then at third stage talukas may be divided into
villages and after selecting few villages we reach final
stage where we get small concentrated areas where
detailed enquiry is done.
18. Stratified sampling
Divide the population by certain characteristics into
homogeneous subgroups (strata) (e.g., UI PhD students,
Masters Students, Bachelors students).
Elements within each strata are homogeneous, but are
heterogeneous across strata.
A simple random or a systematic sample is taken from each
strata relative to the proportion of that strata to each of the
others
E.g.: If we conduct sample survey of lung cancer for 1000 smokers,
out of which
• 300 smoke pipe
• 500 smoke cigarette
• 200 smoke bidi
Suppose we have to select a sample of 250 i.e one fourth of total SMOKERS .
So we need to select one forth of every strata i.e. 75,125 and 50
19. It is generally some difference between a static
value (the value obtained from samples) and
its corresponding parameter (the value
obtained from population)
This difference is called SAMPLING ERROR.
Sampling Error