STATISTICAL
ANALYSIS

Statistical analysis
is essential for
professional marketing
research because
it’s used to
measure differences
or associations between
variables. ARI uses
statistical techniques
with moderate complexity
to develop useful
insights and focused
recommendations.
These techniques
are rooted in reality
and proven through
academic publications.
Reliance on statistics
offered from do-it-yourself
surveys is inadequate
for important decisions
because theses services
seldom test the significance
of the results. The
statistical procedures
most frequently used
on ARI marketing
projects include
the following:

(1) Univariate techniques
are used for analyzing
data when there is
a single measurement
of each element in
the sample or when
there are several
measurements on each
element but each
variable is analyzed
in isolation. These
techniques focus
on averages and distributions.
Sub classifications
of univariate techniques
are based on the
way data is measured.
i.e. metric or nonmetric.
A satisfaction rating
based on a 5-point
scale is an example
of metric data. Survey
questions with a
yes or no response
offer data in a nonmetric
format.

(1a) Analysis of
variance (ANOVA)
is a common univariate
technique because
it’s often
used to measure
differences between
groups or business
units. ANOVA is
also used to determine
the significance
of performance differences
when they do exist.
Analysis of variance
is very useful for
customer satisfaction
surveys because
it helps clients
identify meaningful
differences. To
see a detailed example
of this procedure
click
here.

(1b) A cross tabulation
performed with a
chi-square test
is another univariate
technique which
is used to reveal
associations between
two or more marketing
variables. The chi-square
value is used to
test the significance
of the association.
This statistical
technique is often
utilized to identify
customer types who
are more or less
likely to be influenced
by a promotional
strategy. For example,
is there an association
between motivation
and the reach of
a specific advertising
medium? To see a
detailed example
of this procedure
click here.

(2) Multivariate
techniques are used
for analyzing data
when there are two
or more measurements
of each element and
the variables are
analyzed simultaneously.
These techniques
concentrate on the
degree of relationships
between marketing
variables of interest.
Sub classifications
of multivariate techniques
are based on dependence
and interdependence
techniques.

(2a) Regression
analysis is a powerful
predictive analysis
technique in which
one or more variables
is used to predict
the level of another
based on a straight
line formula. This
is classified as
a dependence technique
because one or more
variables can be
identified as dependent
variables and the
remaining as independent
variables. Regression
analysis is frequently
used to measure
the significance
and strength of relationships
between marketing
variables. i.e.
sales and advertising
expenditures. To
see a detailed example
of this procedure
click here.

(2b) Factor analysis
is one of the more
complex statistical
techniques because
a whole set of interdependent
relationships among
variables is examined.
Factor analysis
is primarily used
to identify the underlying
dimensions which
explain the correlations
among a set of marketing
variables. This
insight is useful
for market segmentation
because it can be
utilized to identify
the underlying variables
on which to group
customers. To see
a detailed example
of this procedure
click here.