Scribd
Upload
65 views

Session-5-Measuring and Scaling Concepts

This document discusses various concepts related to measurement and scaling of variables in quantitative research. It covers: 1) The importance of measuring variables in a valid and reliable way in order to test hypotheses and answer research questions. 2) Operationalization which involves reducing abstract concepts into observable and measurable behaviors. 3) Different levels of measurement scales including nominal, ordinal, interval, and ratio scales and examples of each. 4) Ensuring good psychometric properties of scales through validity which tests accuracy of measurement, and reliability which tests consistency of measurement.

Uploaded by

Muhammad Gulfam
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
65 views

Session-5-Measuring and Scaling Concepts

This document discusses various concepts related to measurement and scaling of variables in quantitative research. It covers: 1) The importance of measuring variables in a valid and reliable way in order to test hypotheses and answer research questions. 2) Operationalization which involves reducing abstract concepts into observable and measurable behaviors. 3) Different levels of measurement scales including nominal, ordinal, interval, and ratio scales and examples of each. 4) Ensuring good psychometric properties of scales through validity which tests accuracy of measurement, and reliability which tests consistency of measurement.

Uploaded by

Muhammad Gulfam
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 20

Session-5

Measurement and scaling of variables


Session-5
Session- 4 How a quantitative research is carried on scientifically

Lit-
Review
measurement of variables

• Measurement of the variables is an integral part of scientific

research. Unless the variables are measured in some way

(which is valid and reliable), we will not be able to test our

hypotheses and find answers to our research questions.


Quantification/ operationalization of variables of study to measure
them
• Easily measurable variables ( Age, gender, qualification, weight , height etc)
• Abstract concepts, hard to define or measure ( Loyalty, personality, anger, love ,
power, learning etc).

• Reduction of abstract concepts into visible behaviors/measurable form is called


operationalization.

• Aggression (verbal, behavioral)


• What are some observable behaviors related to aggression?????????
• Shouting, aggression, abusive language, breaking things, sulking,
Operationalization of a construct
1-Start with an operational definition: An operational definition entails a concise
description of concepts/behaviors associated with a specific variable to facilitate the
collection of meaningful and standardized data.

Give a clear definition of at least your IV , DV and mod, med (literature review).

2-Choose the scale (from available literature)/ design scale items/

• Already available scales Lucky you 

• Develop your own scale (a rigorous process in academic research)

In applied research, questionnaire could be self developed/ templates are


available/ready made scales/ survey monkey. com
CP- How will you operationalize customer satisfaction
• To operationalize the concept of customer satisfaction, Write down any 5
behaviors/ attitudes which come in your mind.

• Gr-1 repeated visits, consistently high ratings, recommending our products,


lesser complaints, long term customers.

• Gr-2 positive reviews, frequent visits, recommendations, good mood


• Frequency of site visits.

• Gr-3 rebuy, pos reviews, social media following, lesser complains


• Refer the brand
Scale for creative climate
5= SD, 4=D, 3= N, 2=A, 1=SA

1 2 3 4 5
Scales
• There are four fundamental levels of measurement on which variable
could be measured.

• 1- Nominal
• 2- Ordinal
• 3-Interval
• 4-Ratio

• Different measurement levels are used to capture data in the form of 
surveys and questionnaires.
Nominal and Ordinal scales
• Nominal ( categorical scale) scale, is • Ordinal Scale not only assigns labels to the
used for labeling variables into distinct variables but also measures the rank or order of
classifications and doesn’t involve a the variables. It does not tell the distance
quantitative value or order. between them.
• Low-income, middle-income, high-income
• Male Female • Undergrad grad post-grad
• How satisfied are you with our services?
• Muslim, Non Muslim • Very Unsatisfied – 1
• Unsatisfied – 2
• Neutral – 3
• “Apple” “Samsung” “Q Mobile” • Satisfied – 4
• Very Satisfied – 5
(An absolute zero is the defining characteristic which differentiating between ratio and
interval scales).

• The Ratio Scale is the highest level scale


• Interval scale is defined as a numerical scale that allows the researcher to classify or
where the order of the variables is known as identify the objects, rank-order the objects
and compare the intervals or differences.
well as the computable difference between • Naturally occurring 0 point/ absolute
these variables. Relative differences, No meanings
naturally occurring 0 point.
• Family members (>70)
• 80 C----40 C is the same as 70C and 30C • What is your weight in kilograms?
• IQ/ Personality 30 score (A) – 60 score (B) • Income

• ????
CP- Question
• Nominal, Ordinal, interval, Ratio

1. Scale measuring the locality of the respondents.


2. Scale measuring the order of importance you give to 4 make up
brands.
3. Scale measuring the mood on a scale of 1-10
4. Scale measuring the income of a house hold
5. Scale measuring the number of cars a household owns
6. Scale measuring pain on a scale of 1-5. 12345
Types of scales .........1-Rating scales (categories)
• Nominal Level

• Dichotomous scale yes /no, male /female

• Category Scale Lahore, Karachi, Islamabad, Peshawar, Quetta

• Semantic differential scale (attitudes, opinions/ mood)

motivated………………………….not motivated
Rating scales (opinions and attitudes) at interval level of
measurement

• Numerical Scale How satisfied are you with our new package plan?
• Extremely satisfied 7 6 5 4 3 2 1 Extremely Dissatisfied

• Itemized rating scale


(with anchors)
The measurement of attitudes
Ranking scales ordinal level
• Forced choice
• The forced choice enables respondents to rank objects relative to one
another, among the alternatives provided. This is easier for the
respondents, particularly if the number of choices to be ranked is
limited in number.
• Rank the following news channels that you would like to buy in the
order of preference, assigning 1 to the most preferred choice and 5 to
the least preferred.

• Geo News, ARY News, Express News, Dawn News, Neo News
Exercise 7.5
Develop and name the type of measuring instrument you would use to tap the following:

a. Which brands of tea is preferred by consumers, Tapal or Lipton?

b. Among the three types of exams – multiple choice, essay type, and a mix of both – which

is the one preferred most by students?

c. To what extent do individuals agree with your definition of accounting principles?

d. How much people like an existing organizational policy.

e. The selection preferences of students for math, accounting, management, and law
Goodness of a scale ....validity and reliability (psychometric
properties)
• Validity (accuracy) is a test of how well an
 Content validity ensures that the
instrument that is developed measures measure covers the whole domain of the
the particular concept it is intended to variable of interest

measure. In other words, validity is  includes an adequate and representative


concerned with whether we measure the set of items that tap the concept.
right concept.
• Example:
• Content validity Face validity • Intelligence ( verbal, analytical, math
metical, memory)
• Predictive Validity

• Convergent validity Discriminant Validity • SAT test/ GMAT/GRE


Reliability (F) Consistency
•
• Examples:
• A scale should show
consistent results over • Weight measurement
time.

• Recruitment tests
• Split half
• Test retest
• SAT scores
Quiz on Monday- All week’s work- MCQs and short questions- In class
quiz

You might also like