2. HYPOTHESIS TESTING
Hypothesis tests are designed to statistically determine
the probability that a certain hypothesis would be
considered accepted at a certain significance level or
not.
The purpose of hypothesis testing is to determine
whether there is enough statistical evidence in favor of a
hypothesis.
3. Research Question ?
“OR”
Hypothesis?
• Research starts with a research question or hypothesis.
• These are tentative generalization regarding a relationship between
two or more variables.
• There is slight difference between the research question and
hypothesis. Hypothesis makes prediction about experimental
outcome, research question does not.
4. • Research question may take the form of a simple question about
the relationship between two or more variables.
• It should be as specific as possible. In some cases, you may
make two or more research questions to cover a complex topic.
• Examples:
• How much hours per day students spend on social networking
sites?
• Do the college students with firm career goals achieve higher
GPAs than the students who do not have firm career goals?
RESEARCH QUESTION
5. HYPOTHESIS
A tentative generalization regarding the relationship
between two or more variables that predict an
experimental outcome.
Example:
The level of attention paid to radio commercials is
positively related to the amount of recall of
commercial.
6. HYPOTHESIS
• A hypothesis is a statement that can be proved or disproved.
• Hypothesis translates the research question into a prediction of
expected outcomes.
1. How much hours per day
students spend on social
networking sites?
2. Do the college students with firm
career goals achieve higher
GPAs than the students who do not
have firm career goals?
1. Students who spend more time
on social networking sites are
considered to be socially alienated.
2.It is hypothesized that college
students who have firm career
goals achieve higher GPAs than
those who do not have firm career
goals.”
7. CRITERIA FOR GOOD HYPOTHESIS
Should be compatible with current knowledge in the area
It should follow logical consistency
A good hypothesis must be stated briefly and clearly
A hypothesis should be testable
8. NULL HYPOTHESIS
(Hypothesisof no difference)
• It asserts that the statistical difference or relationship being
analyzed are due to chance or random error.
• The null hypothesis (H0) is the logical alternative to the research
hypothesis (H1).
Research Hypothesis: The level of attention paid to
radio commercials is positively related to the amount of
recall of commercial.
Null Hypothesis: The level of attention paid to radio
commercials is not positively related to the amount of
recall of commercial.
9. BENEFITS OF HYPOTHESIS
4 MAJOR BENEFITS
• PROVIDE DIRECTION OF STUDY
• ELIMNINATE TRIAL AND ERROR
RESEARCH
• HELP RULE OUT INTERVENING AND
CONFOUNDING VARIABLES
• ALLOW FOR QUANTIFICATION OF
VARIABLES
10. TESTING HYPOTHESIS
• In hypothesis testing or significance testing, the
researcher either rejects or accepts the null hypothesis.
• That’s is, if H0 is accepted, it is assumed that H1 is
rejected.
• If H0 is rejected,H1 must be accepted.
11. Example
• Consider a study of the math competency of a group of subjects who
receive a special learning treatment, possibly a series of television
programs on mathematics. It is hypothesized that this group after
viewing the programs, will have scores on a standardized math test
significantly different from those of the remainder of the population that
has not seen the programs.
Research Hypothesis H1=
Null Hypothesis H0=
Students who have received special
mathematics learning TV programs will
score high on Math competency test.
Special mathematics learning TV programs
makes no impact on score obtained in Math
competency test.
12. Significance Level
• To determine the statistical significance of a research study,
the researcher must set a probability level or significance
level, against which the null hypothesis is tested.
If the results of the study indicate a probability lower than
this level, the researcher can reject the null hypothesis.
If the results of the study indicate a probability higher
than this level, the researcher must support the null
hypothesis.
13. One Tail Test:
A one-tail test predicts that the results
will fall in only one direction- either
positive or negative.
Two Tail Test:
It does not predict any direction. Two
tail tests are generally used when little
information is available about the
research area.
14. One -Tailed Alternative=
Men earn more than women.
.
Two-Tailed Alternative=
Wages of men and women are not
equal.
Research Hypothesis H1 =
Wages of men and women are equal.
Null Hypothesis H0 =
Men earn more than women.
16. Critical region/region of rejection
In a theoretical sampling distribution,
the proportion of the area in which the
null hypothesis is rejected is called the
region of rejection. This area is defined
by the level of significance chosen by
the researcher. If the .05 level of
significance is used, then 5% of the
sampling distribution becomes the
critical region. Conversely, the null
hypothesis is retained in the region
between the two rejection values.
18. Finding Z (Standard score) value
• It follows that corresponding z values that define the
region of rejection are those that cut off 47.5% (.4750) of
the area from “m” to each end of the tail.
• To find this z-value, we use table:
19. Table 3: Areas under the Normal Curve. Proportion of Area under the Normal
curve between the Mean and a z distance from the Mean.
20. Determination of region of rejection
These values are used to determine the region of rejection.
- 1.96 αm + μ = lower boundary
+ 1.96 αm + μ = upper boundary
Where:
αm = Standard deviation of the distribution
μ = Population mean
Assume population mean for math competency is 100 and standard deviation is
15.
Mean math’s competency scores are:
-1.96(15) + 100 = 70.60
+1.96(15) +100 = 129.40
22. CONCLUSION
• If the research study produces a result between 70.60 and
129.40, a result between 70.60 and 129.40, the null
hypothesis can not be rejected. TV programs had no
significant effect on math competency levels.