HYPOTHES
IS
TESTING
PRESENTED BY: JEREMIAH T. CARLOS
OUTLINE
Introduction and Significance
Hypothesis and Hypothesis Testing Defined
Characteristics of a Good Hypothesis
Functions of the Hypothesis
Types of Hypotheses
Alpha and Beta
p-level
Type I and Type II Errors
Legal Analogy
Hypothesis Testing Flow Chart
References
INTRODUCTION AND SIGNIFICANCE
Almost every great step in the history of science has
been made by the anticipation of nature, that is, by the
invention of hypotheses which, though verifiable, often
had very little foundation to start with.
T. H. Huxley
Hypotheses are important because
they bring clarity, specificity and focus to a research
problem
they narrow the field of investigation
they tell a researcher what specific information to collect
they provide greater focus to the research problem
(Kumar, 2005)
HYPOTHESIS AND
HYPOTHESIS TESTING DEFINED
Hypothesis is
a conjectural statement of the relationship between
two or more variables (Kerlinger, 1986).
a tentative explanation for certain behaviors,
phenomena, or events which have occurred or will
occur (Gay, 1976).
a testable statement of a potential relationship
between two or more variables (McGuigan, 1978).
the most specific statement of a problem that
states the researchers expectations concerning
the relationship between the variables in the
research problem (Sevilla, et al., 1988).
HYPOTHESIS AND
HYPOTHESIS TESTING DEFINED
Hypothesis Testing
is
statistically analyzing data from a sample in
order to make assumptions about a
population (Heffner, 2004).
determining how much error is associated
with sample data and then comparing the
amount of estimated error with an acceptable
level of error (Heffner).
sometimes referred to as the NeymanPearson framework, in honor of Jerzy Neyman
and Egon Pearson (Siegrist, 1997).
CHARACTERISTICS OF
A GOOD HYPOTHESIS
Good Hypothesis
A
should
be
1. reasonable
2. stating in definite terms the relationship
between variables. (e.g. cause and effect,
correlation, measures of difference, etc.)
3. testable; and
4. following the findings of previous studies.
FUNCTIONS OF THE HYPOTHESIS
The
1.
2.
3.
Hypothesis
introduces the researchers thinking at
the start of the study.
structures the next stages or procedures
of the study.
helps to provide the format for the
presentation, analysis and interpretation
of the data in the thesis.
TYPES OF HYPOTHESES
1.
Null Hypothesis (H0 )- states that there is no
existence of effect, interaction, relationships or
difference between two groups.
H0 : 1 = 2
Examples:
There
is no significant effect of three feeding
schemes on the weight of malnourished eight-yearold children.
The mean for the boys equals that of the girls on the
Torrance Verbal Test of originality.
Sex is independent of positiveness or negativeness
towards some concepts among freshman high school
boys and girls.
TYPES OF HYPOTHESES
2.
Alternative Hypothesis (H1 )- the
operational statement of the research
hypothesis.
H1 : 1 2 or H1 : 1 < 2 or H1 : 1 > 2
A.
B.
Deductive Hypothesis expectation based
on a theory
Inductive Hypothesis based on
observations of behavior.
TYPES OF HYPOTHESES
3.
Non-Directional Hypothesis does not state a
direction or nature of difference; involves two-tailed
tests.
H0 : 1 = 2 or H1 : 1 2
4.
Directional Hypothesis an alternative hypothesis
that predicts the results of one condition will be
greater or less than another, rather than a
prediction that they will simply differ; involves onetailed tests (White and McBurney, 2013).
H1 : 1 < 2 or H1 : 1 > 2
TYPES OF HYPOTHESES
Source: Business Research Methods, 6th Ed., Cooper and Schindler, 1998
ALPHA and BETA
Alpha () refers to the probability of making
a Type I Error. It is determined before the
research is conducted (typically .05 or .01).
Beta () represents the possibility of making
a Type II Error and is called Power; unknown
prior to research although there are statistics
that attempt to estimate it (Heffner, 2004).
p-level
p the probability of making an error
No research is ever 100% error free, such that
the researcher sets the alpha () level, or the
acceptable level of error, prior to analyzing data.
Hence, if the probability of error (p) is less than
the accepted level of error (), then the research
is significant because p < .
Examples:
The results are significant (p<.05); or
The results are significant at the .05 level.
TYPE I AND TYPE II ERRORS
Type I Error rejecting a null hypothesis
when it is true.
Type II Error accepting a null hypothesis
when it is false. (White and McBurney, 2013)
STATE OF NATURE
DECIS
ION
Null hypothesis is
true
Alternative
hypothesis
is true
Accept null
hypothesis
Correct decision
Wrong decision
Type II Error ()
Reject null
hypothesis
Wrong decision
Type I Error ()
Correct decision
LEGAL ANALOGY
Source: Business Research Methods, 6th Ed., Cooper and Schindler, 1998
HYPOTHESIS TESTING
FLOW CHART
Develop research hypothesis & null hypothesis
Develop research hypothesis & null hypothesis
Set alpha (usually .05)
Set alpha (usually .05)
Calculate power to determine the sample
Calculate power to determine the sample
size
size
Collect data & calculate statistic and p
Collect data & calculate statistic and p
P < .05
P < .05
Compare p to alpha (.05)
Compare p to alpha (.05)
Reject null hypothesis
Reject null hypothesis
P > .05
P > .05
Fail to reject null hypothesis
Fail to reject null hypothesis
Source: Hypothesis Testing and Inferential Statistics, Masuda, 2000
REFERENCES
Cooper and Schindler. Business Research Methods. Sixth Edition.
Singapore: McGrawHill, 1998.
Heffner, Christopher. Statistics Primer: Hypothesis Testing. Heffner
Media Group, Inc., 2011. Web. 22 November 2013. <
http://allpsych.com/stats/unit4/22.html>.
Kumar, Ranjit. Research Methodology. Second Edition. New Delhi: Sage
Publications,
2005.
Masuda, Takahiko. Chapter 8: Hypothesis Testing and Inferential
Statistics. University of Michigan, 2000. Web. 22 November 2013.
<
http://umich.edu/~psycours/381/masuda/381notes/ch8/index.htm>.
Sevilla, Ochave and others. An Introduction to Research Methods. First
Edition. Manila:
Rex Bookstore, 1988.
Siegrist, Kyle. Hypothesis Testing: Introduction. University of Alabama
Huntsville, 1997. Web. 22 November 2013.
<
http://www.math.uah.edu/stat/hypothesis/Introduction.html>.
White and McBurney. Research Methods. 9th International Edition.
END OF PRESENTATION.
THANK YOU FOR YOUR ATTENTION.
