SOS IN PHYSICAL EDUCATION

JIWAJI UNIVERSITY, GWALIOR

FINAL SEMINAR PRESENTATION

TYPES OF VARIABLES AND

EXPERIMENTAL DESIGN IN RESEARCH

Presented by:
ANANT PRATAP SINGH
MPEd. 4TH sem
Roll.no:201616918
VARIABLE
 Variable any characteristic which is subject to change
and can have more than one value such as age,
intelligence, motivation, gender, etc.
 In research, variables are any characteristics that can
be take on different values, such as height, age,
temperature, or test scores. Researchers often
manipulate or measure independent and dependent
variables in studies to test cause-and-effect
relationship.
TYPES OF VARIABLES
 Dependent variable
 Independent variable

 Intervening /mediating variable

 Organismic variable

 Control/constant variable

 Interval variable

 Ratio variable

 Nominal/ categorical variable

 Ordinal variable

 Dummy variable

 Preference variable

 Extraneous variable
DEPENDENT VARIABLE
 Variable affected by the independent variable.
 It responds to the independent variable.

EXAMPLE:
 A test score could be a dependent variable because it
could change depending on several factors such as;
 How much you studied.

 How much sleep you got the night before you took the
test.
 How much hungry you were when you took it.
INDEPENDENT VARIABLE
 Variable that is presumed to influence other variable.
 It is the presumed cause, where as the dependent variable
is the presumed effect.
 EXAMPLE:
 “How stress affects mental state of human beings ?”
 Independent variable -Stress
 Dependent variable - Mental state of human beings

 You can directly manipulate stress level in your human

subjects and measures how those stress level change mental
state.
OTHER NAMES FOR DEPENDENT AND
INDEPENDENT VARIABLES
DEPENDENT VARIABLES INDEPENDENT VARIABLE

 Explained  Explanatory
 Predictand  Predictor
 Regressand  Regressor
 Response  Stimulus
 Outcome  Covariate
 Controlled  Control
INTERVENING/MEDIATING VARIABLE
 It is a variable whose existence is inferred but it cannot be
measured.

 EXAMPLE:

 “Determining the effect of video clips on learning ability of

students of MPEd.”

 The association between video clips and learning ability needs

to explained.

 Other variables intervene:

 Such as anxiety, fatigue, motivation, improper diet, etc.
INTERVENING/MEDIATING VARIABLE
(CONTD.)
 It is caused by the independent variable and is itself a
cause of the dependent variable.
 EXAMPLE:

 Higher education typically leads to higher income.”

 Higher education – (independent variable)

 Higher income – (dependent variable)
 Better occupation – intervening variable

 It is casually affected by education and itself affect

income.
ORGANISMIC VARIABLE
 Any characteristics of the research participant/individual
under study that can be used for classification.

 Such as personal characteristics of gender, height,

weight, age, in behavioral sciences.
CONTROL/CONSTANT VARIABLE
 It is variable that is NOT allowed to be changed
unpredictably during an experiment.
 As they are ideally expected to remain the same, they are
also called “CONSTANT VARIABLES” .
 EXAMPLE:

• “Voltage from a power supply”

 If you are examining “How electricity affect

experimental subjects” you should keep the voltage
constant, otherwise the energy supplied will change as
the voltage will be changed.
INTERVAL VARIABLE
 Interval variable have a numerical value.
 These have order and equal interval.

 They allow not to rank order the item that are measured
but also to quantify and compare the magnitudes of
difference between them.
 EXAMPLE:

 Suppose you have a variable such as monthly income that

is measured in rupees, and we have three people who
make.
 Rs 10,000

 Rs 15,000 and

 Rs 20,000
RATIO VARIABLE
 A ratio variable is similar to an interval variable with
one difference the ratio makes sense.
 EXAMPLE:

 Lets say respondents were being surveyed about their

stress level on a scale of (0-10).
 A respondent with a stress level of 10 should have twice
the stress experienced as a respondent who selected a
stress level of 5.
 Age, height, weight ate also good examples of ratio
variables. Someone who is 6’.0” tall is twice as tall as
someone who is 3’.0” tall.
 NOMINAL/CATEGORICAL VARIABLE
 They can b e measured only in terms of weather the
individual items belong to certain distinct categories.
 We cannot quantify or even rank/order the categories:

 Nominal data has no order.

 One cannot perform arithmetic (+, -, / ,*) or logical

operations (>, <,=) on the nominal data.
 EXAMPLE:
NOMINAL/CATEGORICAL VARIABLE
 Gender (Dichotomous variable)
 Male
 Female

• Marital status:
 Unmarried
 Married
 Divorce
 Widower
ORDINAL VARIABLE
 An ordinal variable is a nominal variable, but its
different states are ordered in a meaningful sequence.
 Ordinal data has order but the interval between scale
points may be uneven.
 Because of lack of equal distance, arithmetic operations
are impossible, but logical operations can be performed
can be performed on the original data.
 A typical example of an ordinal is the socio-economic
status of families.
 “ We know ‘upper middle’ is higher than ‘middle’ but
we cannot say ‘how much higher’.
ORDINAL VARIABLE, EXAMPLE:
 A questionnaire on the time involvement of scientists in
the ‘perception and identification of research
problems’.
 The respondents were asked to indicate their involvement
by selecting one of the following codes:
 Very low or nil

 Low

 Medium

 Great

 Very great

Here, the variable ‘Time Involvement’ is an ordinal

variable with 5 states.
 EXPERIMENTAL
RESEARCH DESIGN
A blueprint of the procedure that enables the researcher to
test his hypothesis by reaching valid conclusion about
relationships between independent and dependent variables.
It refers to the conceptual frame work within which the
experiment is conducted.
Experimental research design is one of the founding
QUQNTITATIVE RESEARCH METHODS
 Experimental research designs are concerned with
examination of effect of independent variable on
dependent variables, where the independent variable is
manipulated through treatments (s) and the effect of
those treatments are observed on dependent variable.
 It is defined as ‘observations under controlled
conditions’.
 TYPES OF EXPERIMENTAL
RESEARCH DESIGN
 Pre experimental research design.(very low control, used for
beginners)
 (SINGLE TIME TESTING)
 True experimental research design.(full control, all conditions
fulfilled)
 (EXPERIMENTAL AND CONTROL GROUP SELECTES
RANDOMLY)
 Quasi experimental research design/Ex-post facto research
design.(one condition is not fulfilled) (RANDOMIZATION)
 (GROUP ARE NOT SELECTED RANDOMLY)
 Other experimental research design.
 (TOO MUCH COMPLICATION)
 CLASSIFICATION OF EXPERIMENTAL
RESEARCH DESIGN
S.NO Pre ERD True ERD Quasi ERD Other ERD
1. Single short Randomized Reverse design Factorial design
(one group) group design (one group) (two or more
(two or more groups)
groups)
2. Single group Pre post test Time series Latin square
pre-post test randomized design design
(one group) design (one group) (two or more
(two or more groups)
groups)
3. Static group post Solomon 3 or 4 Non equivalent
test group design design
(two group) (3-4 group) (two or more
groups)
 DESIGNS
SINGLE GROUP (ONE GROUP)
 In this design, single group is experimented and only one
observation/test is taken for conducting research work.
 Design:

Single group- Treatment-Post test

RD: (G-T-O post)
Single group (G);Treatment (T);Observation (O)
 Advantages:

 Less time taking

 Less expensive

 Disadvantages:

 Lack of group

 Lack experiments
 THE EXPERIMENTAL RESEARCH
DESIGNS
Single group pre post GERD (one group)
 In this design, single group is used. Pre test is taken before
experiment/treatment and post test is taken at last.
 Design:

Single group-pre test-treatment-post test

(G-O pre-T-O post)
 Advantages:

 Investigate effect of experiment.

 Less time taken.

 Less expensive.

 Disadvantages:

 Lack of groups.

 Lack experiments.
STATIC GROUP ERD (TWO GROUP)
 In this design, two different groups are used, but only
one of them is given the experimental treatment and at
the both group post test is taken.
 Design:

Experimental group: Treatment-post test

(G-T-O post)
Control group: No treatment –post test
(G-X-O post)
X= No treatment
 TRUE EXPERIMENTAL RESEARCH DESIGNS
POST TEST RANDOMIZED GROUP ERD
(TWO OR MORE GROUPS)

 This design is similar to the static group design except than an

attempt is made to selected two or more groups randomly using
different random sampling technique.
 Only experimental groups is given treatment and at the end all the
groups post test is done.
 Their may be more than one experimental as well as control group.
 Design:
Experimental group 1: treatment one-post test
(RG-TI-O post)
Experimental group 2: treatment two-post test
(RG-T2-O post)
Control group 1: no treatment-post test
(RG-X-O post)
POST TEST RANDOMIZED GROUP ERD
RANDOM
ASSIGNME
NT
CON PO
TRO
PR TR ST-
EX L
EA TES
P.G ET GRO
T
TM UP
RO ES EN
UP T T
POST-TEST
 EXPERIMENTAL RESEARCH DESIGNS
PRE-TEST, POST TEST CONTROL GROUP DESIGN

RANDOM
POS
LY
TRE
SELECTES
T-
EXPERIM
ENTAL ATM
GROUP TES
(PRE-
TEST)
ENT T
RANDOMLY
SELECTED
CONTROL
POST-
GROUP
(PRE-TEST) TEST
 SOLOMAN ¾ ERD
(THREE OR FOUR GROUPS)
 This type of design is used on large population only. The
groups are randomly divided followed by the pre test
which is followed by the treatment and the final test.
 This can also includes post test only or pre test post test
random design both depending upon the situations
separately or simultaneously.
 The comparison will be made between the group
vertically in case of pre test and post test.
DESIGN:
Experimental group 1: Random group –pre test-treatment-
post-test
(RG-O pre-T1-O post)
Control group 1: Random group-pre test-post test
(RG-O pre-X-O post)
Experimental group 2: Random group-no pre test-
treatment-post test
(RG-X-T-O post)
Control group 2: Random group-no treatment-no
treatment-post test
(RG-X-X-O post)
 SOLOMON 4 ERD

EXP.G TREA
PRE POST-TEST
ROUP TME
1 TEST NT

CONTRO
L GROUP
PRE POST-
1 TEST TEST
RANDOM
ASSIGNMENT
EXP.G TREA
POST-
ROUP TME
TEST
2 NT

CONTROL POST-
GROUP 2 TEST
 QUASI EXPERIMENTAL RESEARCH
DESIGNS
Reverse designs ERD (one group)
 This is the variation of single group design but more than
one treatment is given to the group. An initial test is
taken followed by the application of the treatment 1
which is followed by re-test followed by application of
treatment 2 and then final test.
 Design:

Experimental group: group-pre test-treatment-post test

(G-O1-T1-O2-T2-O3)
 TIME SERIES DESIGN (ONE GROUP)
 In this design only one group is evolved but many
observations will be recorded to see the effect of
treatment or change due to the treatment. Many test are
taken followed by a treatment which is again followed
by many post tests.
 Design:

Experimental group: group-pre test-pre test 2-pre test 3-

treatment-post test 1-post test 2-post test 3
(G-O pre1-O pre2-O pre3-T-O post1-O post2-O post3)
 NON EQUIVALENT DESIGN
(TWO OR MORE GROUPS)
 It is structured like a pre test-post test randomized
experiment, but it lacks the key feature of the random
selection of group.
 Design:

Experimental group: group-pre test-treatment-post test

(G-O pre-T-O post)
Control group: group-pre test-no treatment-post test
(G-O pre-O post)
 OTHER EXPERIMENTAL RESEARCH
DESIGNS
Factorial ERD (two or more groups)
 A factorial design is often used by scientists wishing to understand
the effect of two or more independent variable upon a single
dependent variable.
 Design:

Experimental group 1: random group-treatment1, 1-post test

(RG-T1, 1-O post)
Experimental group 2: random group-treatment1,2-post test
(RG-T1, 2-O post)
Experimental group 3: random group-treatment2, 1-post test
(RG-T2, 1-O post)
Experimental group 4: random group-treatment2, 2-post test
(RG-T2, 2-O post)
THANK YOU…