STATISTICAL TECHNIQUE FOR MANAGERS
MODULE 5: SAMPLING METHODS AND SAMPLING DISTRIBUTIONS s
Sampling Theory and Tests of Significance: The basics of Sampling, Sampling
Procedures: Random and Non-Random methods-Sample size determination-
Sampling distribution-Central Limit Business Applications of Hypothesis Testing -
Large Sample test, Small Sample Test. Chi-Square Analysis: Test for a specified
Population variance, Test for Goodness of fit, Test for Independence of Attributes.
Analysis of Variance- One Way and Two Way ANOVA
Sampling Theory and Tests of Significance
Sampling: Sampling is the process of
selecting a subset of individuals or items from a
larger group (population) to estimate
characteristics of the whole group.
Importance of Sampling:
Sampling allows researchers to draw conclusions
about a population
It is cost-effective, less time-consuming and often
more practical than conducting a census.
Difference between Population and Sample
Aspect Population Sample
Entire group under Subset of the population
Definition
consideration selected for study
Size Typically larger Smaller
Represents a portion of the
Representativeness Represents the entire group
population
Used for generalization and Used to make estimates about
Purpose
inference the population
Often expensive to collect data Less expensive as data
Cost
for the whole population collection is for a smaller group
May require more time to Requires less time for data
Time
collect data collection
Sampling Distribution
• Sampling distribution: refers to the distribution
of a statistic (such as the mean, median and
variance etc.) calculated from multiple samples
of the same size drawn from the same
population.
• It provides information about the possible values
and probabilities associated with a particular
statistic when calculated from various samples.
• Sampling distribution helps in making inferences
about the population parameters based on sample
statistics.
• It allows us to assess the variability of a statistic
across different samples.
Central Limit Theorem (CLT):
• The Central Limit Theorem (CLT) is a statistical
principle that states: "When you add together a
large number of random values, their average
tends to follow a bell-shaped curve, no matter
what the original shape of the values was."
The CLT states that, regardless of the shape of the
population distribution, the sampling distribution of
the sample mean will be approximately normally
distributed if the sample size is sufficiently large
(usually n ≥ 30).
Normal bell-shaped curve
Properties of Sampling Distribution:
Mean:
• The mean of the sampling distribution of the sample mean is equal to
the population mean.
Standard Deviation (Standard Error):
• The standard deviation of the sampling distribution of the sample mean
also known as the standard error and is equal to the population
standard deviation divided by the square root of the sample size (√n).
Shape:
• The shape of the sampling distribution becomes approximately normal
as the sample size increases, even if the population distribution is not
normal.
Cantering and Spread:
• The sampling distribution is centered around the population parameter
being estimated (e.g., population mean) and its spread decreases with
increasing sample size.
Effect of Sample Size:
• Increasing the sample size tends to make the sampling distribution
narrower and more concentrated around the population parameter.
• A larger sample size leads to a more precise estimation of population
parameters.
HYPOTHESIS TESTING
• The science of collecting, organizing, analyzing and
interpreting data in order to make decisions.
• Statistics is used to describe the data set and to draw
conclusion about the population from the data set.
Statistical methods are of two types:
Descriptive Method: This method uses graphs and
numerical summaries
Inferential Method: This method uses
confidence interval and Hypothesis
Testing for significance test which are part of
applied statistics.
HYPOTHESIS TESTING
• Hypothesis is a idea about a group or population.
• Hypothesis refers to an educated guess or assumption
that can be tested.
• Hypothesis is formulated based on previous studies.
1. NULL HYPOTHESIS (H0)
2. ALTERNATE HYPOTHESIS ( H1)
There is no significant difference or relationship
Null Hypothesis
between the variables in the population.
H0: μ1=μ2
There is a significant difference or relationship
Alternative Hypothesis
between the variables in the population.
H1: μ1≠μ2
Significan
Test ce
Statistic Critical Level
Decision (2.5) Value p-value (0.012) (0.05) Decision Rule
1.96
Reject the null hypothesis.
Reject H0 Yes (for Yes (p < α) Yes
Evidence supports H1.
α=0.05)
1.96
Fail to Fail to reject the null
No (for No (p ≥ α) Yes
Reject H0 hypothesis.
α=0.05)
Accept the null hypothesis.
Accept H0 No N/A Yes (p < α) Yes There is no evidence for
H1.
Steps in Hypothesis testing
Step Description
Null Hypothesis (H0): The average time it takes to complete
a task is 10 minutes.
1. Formulate a Hypothesis
Alternative Hypothesis (Ha or H1): The average time is not
10 minutes.
2. Set up a Suitable Choose a significance level, such as α = 0.05, indicating a 5%
Significance Level (α) chance of rejecting the null hypothesis when it's true.
Select the appropriate test statistic and distribution
3. Select the Criterion
Z test, t-test , F test and Chi square X2
Collect data from a sample of tasks completion times and
4. Compute calculate the sample mean, standard deviation and standard
error
Compare the test statistic to critical value(s) or compute p-
5. Make Decision
value and compare it with α.
Make Decision
Null Hypothesis Alternative
Decision (Ho) Hypothesis (H1) Decision Rule
Accept Supported Not supported p-value > alpha
Reject Not supported Supported p-value ≤ alpha
Two types of errors in the hypothesis testing
Decision to Reject H0 Decision to Retain H0
Error Type
(Null Hypothesis) (Null Hypothesis)
Incorrectly rejecting a Correct decision
Type I Error
true null hypothesis (no error)
Correct decision Incorrectly retaining a
Type II Error
(no error) false null hypothesis
Decision
Error Type Interpretation Probability Symbol
Outcome
Reject H0 when False positive or
Type I Error α
H0 is actually true false alarm
Retain H0 when
Missed detection
Type II Error H0 is actually β
or false negative
false