BASIC

INFERENTIAL

TESTS FOR

RESEARCH
 BASIC
INFERENTIAL
TESTS FOR
R E S E A R C H
There are two types of statistical
tests:(a) parametric, and (b)
nonparametric
In parametric tests, the population
is normally distributed.
BASIC Parametric
tests involve interval
INFERENTIAL or ratio data;
but it can be used for continuous
TESTS FOR
data obtained from a population
R E S E A R C H
with a normal distribution. Hence,
these are powerful and useful
statistics for research, particularly
for testing the null hypothesis.
 TWO WAYS TO MAKE
A STATISTICAL
ESTIMATIO INFERENCE
N
• Estimation is a crude method of making
population inferences.
• It makes inferences about the population
using sampling distribution tools such as
random sampling.
• Inferences based on the results of random
sampling alone are not only difficult, but
more often than not, inaccurate.
 TWO WAYS TO MAKE
A STATISTICAL
INFERENCE
Hypothesis Testing
• is the statistical tool that determines whether there is a
statistically significant difference between two or more
groups, or whether there is a statistically significant
relationship between two or more variables.
• it is a better statistical method which is utilized to
study the samples taken from the population and make
generalizations about the population from which they
were selected.
•However, inferential statistics do not
only make inferences about the
population from the sample data, but
also estimate parameters from sample
statistics on the top of studying or
testing the null hypothesis, a
research hypothesis.
•A sample statistic is the value of a statistic in a
sample. While, a parameter is the true value of the
same statistic in a sample. The discrepancy between
sample statistics and the population parameters,
otherwise known as the sampling error, is an
important issue in inferential statistics as discussed
earlier. This measures how well or close a particular
estimate of interest captures the true value in the
population. It is also called the margin of error, or in
• Confidence level is the number of times out of 100 that the
true value will fall within the confidence interval.
• Confidence Interval pertains to the calculated range for the
true value, based on the relative sizes of the sample and the
population.
• Confidence levels, which indicate the level of acceptable error,
are based on the concept of the normal curve and probabilities.
The level of confidence is set either at 90%, 95% or 99%. The
95% confidence level means that 95 times out of 100 the true
value will fall within the confidence level. While confidence
•For example, in a study of clients' satisfaction on the
performance of lawyers in a law firm, the researcher
calculated the confidence interval at 5% and set the
confidence level at 90%. As a result, it was found out
that 75% of the clients are satisfied. This means that
the true value is expected to be between 70% to 80%
of customers. However, on the confidence level, it
means that ten times out of 100 the true value will not
fall within the confidence interval considering that only
ninety 100 the true value will land within the
• Statistical analysis is employed to check and describe the existence of
relationship between two variables under study. A statistical test of
significance is exploited to do this important research task. In the final
analysis, it is used to decide on whether
(a)the relationship is the result of chance variations in the selection of
the sample, and thus, the relationship found in the sample data does
not describe a true relationship between variables in the population; or
(b) the relationship is too strong to be explained by sampling error, and
thus, most likely reflects a genuine variable in the population.
However, the statistical analysis for research does not per se prove
anything. It aims to cast doubt on the argument, and therefore,
provide a sufficient basis for reaching the opposite conclusion that
• There could be no statistical test of significance if there is no
null hypothesis. A null hypothesis should be formulated in a
standard statistical way, to wit: that there is no relationship
between the two variables or that there are no differences
between two measures of central tendency. For example, there
is no relationship between variable "x" and variable "y" or there
is no difference between the means or percentages in one
sample compared to another sample. This is called a
hypothesis, which is a statement to be made subject of a test in
a research study. The hypothesis to be tested is known as the
•On the basis of hypothesis testing, the null hypothesis
is either rejected or accepted. It is accepted when the
difference or relationship found are due to chance
variations. This means that the independent variable
had no effect on dependent variable or that the two
means are not statistically different. On the other hand,
it is rejected when the difference or relationship is too
large to have occurred due to chance. It means that
there exists a real relationship or difference between
• The decision to reject Ho and accept Ha or accept Ho
and reject Ha is always exposed to the risk of making an
error. It is not an error free undertaking. Pagoso, et al.
(1992) summarizes the following possibilities of error:
•Type I error- reject the null hypothesis
(action) when in fact the null hypothesis (Ho)
is true (actual condition) and therefore the
alternative hypothesis (Ha) is false.
•Type II error- accept the null hypothesis
(action) when in fact the null hypothesis (Ho)
is false (actual condition) and therefore the
•Simply put, when null hypothesis is rejected, the
alternative hypothesis becomes the true hypothesis. It
means that the alternative hypothesis may be true. On
the other hand, when the null hypothesis is accepted, the
alternative hypothesis may be false. According to
Pagoso, et al. (1992), "the type I error is called producers'
risk because it means the articles produced may be
rejected and not sold"; whereas "the type II error is called
consumers' risk because the consumer may accept
•Thus, in carrying out the test of difference, the
two hypotheses to study are:
(a) the null hypothesis, which states that results
will be due to chance; and
(b) the experimental (alternate) hypothesis,
which predicts that the results are due to
manipulation of the independent variables.
•To accept or reject the null
hypothesis, the level of
significance (α) or p value of the
result is used in the research
study as (α = 0.05 or α = 0.01).
The typical cut-off
points for p value are
shown below:
• The probability level, also called the "p" level or value, is used
to guide the researcher in making decision on whether to
accept or reject a null hypothesis. First, a p level of .05
indicates that the observed finding could have occurred due to
chance variations in sample selection in less than 5 times out
of 100. It means that the researcher accepts the 5% chance in
100 that he/she would reject the null hypothesis when it should
be accepted. With 5% significance level, it means that the
researcher has 95% confident level that he/she has made the
right decision. With 1% significance level, the researcher has
Social research accepts .05 or 5% level of confidence as the
minimum level for rejecting a null hypothesis. When a p level is
greater than 5%, the null hypothesis is accepted. A p value of
0.05 is usually used to distinguish between the significant and
non- significant relations and 95% confidence interval is used to
show the probable margin of error. If 5 percent probability is
expressed as p<0.05, and the null hypothesis is rejected, the
researcher is pretty confident (95%) of the conclusions. The
smaller the p value, the stronger the evidence that null
hypothesis is not true or to be rejected in favor of the alternate
 The procedures for hypothesis
testing in research and thesis
writing are as follows:
(a) Make Ho and Ha in standard statistical way;
(b) Choose the value of a prior to the test;
(c) Calculate the value of the test statistics;
(d) Compare the calculated value (Ca) with a table of the
critical values of the test statistic (Cr);
(e) If the calculated value is less than the critical value
(Ca<Cr), accept the null hypothesis (Ho); or If the
calculated value is greater than or equal to the critical
value (Ca> or = Cr), reject the null hypothesis (Ho) and
accept the alternate hypothesis.
THANK
YOU!