What is Hypothesis
A hypothesis is a claim( assertion,
statement, belief or assumption)
about an unknown population
parameter value.
A Hypothesis is the statement or an
assumption about relationships
between variables.
Interesting Hypothesis
Question Hypothesis
Does a training program in People who take a driver safety course will
driver safety result in a have a lower accident rate than those who
decline in accident rate? do not take the course.
Who is better in math, men Men are better at math than women.
or women?
What is the relationship Cell phone use is higher for younger adults
between age and cell phone than for older adults.
use?
Is there a relationship Income increases with years of education.
between education and
income?
Can public education reduce The number of AIDS cases is inversely
the occurrence of AIDS? related to the amount of public education
about the disease.
Rationale of Hypothesis Testing
To test theories: Hypothesis testing allows researchers to test the
validity of theories and to determine if they are supported by
empirical evidence.
To make objective decisions: Hypothesis testing provides a
structured approach to making objective decisions about the data and
helps to reduce the influence of subjective biases or personal beliefs.
To draw conclusions: Hypothesis testing allows researchers to draw
conclusions about the population being studied based on a sample of
data.
To identify relationships: Hypothesis testing can help identify
relationships between variables and can be used to determine the
strength and direction of those relationships.
Rationale of Hypothesis Testing
To make predictions: Hypothesis testing can be used
to make predictions about future observations or data.
To improve accuracy: Hypothesis testing can help
improve the accuracy of research findings by
identifying and correcting for errors or biases in data
collection or analysis.
To increase confidence: Hypothesis testing provides a
way to quantify the level of confidence that can be
placed in research findings, and can help to increase the
credibility of the research.
Qualities of Good Hypothesis
Testable: A good hypothesis is testable and can be
supported or refuted through empirical evidence.
Specific: A good hypothesis is specific and focused on a
particular relationship or phenomenon, rather than being
too general or vague.
Falsifiable: A good hypothesis is falsifiable, meaning that
it can be shown to be false if it is not supported by the data.
Relevant: A good hypothesis is relevant to the research
question and is based on existing knowledge and
observations.
Qualities of Good Hypothesis
Clear and concise: A good hypothesis is clear and concise,
with well-defined variables and a clear statement of the
predicted relationship or outcome.
Logical: A good hypothesis is logically sound and based on a
clear understanding of the underlying theory or conceptual
framework.
Realistic: A good hypothesis is realistic and feasible, given
the available resources and data.
Testable with available data: A good hypothesis can be
tested using the available data and research methods, without
requiring additional data or resources.
Types of
•
Hypothesis
Null Hypothesis (H ) 0
• Alternative Hypothesis (Ha or H1)
Each of the following statements is an example of a null
hypothesis and alternative hypothesis.
Null and Alternative Hypothesis
Null Hypothesis: This is the default hypothesis that there is no
significant difference or relationship between the variables being
studied. It is usually denoted by "H0". For example, "There is no
significant difference in test scores between students who studied
with flashcards and those who studied with textbooks."
Alternative Hypothesis: This is the opposite of the null
hypothesis and suggests that there is a significant difference or
relationship between the variables being studied. It is usually
denoted by "Ha" or "H1". For example, "There is a significant
difference in test scores between students who studied with
flashcards and those who studied with textbooks.“ Alternative
Hypothesis is the research Hypothesis
Developing Null and Alternative Hypotheses
Alternative Hypothesis as a Research Hypothesis
• Example:
A new teaching method is developed that is
believed to be better than the current method.
• Alternative Hypothesis:
The new teaching method is better.
• Null Hypothesis:
The new method is no better than the old method.
Developing Null and Alternative
Hypotheses
Alternative Hypothesis as a Research Hypothesis
• Example:
A new sales force bonus plan is developed in an
attempt to increase sales.
• Alternative Hypothesis:
The new bonus plan increase sales.
• Null Hypothesis:
The new bonus plan does not increase sales.
Developing Null and Alternative
Hypotheses
Alternative Hypothesis as a Research Hypothesis
• Example:
A new drug is developed with the goal of lowering
blood pressure more than the existing drug.
• Alternative Hypothesis:
The new drug lowers blood pressure more than
the existing drug.
• Null Hypothesis:
The new drug does not lower blood pressure more
than the existing drug.
Examples
Null Hypothesis Alternative Hypothesis
Exercise is unrelated to weight
loss. Exercise leads to weight loss
Exposure to classical music Exposure to classical music
has no effect on IQ score. increases IQ score
Extrovert and introverts are Extroverts are healthier than
equally healthy. introverts
People exposed to sensitivity
training are no more tolerant
than those not exposed to Sensitivity training reduces racial
sensitivity training. bias
Steps in Hypothesis Testing
Step-1: State the Null Hypothesis(H0) & Alternative Hypothesis (H1): The first step is to
state the null hypothesis (H0) and alternative hypothesis (Ha) clearly. For example, H0:
There is no significant difference in the mean height of males and females in the population.
Ha: There is a significant difference in the mean height of males and females in the
population.
Step-2 : Selecting a significance level: The level of significance (alpha) determines the risk
of rejecting the null hypothesis when it is actually true. The common alpha value is 0.05,
which means that there is a 5% chance of rejecting the null hypothesis when it is actually
true.
Step-3: Deciding the Test Statistic: The test statistic is a numerical value that is calculated
based on the data and the null hypothesis. The choice of test statistic depends on the type of
data and the hypothesis being tested. For example, if the data is normally distributed, the t-
test or z-test can be used.
Step-4: Calculation of the probability: The p-value is the probability of obtaining the
observed test statistic, or a more extreme one, if the null hypothesis is true. If the p-value is
less than the level of significance, the null hypothesis is rejected.
Step-5: Comparing the probability: Now comparing the probability thus calculated with
the specified value for α , the significance level. If the calculated probability is equal to or
smaller than the α value then reject the null hypothesis but if the calculated probability is
LIMITATIONS OF THE TESTS OF HYPOTHESIS
Sampling error: The test of hypothesis is based on a sample of data
that is assumed to be representative of the population. However,
sampling error can occur, which means that the sample may not
accurately represent the population. This can lead to incorrect
conclusions.
Assumptions: The test of hypothesis is based on certain assumptions
about the data, such as normality and independence. If these
assumptions are not met, the results of the test may not be accurate.
Type I and Type II errors: The test of hypothesis is based on a
significance level, which determines the risk of rejecting the null
hypothesis when it is actually true (Type I error) and the risk of
failing to reject the null hypothesis when it is actually false (Type II
error). Researchers must balance these risks and choose an
appropriate significance level.
LIMITATIONS OF THE TESTS OF HYPOTHESIS
Effect size: The test of hypothesis does not provide
information about the size of the effect. Even if a
significant difference is found, it may not be practically
or clinically meaningful.
Multiple comparisons: When testing multiple
hypotheses or comparing multiple groups, the probability
of making a Type I error increases. Researchers must
adjust the significance level or use appropriate methods
to control for multiple comparisons.
Ethics: The test of hypothesis can sometimes be used in
unethical ways, such as selectively reporting results or
manipulating data to obtain a desired outcome.
LIMITATIONS OF THE TESTS OF
HYPOTHESIS
(i) The tests should not be used in a mechanical fashion. It should be kept in
view that testing is not decision-making itself; the tests are only useful aids
for decision-making. Hence “proper interpretation of statistical evidence is
important to intelligent decisions.”
(ii) Test do not explain the reasons as to why does the difference exist, say
between the means of the two samples. They simply indicate whether the
difference is due to fluctuations of sampling or because of other reasons but
the tests do not tell us as to which is/are the other reason(s) causing the
difference.
(iii) Results of significance tests are based on probabilities and as such
cannot be expressed with full certainty. When a test shows that a difference is
statistically significant, then it simply suggests that the difference is probably
not due to chance.
(iv) Statistical inferences based on the significance tests cannot be said to be
entirely correct evidences concerning the truth of the hypotheses. This is
specially so in case of small samples where the probability of drawing erring
inferences happens to be generally higher. For greater reliability, the size of
samples be sufficiently enlarged.
Precautions taken by researcher while drawing inferences
from hypothesis test
Consider the context: Results of a hypothesis test should always be interpreted in the context of the
research question and the data collected. Researchers should not generalize their results beyond the
scope of their study.
Check assumptions: Researchers should check the assumptions of the statistical test used to make sure
that they are met. If the assumptions are not met, the results may not be valid.
Look at effect sizes: Hypothesis tests do not provide information about the size of the effect.
Researchers should consider effect sizes to determine the practical significance of their results.
Avoid multiple comparisons: Multiple comparisons can increase the probability of making a Type I
error. Researchers should adjust their significance levels or use appropriate methods to control for
multiple comparisons.
Use appropriate statistical tests: Researchers should use appropriate statistical tests for their research
questions and data. Using the wrong test can lead to inaccurate results.
Acknowledge limitations: Researchers should acknowledge the limitations of their study and the test
used. This includes issues such as sampling bias, measurement error, and the limitations of statistical
tests.
ERRORS IN HYPOTHESIS TESTING
Type I Error
Type I error occurs when we reject the null hypothesis when it is true.
It is also known as a false positive error. The probability of making a
Type I error is equal to the level of significance (alpha), which is
typically set at 0.05. In other words, if the level of significance is 0.05,
there is a 5% chance of making a Type I error.
For example, suppose a medical researcher is testing a new drug and
wants to see if it is effective in reducing blood pressure. The null
hypothesis is that the drug has no effect on blood pressure. If the
researcher rejects the null hypothesis and concludes that the drug is
effective, even though it is not, this is a Type I error.
Type II Error
Type II error occurs when we accept the null hypothesis though it is false. It is
also known as a false negative error. The probability of making a Type II
error is denoted by beta (β). The power of a statistical test is equal to 1 - β. In
general, the higher the power, the lower the chance of making a Type II error.
For example, continuing with the same example of the medical researcher
testing a new drug, if the researcher fails to reject the null hypothesis and
concludes that the drug is not effective, even though it is, this is a Type II error.
The risk of making Type I and Type II errors is related. If the level of
significance (alpha) is decreased, the risk of making a Type I error is
decreased, but the risk of making a Type II error is increased. Similarly, if
the sample size is increased, the risk of making a Type II error is
decreased, but the risk of making a Type I error is increased.
Simple and Composite Hypothesis
Simple hypothesis: A simple hypothesis is a hypothesis that makes a precise statement about a
population parameter. It is a hypothesis that specifies a particular value or range of values for the
population parameter being tested. For example, a simple hypothesis could be "The mean height
of adult females in the United States is 65 inches."
Composite hypothesis: A composite hypothesis is a hypothesis that allows for more than one
possible value or range of values for the population parameter being tested. It is a hypothesis
that specifies a range of values or a set of values for the population parameter being tested. For
example, a composite hypothesis could be "The mean height of adult females in the United
States is not equal to 65 inches." This composite hypothesis includes all possible values of the
mean height except for 65 inches.
In hypothesis testing, the null hypothesis is usually a simple hypothesis, while the alternative
hypothesis is often a composite hypothesis. The null hypothesis is the hypothesis that is being
tested against the alternative hypothesis. It is usually the hypothesis that there is no difference or no
effect in the population, or that any difference or effect is due to chance. The alternative hypothesis
is the hypothesis that the researcher is interested in proving, and it is usually a statement that
contradicts the null hypothesis.
One Tailed and Two Tailed Test
One-tailed test: A one-tailed test is a hypothesis test in which the alternative hypothesis is directional,
meaning it specifies either an increase or a decrease in the population parameter being tested. For
example, a one-tailed test could be used to test whether a new drug increases the average response time
of participants in a cognitive task. The alternative hypothesis in this case would be that the drug
increases response time, and the null hypothesis would be that there is no difference in response time
between the drug and placebo groups. A one-tailed test is typically used when the researcher has a
strong prior belief about the direction of the effect.
Two-tailed test: A two-tailed test is a hypothesis test in which the alternative hypothesis is non-
directional, meaning it simply states that there is a difference or an effect in the population parameter
being tested, without specifying the direction. For example, a two-tailed test could be used to test
whether a new drug has any effect on the average response time of participants in a cognitive task. The
alternative hypothesis in this case would be that there is a difference in response time between the drug
and placebo groups, and the null hypothesis would be that there is no difference. A two-tailed test is
typically used when the researcher does not have a strong prior belief about the direction of the effect.
In both one-tailed and two-tailed tests, the level of significance (alpha) is set in advance, and the p-
value is calculated based on the sample data. If the p-value is less than or equal to the level of
significance, the null hypothesis is rejected, and the alternative hypothesis is accepted. If the p-value is
greater than the level of significance, the null hypothesis is not rejected.
Acceptance and Rejection Region
Acceptance region: The acceptance region is the range of values of the test statistic
that indicates that the null hypothesis cannot be rejected at the given level of
significance. The acceptance region is usually centered around the null hypothesis
value and can be defined by setting a critical value or a range of values for the test
statistic. If the calculated test statistic falls within the acceptance region, the null
hypothesis is not rejected, and the result is considered inconclusive.
Rejection region: The rejection region is the range of values of the test statistic that
indicates that the null hypothesis can be rejected at the given level of significance.
The rejection region is usually defined by setting a critical value or a range of values
for the test statistic that is unlikely to occur if the null hypothesis is true. If the
calculated test statistic falls within the rejection region, the null hypothesis is
rejected, and the result is considered significant.
The choice between the acceptance and rejection regions depends on the level of
significance, which is usually set at 5% or 1%.
Descriptive statistics: Descriptive statistics are used to describe and
summarize the main features of a dataset. These statistics are used to
provide a clear understanding of the characteristics of the data,
including its central tendency, variability, and distribution.
Descriptive statistics can be used to calculate measures such as
mean, median, mode, range, variance, and standard deviation.
Inferential statistics: Inferential statistics are used to draw
conclusions and make inferences about a population based on a
sample of data. Inferential statistics are used to test hypotheses and
make predictions about the population based on the data collected
from a sample. The aim of inferential statistics is to generalize the
findings from the sample to the larger population.
Important statistical measures often used to summarize survey/research data
Central tendency: This refers to the measure of the middle or central point of a
dataset. The three main measures of central tendency are mean, median, and mode.
Dispersion: This refers to the measure of how spread out the data is. Common
measures of dispersion include range, variance, and standard deviation.
Skewness and kurtosis: Skewness is a measure of the asymmetry of the
distribution, while kurtosis measures the degree of peakedness or flatness of a
distribution.
Correlation: This refers to the measure of the relationship between two variables.
Correlation coefficients such as Pearson's correlation coefficient can be used to
quantify the strength and direction of the relationship between two variables.
Regression analysis: This refers to the statistical technique used to model the
relationship between a dependent variable and one or more independent variables.
What is ANOVA
ANOVA stands for Analysis of Variance. It is a statistical method used to
compare the means of two or more groups to determine if there is a significant
difference between them.
ANOVA is used when there are three or more groups, and it compares the
variance within groups to the variance between groups. If the variance
between groups is much larger than the variance within groups, it suggests
that there is a significant difference between the groups.
ANOVA can be used in a variety of situations, such as in experimental designs
to test the effect of a treatment on a dependent variable, or in observational
studies to compare the means of different groups.
There are different types of ANOVA, including one-way ANOVA, two-way
ANOVA, and repeated measures ANOVA.
Assumptions of ANOVA
Normality: Sample are drawn from normally
distributed population
Homogeneity of variance: Populations from
which samples are drawn have equal variance.
Independence: Samples are randomly drawn from
populations and are independent of each other,.
Random sampling: The samples are randomly
selected from the population. This means that each
observation has an equal chance of being selected.
Additive property of ANOVA
The additive property of ANOVA refers to the fact that the total variance
in a dataset can be broken down into components that are due to different
sources of variation.
In a one-way ANOVA, for example, the total variance in the dependent
variable can be decomposed into the variance between groups and the
variance within groups.
This property allows us to identify the sources of variation that are most
responsible for the differences between groups.
By comparing the variance between groups to the variance within
groups, ANOVA allows us to determine whether the differences between
groups are due to chance or to a real effect of the independent variable.
ANOVA vs. Sampling
First, ANOVA allows us to compare the means of more than two groups, whereas
sampling is typically used to compare the means of only two groups. This means
that ANOVA is more versatile and can be used in a wider range of research
settings.
Second, ANOVA allows us to test the significance of differences between groups
while controlling for other sources of variation, such as individual differences or
extraneous variables. This is not possible with simple sampling techniques, which
may be confounded by other sources of variation.
Finally, ANOVA allows us to estimate the effect size of the differences between
groups, which provides a more meaningful measure of the strength of the effect
than simply reporting whether or not the differences are statistically significant.
This can help researchers to interpret their results more accurately and to draw
more meaningful conclusions from their data.
One-way ANOVA: One-way ANOVA is used to determine if there is a significant
difference between three or more groups. It involves comparing the means of each
group and their variance. The null hypothesis assumes that the means of all groups are
equal. If the p-value is less than the chosen significance level (typically 0.05), the null
hypothesis is rejected and it can be concluded that there is a significant difference
between at least one of the groups. E.g Testing whether there is a difference in exam
scores between three different schools
Two-way ANOVA: Two-way ANOVA is used when there are two independent
variables or factors. It involves testing the main effects of each variable and their
interaction. The main effect of each variable refers to the difference between the
means of each level of that variable, while the interaction effect refers to the
combined effect of both variables. The null hypothesis assumes that there is no
significant difference between groups due to either factor or their interaction. If the p-
value is less than the chosen significance level, the null hypothesis is rejected, and it
can be concluded that there is a significant difference due to at least one factor or
their interaction. E.g. Testing whether there is a difference in exam scores between
different schools and different genders of students, as well as whether there is an
interaction between these two factors
One-Way ANOVA Two-Way ANOVA
Number of Factors One factor Two factors
Differences in means between three or more Differences in means due to two factors and
Test for
groups their interaction
Only one main effect - the difference between Two main effects - the difference between
Main Effects
group means levels of each factor
Interaction No interaction effect tested Interaction effect tested
No significant difference due to factors or their
Null Hypothesis All group means are equal
interaction
F = ((Between-group variance + Within-group
F = (Between-group variance / Within-group variance due to factor A + Within-group
Formula
variance) variance due to factor B) / Within-group
variance)
df between factors = a-1, df within factors =
Degrees of Freedom df between = k-1, df within = N-k, total df = N-1
(a-1)(b-1), total df = ab-1
Independence of observations, normality of the
Independence of observations, normality of the
Assumptions population, equal variances, no interaction
population, equal variances
between factors
Testing whether there is a difference in exam
Testing whether there is a difference in exam scores between different schools and different
Example
scores between three different schools genders of students, as well as whether there is
an interaction between these two factors
