Dr.
Salman Iqbal
WEEK 8
Dr. Salman Iqbal
Week 8
Questionnaire
development and Scales
Questionnaire development
• Demographic
• Likert scale items
Scales
• A scale is a tool or mechanism by which individuals are
distinguished as how they differ from one another.
– To measure consumers’ attitude towards soft drinks
– We need to classify our objects/consumer to decide
how unfavourable or favourable is soft drink for them,
one way is introducing some scale
– A scale from 1 to 5 or 1 to 7, including strongly agree to
strong disagree, for instance, I enjoy having a soft drink on
five point scale. OR, I prefer tap water to soft drink,
Type of scales
• Nominal Scale
• Ordinal
• Interval
• Ratio
• Nominal scale:
– Categorise the object, For instance: Gender
• Assigning number to each group
– Country of residence
– Department(marketing/finance/R&D)
• Ordinal scale:Choice of order
– It shows the order, difference among the categories
– For instance, from best to worst, or numbers
assigning weight to those numbers
– Job characteristics: interact with others, learn skills,
complete whole task, work independently, serve others.
– Consumer preferences to type of soft drinks, Computer
brand preferences (Apple, HP, Dell, SONY, Toshiba)
• Nominal scale provide a qualitative and a
descriptive information.
• Ordinal scale provided some more information.
• However, in terms of Data analysis, performing
arithmetic means, Standard deviation ( to calculate
the magnitude of difference), Interval scale is used
• For instance, use this scale for job characteristic
example and computer brand example. Difference
is same between the points, 1 to 2, 3 to 4...
How does each scale help?
• Nominal (gender, ethnicity, major, religion)
• Ordinal (age groups, grade level, ranks)
• Interval (Fahrenheit, Celsius)
• (Celsius: Yesterday 25 Today 50) (Fahrenheit: 77, 122)
• Ratio (time, number, age, height, weight)
Inches (25/50) Centimeters (62.5/125)
• Ratio scale: is used in business research when
exact numbers on objective are used
– For instance, How many courses you took
before this semester
– How many organisations did you work for
before joining this company
– How many retail outlets do you operate
– The answer could be from zero to any
number.
Important .
Rather than developing your own
questions, it often makes sense to use or
adapt existing scales (Schrauf and Navarro
2005). Since scaling techniques were first
used in the 1930s, literally thousands of
questions and scales have been developed
to measure attitudes and personality
dimensions and to assess skills and abilities.
Rank
Summary
Open ended example
Summary
Close ended
List category
Dr. Salman Iqbal
Validity & Reliability
Four Aspects of Reliability:
• 1. Stability
• 2. Reproducibility
• 3. Homogeneity
• 4. Accuracy
1. Stability
• Consistency across time
– repeating a measure at a later time to examine
the consistency
– Compare time 1 and time 2
2. Reproducibility
• Consistency between observers
• Equivalent application of measuring device
– Do observers reach the same conclusion?
– If we don’t get the same results, what are we
measuring?
• Lack of reliability can compromise validity
3. Homogeneity
• Consistency between different measures of
the same concept
– Different items used to tap a given concept show
similar results – ex. open-ended and closed-ended
questions
4. Accuracy
• Lack of mistakes in measurement
• Increased by clear, defined procedures
– Reduce complications that lead to errors
• Observers must have sufficient:
– Training
– Motivation
– Concentration
Types of Validity
• 1. Face validity
• 2. Content validity
• 3. Construct validity
– A. Testing of hypotheses
• Convergent validity
• Discriminant validity
Face Validity
• Subjective judgment of experts about:
– “what’s there”
– Do the measures make sense?
• Compare each item to conceptual definition
– Do it represent the concept in question?
• If not, it should be dropped
– Is the measure valid “on its face”
Measure??
Measure..?
Time horizon
• Cross-sectional • Longitudinal
Observation at one • several observations of the
point a certain time same subjects over a period of
frame) time
• To Measure the cholesterol • To Measure the cholesterol
levels of the patients (no levels of the patients
past history) periodically
• Benefit: compare different • Benefit: Establish cause and
variables, Age/gender/income effect, a true picture,
with cholesterol levels predict and suggest
• Limitation: cross-sectional • Disadvantage: costly and
studies may not provide time consuming, hence
definite information about some time out of scope of
cause-and-effect relationships. small studies
