Page 161 - SREENARAYANAGURU OPEN UNIVERSITY

sample data analysis. The researcher must ensure that the samples
chosen for the study are representative and adequate in number in ◆ Estimation
order to arrive at accurate estimates of parameters. The estimation of population
can be done using two methods- Interval estimation and Point parameters
estimation.
ii. Test of hypotheses: Inferential statistics focuses on the
numerous significance tests for testing hypotheses to ascertain the
degree of validity with which evidence can be used to support
a conclusion or series of conclusions. Testing of hypotheses for
inferential statistics can be done using the following two types of
tests:

◆ Parametric tests - Parametric tests assume that the data follows
a specific distribution, usually the normal distribution. These
tests make assumptions about the population parameters, such ◆ Assumptions about
as mean and variance. Parametric tests are powerful when population
these assumptions are met, but they may not be accurate when
the assumptions are violated. The most common parametric
test includes t-test, z-test, ANOVA, MANOVA, regression
etc..

◆ Non-parametric tests- Non-parametric test are those tests
which can be used for ordinal and nominal data. It does
not make assumptions about population such as normality
in distribution and randomness like parametric tests. They ◆ No assumptions
are used when the data doesn’t meet the assumptions of about population
parametric tests, such as when the data is not normally
distributed or when it includes outliers. The most common
non-parametric tests are Mann-Whitney U test, Kruskal-
Wallis H test, Friedman test, Wilcoxon signed-rank test etc..
iii. Based on the number of variables considered for analysis,
data analysis can be classified as:

a. Univariate analysis- This sort of analysis describes the data
on one variable. ‘Uni’ means one and ‘variate’ means vari-
able, so in univariate analysis, there is only one dependable ◆ One variable
variable. The objective of univariate analysis is to derive the
data, define, summarise it, and analyse the pattern present in
it. In a data set, it explores each variable separately. It is pos-
sible for two kinds of variables- categorical and numerical.

b. Bivariate analysis- This sort of analysis describes the data
on two variables. ‘Bi’ means two and ‘variate’ means vari- ◆ Two variables
able, so here there are two variables. The analysis is related
to cause-and-effect relationship between the two variables.
c. Multivariate analysis- Multivariate analysis is required

SGOU - SLM - MCom Research Methodology 155