The null hypothesis states that there is no association between the predictor and outcome
variables in the population (There is no difference between tranquilizer habits of patients with
attempted suicides and those of age- and sex- matched ―control‖ patients hospitalized for other
diagnoses). The null hypothesis is the formal basis for testing statistical significance. By starting
with the proposition that there is no association, statistical tests can estimate the probability that
an observed association could be due to chance.
The proposition that there is an association — that patients with attempted suicides will report
different tranquilizer habits from those of the controls — is called the alternative hypothesis. The
alternative hypothesis cannot be tested directly; it is accepted by exclusion if the test of statistical
significance rejects the null hypothesis.
Types of hypotheses-
a.Inductive is a generalization based on specific observations.
b.Deductive is derived from theory and provides evidence that supports, expands, or contradicts
c.Nondirectional - states that relation or difference between variables exists.
d.Directional - states the expected direction of the relation or difference.
e.Null - states that there is no significant relation or difference between variables.
Daniel W. W. In: Biostatistics. 7th ed. New York: John Wiley and Sons, Inc; 2002. Hypothesis
testing; pp. 204–294.
As mentioned previously, a hypothesis is a tool of
It is a tentative and formal prediction about the relationship
between two or more variables in the population being
studied, and the hypothesis translates the research
question into a prediction of expected outcomes.
So…a hypothesis is a statement about the relationship
between two or more variables that we set out to prove or
disprove in our research. study.
To be complete the hypothesis must include three
The relationship between the variables.
A hypothesis should be:
stated clearly using appropriate terminology;
a statement of relationships between variables;
limited in scope (focused).
Examples of a hypothesis are:
Health Education programmes influence the number
of people who smoke.
Newspapers affect people's voting pattern.
Attendance at lectures influences exam marks.
Diet influences intelligence.
Types of hypotheses
There are different types of hypotheses:
Simple hypothesis - this predicts the relationship
between a single independent variable (IV) and a
single dependent variable (DV)
Lower levels of exercise postpartum (IV) will be
associated with greater weight retention (DV).
IV = independent variable
D V = dependent variable
Complex hypothesis - this predicts the relationship
between two or more independent variables and two
or more dependent variables.
1. Example of a complex multiple independent variable
Low risk pregnant women (IV) who:
value health highly ;
believe that engaging in health promoting
behaviours will result in positive outcomes;
perceive fewer barriers to health promoting
are more likely than other women to attend
pregnancy-related education programmes (DV).
2. Example of a complex multiple dependent variable
The implementation of an evidence based protocol for
urinary incontinence (IV) will result in (DV):
decreased frequency of urinary incontinence
decreased urine loss per episode;
decreased avoidance of activities among
women in ambulatory care settings.
Hypotheses can be stated in various ways as long as the
researcher specifies or implies the relationship that will be
Lower levels of exercise postpartum are associated
with greater weight retention.
There is a relationship between level of exercise
postpartum and weight retention.
The greater the level of exercise postpartum, the
lower the weight retention.
Women with different levels of exercise postpartum
differ with regard to weight retention.
Weight retention postpartum decreases as the
woman's level of exercise increases.
Women who exercise vigorously postpartum have
lower weight retention than women who do not.
These are usually derived from theory.
They may imply that the researcher is intellectually
committed to a particular outcome.
They specify the expected direction of the relationship
between variables i.e. the researcher predicts not only the
existence of a relationship but also its nature.
Used when there is little or no theory, or when findings of
previous studies are contradictory.
They may imply impartiality.
Do not stipulate the direction of the relationship.
Associative and causal hypotheses
Propose relationships between variables - when one
variable changes, the other changes.
Do not indicate cause and effect.
Propose a cause and effect interaction between two or
The independent variable is manipulated to cause effect
The dependent variable is measured to examine the effect
the independent variable.
A format for stating causal hypotheses is:
The subjects in the experimental group who are exposed
to the independent variable demonstrate greater change,
as measured by the dependent variable, than do the
subjects in the control group who are not exposed to the
These are used when the researcher believes there is no
relationship between two variables or when there is
inadequate theoretical or empirical information to state a
Null hypotheses can be:
simple or complex;
associative or causal.
Contain variables that are measurable or able to be
They need to predict a relationship that can be 'supported'
supported' based on data collection and analysis.