The Session began with a revision of concepts studied in the last
class. i.e the distinction between chi-square and T-test methods and their applicability under different
situations. For eg : We have two groups of 6 people each with similar age groups. One of these groups
is administered with a drug. The effect of this medicine shall constitute the
output and it shall be a continuous variable.
Sampling
Census is an
example where the population is calculated using a sample .
Sample should be
made aas representative of the population. If this is achieved then a small
sample will do and the cost benefit shall also be achieved.
Target population
is the people we want to target and is smaller than the cenus
Sampling frame is a
list of all the respondents within a sample.
Sample is the smallest representation of the people.
Methods of
Sampling
1)
Probability sampling
a)
Simple Random
b)
Systematic Random : From the
sampling frame we decide a certain number of observations we require at regular
intervals. Which number to start with may be decide in a number of ways.
c)
Cluster Sampling : The sample
is more representative of the population because it has different elements of
population.
d)
Stratified Sampling : In
stratified sampling we have uniformity i.e. we find all our samples at one
place.
Benford's Law : It refers to the frequency refers to frequency distribution of the digits in many real life sources of data. In this distribution,the number 1 occurs as the leading digit about 30% of the time,while larger numbers occurs in that position less frequently,9 as the first digit less than 5% of the time.
Non-Probability methods
1)
Convenience sampling: Convienence sampling is a non-probability sampling where subjects are selected because of thier conveinent accessibility and proximity to the researcher.
2)
Judgement sampling : When take
the opinion of an expert as to where we can get a good sample,it is called judgement sampling.
3)
Quota Sampling : The sample is
in the same proportion of elements as the population
4)
Snowball or reference sampling
: This is a non probabilty technique where existing study subjects recruit future subjects among thier acquaintances.
Written By,
Nishidh Lad
Group Members
Nishidh Lad
PPriyadarshi Tandon
Priyathaam Kireeti
Kalyani Pakala
p.Karteeki
No comments:
Post a Comment