Page 45 - SREENARAYANAGURU OPEN UNIVERSITY

variable that influences the dependent variable even when it is
not explicitly stated in the study. An example of randomised
block design is a researcher investigating the effects of three
different exercise programmes (A, B, and C) on cardiovascular
fitness. In this example, the researcher can make age as the
blocking variable. The researcher divides participants into ◆ Principles of local
three distinct age groups: young adults (18-25 years), middle- control
aged adults (35-50 years), and older adults (60+ years).
Each age group serves as a separate block. Participants are
randomly assigned to one of the three exercise programmes
within each block. This random assignment guarantees that
participants in the fitness programmes for each age group are
distributed equally.

◆ Latin square design
Latin square design enables the researcher to manipulate
two extraneous variables. Each treatment appears an equal
number of times in each row and column in any one ordinate
position. The latin square design is used when there are
multiple treatments or conditions to be compared, while also
controlling for potential confounding factors. The design is
structured like a grid, with each row and column representing
a treatment or condition. Each participant receives one
treatment from each row and column, resulting in a balanced ◆ Controls two
distribution of treatments. The latin square design is useful extraneous variables
when there are factors other than the treatment of interest that
could influence the outcome, such as time, order, or specific
sequences of treatments. For example, a researcher aims to
examine the effects of three different fertilizers (A, B, and C)
on plant growth at three different locations (X, Y, and Z). The
latin square design ensures that each fertilizer is used once
at each location, and each location is used once with each
fertilizer.
The latin square design will be as follows:

Locations Fertilizers

X A B C

Y B C A

C
Z
A
◆ Factorial design B
Factorial designs are used to measure the effect of multiple
independent variables on the dependent variable. Factorial
designs, unlike any other statistical design, allow for variable
interaction. An interaction occurs when the sum of the effects
of two or more variables taken together differs from the sum

SGOU - SLM - MCom Research Methodology 39