Professional Documents
Culture Documents
STUDY POPULATION
SAMPLE
TARGET POPULATION
• A representative sample has all the important characteristics of the population from
which it is drawn
• Nigerian registered voters
• Generalizability refers to the extent to which we can apply the findings of our research to
the target population we are interested in
• 1000 voters 93.4 million voters
NPMCN MD BIOSTATISTIC TOPIC SEMINAR PRESENTATION 18/01/2023 7
INTRODUCTION
• Sampling
• Systematic process of selecting a number of units or participants from a defined study
population
• The procedure by which some members of the population are selected as representatives
of the entire population
• The difference between the two is whether the sample selection is based on
randomization or not
• This technique is more reliant on the researcher’s ability to select elements for a sample
• Outcome of sampling might be biased and makes it difficult for all the elements of
population to be part of the sample equally
• Researcher chooses the sample based on who they think would be appropriate for the study
• Used primarily when there is a limited number of people that have expertise in the area
being researched
• E.g. to select focus group discussion participants
• That is, each participant is given same survey or interview at two or more time
points; each period of data collection is called a "wave"
• Often chosen for large scale or nation-wide studies in order to gauge changes in the
population with regard to any number of variables from chronic illness to job stress
to weekly food expenditures etc.
NPMCN MD BIOSTATISTIC TOPIC SEMINAR PRESENTATION 18/01/2023 14
Non-Probability: Snowball sampling
• Also known as Network or chain referral sampling
• This technique can be used in situations where the population is a small group of individuals
who are completely unknown, difficult to reach/find and with special characteristic or rare
conditions
• Such as homeless people, drug abusers etc.
• Therefore we receive help from the first person who meets criteria for inclusion in a study
and then the person is asked to name others who meet these criteria. This referral technique
goes on, increasing the size of population like a snowball until the targeted sample size has
been attained
NPMCN MD BIOSTATISTIC TOPIC SEMINAR PRESENTATION 18/01/2023 15
Non-Probability: Snowball sampling
• A type of nonprobability sampling which involves the sample being drawn from
that part of the population which is close to hand. That is, readily available and
convenient
• In advance the general composition of the sample’s characteristic (such as age, sex, religion, social
class, income, education etc.) is decided and quota/numbers predetermined
• The population is first segmented into mutually exclusive sub-groups, just as in stratified
sampling
• Then judgement used to select subjects or units non-randomly from each segment until exact
proportions of certain types of data is obtained or sufficient data in different categories is collected
• It is this second step which makes the technique one of non-probability sampling
NPMCN MD BIOSTATISTIC TOPIC SEMINAR PRESENTATION 18/01/2023 19
Non-Probability: Quota Sampling
• The problem is that these samples may introduce bias into research
findings because not everyone gets a chance of selection
• This random element is its greatest weakness and quota versus probability
has been a matter of controversy for many years
• Drawback is person selected may not be representative of the total
population in each category
• Generalization may not be correct
NPMCN MD BIOSTATISTIC TOPIC SEMINAR PRESENTATION 18/01/2023 20
Non-Probability: Quota Sampling
• When every element in the population does have the same probability of selection, this is
known as an 'equal probability of selection' (EPS) design
• Such designs are also referred to as 'self-weighting' because all sampled units are given the same
weight
• This sampling technique uses randomization to make sure that every element of the
population gets an equal chance to be part of the selected sample
NPMCN MD BIOSTATISTIC TOPIC SEMINAR PRESENTATION 18/01/2023 22
SAMPLING METHODS
• Probability (Random) Sampling Methods include:
• Simple random sampling
• Systematic random sampling
• Stratified random sampling
• Multistage sampling
• Multiphase sampling
• Cluster sampling
• Every member of the entire target population has an equal chance of being
selected as a member of the sample
• Disadvantages:
• It is very difficult to achieve (i.e. time, effort and money)
• If sampling frame large, this method may be challenging or impracticable
• Minority subgroups of interest in population may not be present in sample in sufficient
numbers for study
NPMCN MD BIOSTATISTIC TOPIC SEMINAR PRESENTATION 18/01/2023 27
Probability Sampling: Systematic sampling
• A sample selected by listing a population sequentially and choosing members
at regular intervals
• First subject is selected then others are systematically selected through a
predetermined sampling interval
• There must be no periodicity in the population or sampling frame
• Sampling fraction (ratio) = number of units in sampling frame
• Technique usually requires a numbered list
NPMCN MD BIOSTATISTIC TOPIC SEMINAR PRESENTATION 18/01/2023 28
Probability Sampling: Systematic sampling
• DISADVANTAGES:
• Relatively time, effort and money consuming
• Sample may be biased if hidden periodicity in population coincides with that of selection.
• Difficult to assess precision of estimate from one survey
NPMCN MD BIOSTATISTIC TOPIC SEMINAR PRESENTATION 18/01/2023 30
Probability Sampling: Stratified sampling
• This technique divides the elements of the population into small subgroups
(strata) based on the similarity in such a way that the elements within the
group are homogeneous and heterogeneous among the other subgroups
formed
• Then the elements are randomly selected from each of these strata
• When examining multiple criteria, stratifying variables may be related to some, but not to
others, further complicating the design, and potentially reducing the utility of the strata
• In some cases (such as designs with a large number of strata, or those with a specified
minimum sample size per group), stratified sampling can potentially require a larger
sample than would other methods
• Combining any of the earliest described methods earlier in a variety of ways that
help address sampling needs in the most efficient and effective manner possible
NPMCN MD BIOSTATISTIC TOPIC SEMINAR PRESENTATION 18/01/2023 35
Probability Sampling: Multistage sampling
• All ultimate units (houses, for instance) selected at last step are surveyed
NPMCN MD BIOSTATISTIC TOPIC SEMINAR PRESENTATION 18/01/2023 36
Probability Sampling: Multistage sampling
• Not as effective as true random sampling, but probably solves more of the problems inherent
to random sampling
• Multistage sampling used frequently when a complete list of all members of the population
not exists and is inappropriate
• Moreover, by avoiding the use of all sample units in all selected clusters, multistage
sampling avoids the large, and perhaps unnecessary, costs associated with traditional cluster
sampling
NPMCN MD BIOSTATISTIC TOPIC SEMINAR PRESENTATION 18/01/2023 37
Probability Sampling: Multi Phase sampling
• Part of the information collected from whole sample & part from subsample
• Survey by such procedure is less costly, less laborious & more purposeful
NPMCN MD BIOSTATISTIC TOPIC SEMINAR PRESENTATION 18/01/2023 38
Probability Sampling: Cluster sampling
• An example of ‘two-stage sampling’ – a subset of sampling units within
selected clusters are randomly selected for inclusion in the sample
• Disadvantage:
• Sampling error is higher for a simple random sample of same size
• BIAS IN SAMPLING
• Systematic error in sampling procedures that leads to a distortion in the results of the
study
• Causes include
• Volunteer sample group
• Non probability sampling
• Loss of sample subjects in non random manner
• Non response of survey sample members in non random manner
NPMCN MD BIOSTATISTIC TOPIC SEMINAR PRESENTATION 18/01/2023 43
CONCLUSION
• The probability of an event A, p(A), is obtained by summing the probabilities of the outcomes
contained within the event A
• An event is said to occur if one of the outcomes contained within the event occurs
• Compound Event:
An event with more than one possible outcome
• Independent events
• The occurrence of one event does not affect the occurrence of the other event
• This is an event which occurs when and only when the other event fails to occur for mutually exclusive events
For example, If probability of females (F) is 15/40, then the probability of its complement Males (M) is
p(M) = 1 – p(F) = 25/40
NPMCN MD BIOSTATISTIC TOPIC SEMINAR PRESENTATION 18/01/2023 53
INTRODUCTION
• Equally likely: if all outcomes have equal chance of occurrence
• In some situations, notably games of chance, the experiments are conducted in such a way
that all of the possible outcomes can be considered to be equally likely, so that they must be
assigned identical probability values
• n outcomes in the sample space that are equally likely => each probability value will be 1/n
• In the toss of a coin, there are two equally likely outcomes, so the probability of any of the
outcomes in any toss is one in two
NPMCN MD BIOSTATISTIC TOPIC SEMINAR PRESENTATION 18/01/2023 54
PROBABILITY LAWS AND THEORIES
3. Subjective
• Probability is the frequency of favourable event divided by total number of possible outcomes
S ={1, 2, 3, 4, 5, 6}
p(1)=p(2)=p(3)=p(4)=p(5)=p(6)=
• The likelihood of occurrence of an event is calculated based on the information we collected from a series of
repeated trials actually observed or experienced
• Based on Empirical Law of Averages which assumes that the world works in such a way that the relative
frequency with which an event occurs in repeated trials always settles down to a limit
• If we toss a coin 100 times, and we have heads 50 times, we speak of p(H) = 50/100 = 1/2
• This is the probability based on our own judgement. i.e. on our personal experiences
• Measures one’s degree of belief (confidence) or disbelief (doubt) about a person, an event or a phenomenon
occurring
• E.g. there will be a cure for Ebola Viral Disease in the next 10 years
• Nigeria will get better under APC government
Solution
S = {1,2,3,4,5,6}
A = {2,4,6} Even
B = {2,3,5} Prime
= {2}
p(A or B) = p(A) + p(B) – p
= 3/6 +3/6 -1/6 = 5/6
NPMCN MD BIOSTATISTIC TOPIC SEMINAR PRESENTATION 18/01/2023 64
PROBABILITY LAWS AND THEORIES
• Independent events: One event’s outcome does not depend on (or is not
influenced by) the other
• Multiplication Law
• When two events (A and B) or more events are independent, the probability of joint
occurrence (e.g. the occurrence of A and B) is the product of the individual probabilities
p(A and B) = p(A) p(B)
Male 40 20 20 80
Female 60 40 20 120
• In general p(A|B) = =
• Consider the Probability of (25 – 34), given female = p(25 – 34/female)
• p(25 – 34|female) = p(25 – 34 and female)/p(Female)
• p(25 – 34|female) = (40/200)/(120/200)
NPMCN MD BIOSTATISTIC TOPIC SEMINAR PRESENTATION 18/01/2023 68
PROBABILITY LAWS AND THEORIES
• Car Warranties
A company sells a certain type of car, which it assembles in one of four
possible locations. Plant I supplies 20%; plant II, 24%; plant III, 25%; and
plant IV, 31%. A customer buying a car does not know where the car has
been assembled, and so the probabilities of a purchased car being from each
of the four plants can be thought of as being 0.20, 0.24, 0.25, and 0.31.
• Car Warranties
Each new car sold carries a 1-year bumper-to-bumper warranty.
P( claim | plant I ) = 0.05, P( claim | plant II ) = 0.11
P( claim | plant III ) = 0.03, P( claim | plant IV ) = 0.08
For example, a car assembled in plant I has a probability of 0.05 of receiving a claim on its warranty.
Notice that claims are clearly not independent of assembly location because these four conditional probabilities
are unequal
• Bayes’ Theorem
If is a partition of a sample space, then the posterior probabilities of the event conditional on
an event B can be obtained from the probabilities p and p using the formula
p
• Car Warranties
- The prior probabilities
- If a claim is made on the warranty of the car, how does this change these probabilities?
p
p
p
p
• Questions?
• Comments?
• Contribution?