MEASUREMENT SCALE

Dr. Kuldeep SharmA

Hod, Research Centre
K.P.B. Hinduja College
Charni Road, Mumbai
 TOPICS

 Types of data

 Measurement in Research / Measurement Scales

 Scale of Measurement and Test

TYPES OF DATA

• To give a holistic picture of classification data

can divided into two types

• Quantitative data (numerical)

• Qualitative data (descriptive,
categorical/frequency count)
 Types of Data
Data

Qualitative Quantitative
Data Data

Discrete Continuous
Nominal Ordinal
Data Data

Interval
Ratio Data
Data
 Types of Data

Qualitative Data
 Nominal Non
 Ordinal Metric

Quantitative Data
 Interval Metric
 Ratio
Qualitative Data
• Also called descriptive/ categorical data/ frequency
count.
• When the data are arranged in categories on the basis
of their quality and there is gap between two values,
it is called qualitative data, e.g name, religion, marital
status, socioeconomic status, awareness.
• Qualitative data cannot be expressed in numerical
forms.
• Types of Qualitative Data:
1.Nominal Data
2.Ordinal Data
QUANTITATIVE DATA
The data , that can be expressed in numbers/ figures is called
quantitative data. In this the exact measures are possible. It has
two types

(a) Discrete:
Discrete variables can take only certain values and none in
between e. g number of patients in a hospital census may be 178
or 179, but it cannot be in between these two, similarly the number
of syringes used in a clinic in one day or number of children in a
family. It is expressed in whole number.

(b) Continuous:
Continuous variables may take any value (typically between
certain limits). For example age (25.5 years), weight (70.5 kg),
height (1.5 meter) , hemoglobin (12.5 gm), blood pressure
(135/95). It can be expressed in decimals.
 Nominal data

Nominal data are divided into categories, that are

only distinguished by their name and labels and
cannot be classified one above another e.g race,
name , sex, name of country, name of crops, type
of blood. In this type of data there is no
implication of order or ratio.
 Ordinal data

When the categorical data can be placed in

meaningful order on the basis of their quality, it
is known as ordinal data.

In this the exact difference between the two

groups cannot be estimated e.g pain categorized
as mild, moderate and severe.
Interval
Interval data are like ordinal data in that they can be
placed in a meaningful order.

• The categories are arranged in equally spaced units and

there is no absolute zero point e.g temperature where 0
˚ C does not mean no temperature but is equal to 32 ˚ F
or 273 K ( Kelvin scale ).

• However because interval scales do not have an

absolute zero, ratio of scores are not meaningful e.g
100 ˚ C is not twice as hot as 50 ˚ C, because 0 ˚ C does
not indicate a complete absence of heat.
 Ratio
• A ratio data has the same properties as an interval
data; however, because it has an absolute zero,
meaningful ratios do exist.

• Most biomedical variables form a ratio data e.g

weight in grams or pounds, time in seconds or days,
blood pressure in millimeters of mercury, and pulse
rate in beats per minute are all ratio scale data.
Measurement means...............

1. Assigning numbers to observation

2. Must have rule for assigning numbers

3. We don’t Measure object, we measure characteristics

4. Standardizes rules and applied uniformity

Types of Measurement

• Nominal Scale

• Ordinal Scale
• Interval Scale

• Ratio Scale
Nominal Scale

Nominal
Scale

Polytomous/
Dichotomous
Multinomial

Tick only Multiple

True Artificial
one Tick
Dichotomous Question
True Dichotomous Questions
1. Gender
2. Day/ Night
3. Sun/ Moon
4. Dead/ Alive

Artificial Dichotomous Questions

1. Pass or Fail
2. Wrong or Right
3. Kashmir
4. History
Polytomous Question
One Tick Questions
Where is your birth place? More Example.......
1. Delhi 1. Occupation
2. Mumbai 2. Qualification
3. Chennai 3. Income
4. Kolkata

Multiple Tick Questions

Which place you like? More Example........
1. Paris 1. Movie
2. London 2. Colour
3. Rome 3. Newspaper
4. Berlin
Ordinal Scale

1. It can be arrange in order, rank, sequence or

hierarchy
2. Unknown distance between options
3. The distance is not meaningful
4. Raking or order has no absolute zero
5. No mathematical equations can be used
Some Examples..............

1.Shirt Size (S, M, L, XL)

2.Qualification (SSC, HSC, G, PG)
3.Income Group (> 1, 1< >3, 3< >5)
4.Grade (O, A, B, C,)
5.Age group......
Interval Scale

1.Equal and definite interval between

two measurement
2.Arbitrary or No Absolute Zero
3.Cannot determine as scale
Ratio Scale

1.Highest Level of Measurement

2.Fixed or Absolute Zero
3.Exact Value
4.All mathematical operation can be carried out.
5.Can be converted into Interval, Ordinal and
Nominal Scale
Some Examples..............

1.Height (__)
2.Age (__)
3.Income (__)
4.Weight (__)
5.Marks (__)
 No
(Nominal)
Categorical data
(In Order)

Yes
(if Equal
Space)

No Yes
(Ordinal) (Zero means none)

No Yes
(Interval) (Ratio)
WHAT IS THE LEVEL OF MEASUREMENT FOR EACH
OF THE VARIABLE?

 Student’s Gender
Male…. Female…..

Ans: Nominal
 WHAT IS THE LEVEL OF MEASUREMENT FOR EACH
OF THE VARIABLE?

Ranking of Student by First Year, Second Year,

Third Year ?
Ans: Ordinal
WHAT IS THE LEVEL OF MEASUREMENT FOR EACH
OF THE VARIABLE?


Student Grade point Average ?
Ans: Interval
WHAT IS THE LEVEL OF MEASUREMENT FOR EACH
OF THE VARIABLE?

Number of hours Student Study per week ?

Ans: Ratio
SCALE OF MEASUREMENT AND TEST

Nominal Define an attributes Non-Parametric

Rank the observation as

Ordinal Non-Parametric
score

Distance btw observation Parametric

Interval
in fixed units

Scale Fundamental Zero Point Parametric

TEST OF DIFFERENCE
TEST OF DIFFERENCE
TEST OF RELATIONSHIP
MAKE LIFE SIMPLER………….
Questions ??