Testing the Hypothesis

Session V
A fact is a simple statement that everyone believes. It is innocent,
unless found guilty. A hypothesis is a novel suggestion that no one
wants to believe. It is guilty, until found effective.

Edward Teller,
theoretical physicist
father of the hydrogen bomb
(1908–2003)
Hypothesis Formulation
Hypothesis
Testing and
Research Process
Types of hypotheses
• Null Hypothesis (H0)
• Alternative Hypothesis (Ha or H1)
• Directional Hypothesis
• Non-Directional Hypothesis
• Simple Hypothesis
• Complex Hypothesis
• Associative Hypothesis
• Causal Hypothesis
• Research Hypothesis
• Statistical Hypothesis
The Logic of Hypothesis Testing
• In classical tests of significance, two kinds of hypotheses are used.
• Null Hypothesis (H0)
• Alternate Hypotheses (HA)
Null & Alternate Hypotheses
• The null hypothesis (H0) is used for testing.
• It is a statement that no difference exists between the parameter (a measure taken by a census of the
population or a prior measurement of a sample of the population) and the statistic being compared to it
(a measure from a recently drawn sample of the population).
Alternate
• The alternative hypothesis states that a population parameter does not equal a specified value.
• Typically, this value is the null hypothesis value associated with no effect, such as zero.
• These types of alternative hypotheses correspond with two-tailed and one-tailed tests.
Null Hypothesis (H0):
• There is no significant difference in test scores between students who study for 2 hours and students who
study for 4 hours.
Alternative Hypothesis (Ha or H1):
• There is a significant decrease in anxiety levels among participants who practice mindfulness meditation
compared to those who do not.
Directional Hypothesis/Non-Directional
Hypothesis
• Directional Hypothesis: This type of hypothesis predicts the specific direction of the
relationship between variables. For example, it might suggest that one variable will
increase as the other variable increases.
• Non-Directional Hypothesis: Also known as a two-tailed hypothesis, this type does not
predict a specific direction of the relationship. It simply suggests that there will be a
significant difference or relationship, without specifying whether it will be positive or
negative.
Directional Hypothesis
• As the amount of sunlight exposure increases, the growth rate of plants will also increase.
Non-Directional Hypothesis (Two-Tailed):
• There is a significant difference in performance between male and female students on a
math test.
Simple and complex hypotheses
• Simple Hypothesis: A simple hypothesis predicts a relationship between
two variables without specifying the exact nature of that relationship. It
states that a change in one variable will cause a change in another variable.
• Complex Hypothesis: A complex hypothesis involves predicting the
relationship between three or more variables. It is more intricate than a
simple hypothesis and explores how multiple variables might interact.
Simple Hypothesis
• Increased exercise leads to improved cardiovascular health.
Complex Hypothesis
• The interaction of genetics and environmental factors contributes to the
development of certain medical conditions.
Associative and causal hypothesis
• Associative Hypothesis: This hypothesis suggests a correlation or association between two
variables. It doesn't necessarily imply causation but indicates that changes in one variable
might coincide with changes in another.
• Causal Hypothesis: A causal hypothesis proposes a cause-and-effect relationship between
variables. It suggests that changes in one variable directly influence changes in another
variable.
Associative Hypothesis
• There is a correlation between the amount of time spent studying and the final exam
scores.
Causal Hypothesis
• Higher levels of air pollution lead to an increased incidence of respiratory illnesses in urban
areas.
• Research Hypothesis:
Research & statistical hypothesis
• Research Hypothesis: A research hypothesis is a specific statement that predicts an
expected outcome of a study. It is often derived from a theory and guides the research
process.
• Statistical Hypothesis: In the context of statistical analysis, this type of hypothesis involves
statements about population parameters, such as means or proportions. Researchers use
statistical testing to determine whether these hypotheses are supported by the data.
• Based on previous studies on cognitive development, it is hypothesized that children
exposed to music education will show enhanced problem-solving skills.
Research Hypothesis
• Based on previous studies on cognitive development, it is hypothesized that children
exposed to music education will show enhanced problem-solving skills.
Statistical Hypothesis
• The mean weight of apples in a sample is equal to 150 grams.
One Tailed or Two Tailed
One-tailed tests
• A one-tailed test, or directional test, places the entire probability of an
unlikely outcome into the tail specified by the alternative hypothesis
Advantages and Disadvantages of one-tailed
tests
• Effects can exist in only one direction.
• Effects can exist in both directions but the researchers only care about
an effect in one direction.

Disadvantages
• One-tailed tests have no statistical power to detect an effect in the
other direction.
Two-tailed tests
• A two-tailed test, or non-directional test, considers two possibilities.
Advantages of two-tailed tests
• They can detect both positive and negative effects.
• Two-tailed tests are standard in scientific research where discovering
any type of effect is usually of interest to researchers.
Errors in testing hypothesis
• Type I error (α)
• Type II error (β)
Type I error
• When a Type I error (α) is committed, a true null hypothesis is rejected; the innocent person is unjustly
convicted. (A true null hypothesis is a statement that accurately represents the absence of an effect or
relationship in a research study)
• The value is called the level of significance and is the probability of rejecting the true null.
• A Type 1 error in research, also known as a false positive error or alpha error, occurs when a statistical test
or hypothesis test incorrectly rejects a null hypothesis when it is actually true.
• In other words, it's when the test incorrectly indicates the presence of an effect or relationship that
doesn't exist in reality.
• Type 1 errors are associated with making a "false alarm" or concluding that there is a significant effect
when there isn't one.
• Researchers typically set a significance level, denoted by alpha (α), which represents the probability of
making a Type 1 error. Commonly used significance levels include 0.05 (5%) or 0.01 (1%), depending on the
field and research standards. If you set a significance level of 0.05, for example, it means that you are
willing to accept a 5% chance of making a Type 1 error.
• To minimize the risk of Type 1 errors, researchers need to choose appropriate statistical tests, set
appropriate significance levels, and interpret the results cautiously. It's also crucial to replicate research
findings to confirm their validity and reduce the likelihood of Type 1 errors leading to false conclusions.
Type II error
• a Type II error (β), one fails to reject a false null hypothesis; the result is an unjust acquittal, with the guilty
person going free. (means that in a hypothesis test, researchers do not find enough evidence in their data to
reject the null hypothesis)
• A Type 2 error in research, also known as a false negative error or beta error, occurs when a statistical test or
hypothesis test fails to reject a null hypothesis when it is actually false.
• In other words, it's when the test does not detect a significant effect or relationship that does exist in reality.
• Type 2 errors are associated with failing to recognize a genuine effect or relationship.
• Type 2 errors are complementary to Type 1 errors, and they are influenced by the statistical power of a study.
• Statistical power is the probability of correctly rejecting a null hypothesis when it is false. A high statistical
power means a lower chance of making a Type 2 error.
• Researchers often aim to balance the risk of Type 1 errors (false positives) and Type 2 errors (false negatives)
when designing experiments and conducting hypothesis tests.
• Increasing the sample size, using more sensitive measurement techniques, or adjusting the significance level
can help reduce the risk of Type 2 errors but may increase the risk of Type 1 errors.
• The choice of significance level and sample size should be made based on the specific research goals and the
importance of avoiding Type 1 or Type 2 errors in a given context.
Procedure for testing hypothesis
1. State the null hypothesis
2. Choose the statistical test
3. Select the desired level of significance
4. Compute the calculated difference value
5. Obtain the critical test value
6. Interpret the test.
State the null hypothesis
• Although the researcher is usually interested in testing a hypothesis of
change or differences, the null hypothesis is always used for statistical
testing purposes.
Choose the statistical test.
• To test a hypothesis, one must choose an appropriate statistical test.
• There are many tests from which to choose, and there are at least
four criteria that can be used in choosing a test.
• One is the power efficiency of the test.
Select the desired level of significance
• The choice of the level of significance should be made before we
collect the data.
• The most common level is .05, although .01 is also widely used.
• Other levels such as .10, .025, or .001 are sometimes chosen.
• The exact level to choose is largely determined by how much risk one
is willing to accept and the effect that this choice has on b risk.
• The larger the a, the lower is the b.
Compute the calculated difference value
• After the data are collected, use the formula for the appropriate
significance test to obtain the calculated value.
• Although the computation typically results from a software program,
we illustrate the procedures in this chapter to help you visualize what
is being done.
Obtain the critical test value
• After researchers compute the calculated t , 2 , or other measure,
They must look up the critical value in the appropriate table for that
distribution (or it is provided with the software calculation).
• The critical value is the criterion that defines the region of rejection
from the region of acceptance of the null hypothesis.
Interpret the test
• For most tests if the calculated value is larger than the critical value,
researchers reject the null hypothesis and conclude that the
alternative hypothesis is supported (although it is by no means
proved).
• If the critical value is larger, researchers conclude they have failed to
reject the null.
 Case Study
Marcus Thomas LLC recently conducted an online survey to help Troy-Bilt®, a
leading manufacturer of lawn and garden equipment, understand what
elements in television advertising help build recall and consideration,
specifically within the lawn and garden category. Marcus Thomas formulated
the hypothesis that “consumers who are in the market to purchase a product
process television advertising differently than those who are not in the
market for lawn and garden equipment,” shared Edwige Winans, associate
director of research for Marcus Thomas. Carefully selected participants were
segmented into groups of equal size, based on whether or not they were
planning to purchase lawn or garden equipment (in-market group vs. out-of-
market group). To control for category-specific reactions, Marcus Thomas also
recruited a control group, those who were in the market for paint and those
who were not.
One thousand respondents were asked to watch an eight minute segment of the
DIY Network show “House Crashers” on their computer (to simulate television
watching). Four 30-second television ads were embedded following the first four
minutes of the show (two for lawn or garden equipment and two for paint),
followed by four additional minutes of the show. The study was blind, in that
respondents did not know its purpose. Ad order was randomized for each
respondent to eliminate order bias. In a subsequent online survey, respondents
provided both aided and unaided recall measurements of the television spots (to
assess which elements in the ads were most memorable), as well as information on
how the ads affected perceptions of each brand featured and brand awareness.
Several red-herring questions were introduced to ensure participants would not be
swayed by the perceived purpose of the study or the merchandise category.
• The test did showcase unique differences in the way consumers process
television advertising based on whether they are in-market or not-in-market
for lawn and garden products.
Queries?