Types of Test: Group 3

TYPES OF TEST
Group 3
What is a statistical test?
A statistical test is a way to evaluate the
evidence the data provides against a
hypothesis.
This hypothesis is called the null
hypothesis and is often referred to as H0.

A statistical test is based on two
competing hypotheses: the null

hypothesis H0 and the alternative
hypothesis Ha.
TYPES OF TEST
Two-tailed
and
One-tailed
test
Two-tailed test
 Is associated to an alternative hypotheses for
which the sign of the potential difference is
unknown.
 is a method in which the critical area of a
distribution is two-sided and tests whether a
sample is greater or less than a range of values
 If the sample being tested falls into either of
the critical areas, the alternative hypothesis is

accepted instead of the null hypothesis.
One-tailed test
 Is associated to an alternative hypothesis for
which the sign of the potential difference is known
before running the experiment and the test.
 Is a statistical test in which the critical area of a
distribution is one-sided so that it is either greater

than or less than a certain value, but not both.
 If the sample being tested falls into the one-
sided critical area, the alternative hypothesis will

be accepted instead of the null hypothesis.
TYPES OF TEST
Paired
and
Independent Sample
test
Paired Sample test
 Sometimes called the dependent sample t-
test
 Is a statistical procedure used to determine
whether the mean difference between two

sets of observations is zero.
 Is used when we are interested in the
difference between two variables for the

same subject.
 Often the two variables are separated by
time.
Independent Sample test
 Isalso known as Independent t Test
 Compares the means of
two independent groups in order to
determine whether there is statistical
evidence that the associated
population means are significantly
different.
TYPES OF TEST
Parametric
and
Non-Parametric
test
Parametric Test
 Parametric tests are those that
make assumptions about the
parameters of the population
distribution from which the sample
is drawn. This is often the
assumption that the population
data are normally distributed.
Advantages
allow one to make generalizations from
a sample to a population
they do not require interval- or ratio-
scaled data to be transformed into rank

data.
a parametric test is more able to lead
to a rejection of H0
PARAMETRIC
TEST
ASSUMPTIONS
1. Normal distribution of data
in a graph form, normal
distribution will appear as a
bell curve
is a probability distribution
2. Homogeneity of variance
The population variances
(i.e., the distribution, or
“spread” of scores around
the mean) of two or more
samples are considered
equal.
3. Interval data
Intervaldata always appears in the
form of numbers or numerical
values where the distance between
the two points is standardized and
equal.
Each point is placed at equal
distance from one another.

4. Independence
 The assumption of independence means
that your data isn't connected in any way
 There are actually two assumptions:
 The observations between groups should
be independent, which basically means the

groups are made up of different people.
 The observations within each group must
be independent.
Examples of distributions of data
Non-parametric test
 isalso called distribution-free test because
they do not assume that your data follow a
specific distribution.
 are methods of statistical analysis that do
not require a distribution to meet the

required assumptions to be analyzed
 are based on fewer assumptions (e.g.,
they do not assume that the outcome is

approximately normally distributed)
Advantages
they are valid in a broader range of
situations (fewer conditions of
validity)
do not require a distribution to meet
the required assumptions to be

analyzed
Can often be used in small samples and
for ordinal data

Parametric and Non-
parametric tests for
comparing two or
more groups
NON-PARAMETRIC
PARAMETRIC TEST
TEST
PAIRED T-TEST WILCOXON RANK

The purpose of the test is SUM TEST
to determine whether The Wilcoxon rank-sum
there is statistical
test is commonly used
evidence that the mean
for the comparison of
difference between paired
two groups of
observations on a
nonparametric (interval
particular outcome is
significantly different or not normally
from zero. distributed) data
UNPAIRED T- MANN-WHITNEY
TEST U TEST
An unpaired t-test (also Is used to compare
known as an independent differences between two
t-test) is a statistical independent groups when
procedure that compares the dependent variable is
the averages/means of either ordinal or
two independent or continuous, but not
unrelated groups to normally distributed.
determine if there is a
significant difference
PEARSON’S THE SPEARMAN
CORRELATION RANK-ORDER
COEFFICIENT CORRELATION
Pearson’s correlation COEFFICIENT
coefficient is the test
(Spearman’s correlation, for
statistics that measures short)
the statistical The test is used for either
relationship, or ordinal variables or for
continuous data that has failed
association, between two
the assumptions necessary for
continuous variables.
conducting the Pearson's
product-moment correlation. It
measure of the strength and
direction of association that
ONE-WAY KRUSKAL-
ANALYSIS OF WALLIS TEST
VARIANCE Is used when the
Compares the means assumptions of one-way
ANOVA are not met. Both
between the groups
the Kruskal-Wallis test
you are interested in
and one-way ANOVA
and determines assess for significant
whether any of those differences on a
means are statistically continuous dependent
significantly different variable by a categorical
independent variable (with
from each other.
How do I know if my data is
parametric or nonparametric?
 If the mean more accurately represents the
center of the distribution of your data, and
your sample size is large enough, use
a parametric test.
 If the median more accurately represents
the center of the distribution of your data,
use a nonparametric test even if you have a
large sample size.
THANK YOU
FOR
LISTENING!

Types of Test: Group 3

Uploaded by

Types of Test: Group 3

Uploaded by

TYPES OF TEST

hypothesis and is often referred to as H0.

competing hypotheses: the null

the critical areas, the alternative hypothesis is

distribution is one-sided so that it is either greater

sided critical area, the alternative hypothesis will

whether the mean difference between two

difference between two variables for the

scaled data to be transformed into rank

distance from one another.

be independent, which basically means the

not require a distribution to meet the

they do not assume that the outcome is

the required assumptions to be

for ordinal data

PAIRED T-TEST WILCOXON RANK

You might also like