Business Research C6

RESEARCH METHODOLOGY
Arti Srivastava
Assigning symbols/numbers to characteristics of objects
MEASUREMENT • Consumer preference; Household income
AND SCALING Isomorphic assignment
• Numbers/Symbols should have a standardized meaning
Scale Characteristics
Description Nominal
Ordinal
Order Interval
Ratio
Distance
Origin
Description • Labels that categorize variables: essentially a noun

Nominal
• Purely descriptive – Categorical
Order • Brand of Phone that you use
• Eye color
• Store number
Distance
• Consumer IDs
• What can you do with them? Percentages, Mode
Origin
• Inferential statistics: Non-parametric tests, e.g., Chi-square
Description • A ranking scale

Ordinal • Relative extent of a characteristic- Categorical
Order • Education: High School….Intermediate…Graduate…Post Graduate….Doctorate
• Socioeconomic Status: Upper…Upper Middle…Middle…Lower Middle…Lower
Distance • Quality ranking: Very Poor… Poor… Average…Above Average… Good…Great
• What can you do with them? Percentile, Median
Origin • Inferential statistics: Non-parametric tests, e.g., Correlation, One-Way ANOVA
Description • Numerical data measured using continuous intervals

• Can categorize and order variables- Not true zero
Order Interval • Temperature
• Income Levels: Under $10K, $10-$20K, $20K-$30K…
Distance • IQ test scores
• Preference rating, Innovativeness rating
Origin • What can you do with them? Mean, Standard Deviation
• Inferential statistics: Parametric tests, e.g., Regression, T-Test, Factor Analysis
Description • Numerical data with absolute zero

• Can be used to identify, rank, compare and measure the difference
Order • Height, Weight, Age
Ratio • Price, Cost
Distance • Market share
• Willingness to pay
Origin • What can you do with them? All statistical operations
• Inferential statistics: All Parametric and non-parametric tests
Scaling Techniques
Comparative Scales Non-comparative Scales
Description Nominal
Ordinal
Order Interval
Ratio
Distance
Origin
Scaling Techniques Paired Comparison
Comparative Rank Order
Constant Sum
Scaling techniques
Continuous Rating Likert
Scale
Non-Comparative
Itemized Rating
Scale
Semantic
Paired
Comparison
Brand Coca-Cola Pepsi Sprite Limca
Coca-Cola – 0 0 0
Pepsi 1 – 0 0
Sprite 1 1 - 1
Limca 1 1 0 -
Rank Order
- Rate the features of a fitness tracker in the order of importance for you (No two features can have the same rank)
- Step measurement
- Calorie tracker
- Heartbeat tracker
- Phone syncing
- Notifications of calls and messages
Constant Sum
- Please allocate 100 points to how you would want to spend your vacation budget
- Transport……………………………………………………………………………………………………………………….$
- Lodging ……………………………………………………………………………………………………………………….$
- Food ……………………………………………………………………………………………………………………….$
- Entertainment ……………………………………………………………………………………………………………………….$
- Gifts ……………………………………………………………………………………………………………………….$
Continuous
Rating Scale
Likert Scale
“Please tell us the extent to which the above ad made you feel:
Not at all A little Moderately Highly Very Highly
Uncomfortable
Unsure of yourself
Scared of negative evaluation
Scared of embarrassment
Semantic
Differential
Indicate your overall attitude toward the ad on a four-item scale
Dislike 1 2 3 4 5 Like
Unfavorable 1 2 3 4 5 Favorable
Negative 1 2 3 4 5 Positive
Bad 1 2 3 4 5 Good
Reliability and Validity in Scales of Measurement
- Internal Validity (Accurate?)
- Reliability (Consistent?)
- Cause of the effect?
- Test-retest reliability
- Construct Validity
- Interrater reliability
- Convergent and Discriminant
- Internal Consistency- Cronbach's Alpha
- Manipulation of the intended factor?
- External Validity
- Generalizable?
Cronbach’s Alpha
N– Total number of items in a scale

c – average correlation between items
v – average correlation within the scale
Internal Validity
- Attrition - Blinding
- Demand artifacts - Manipulation
- Confounding - Random selection
- Diffusion
- Random assignment
- Bias
- Strict study protocol
External Validity
- Pre and post test effects - Clear inclusion exclusion criteria

- Sample features - Psychological realism
- Selection bias - Replication
- Situational factors - Calibration of the group
- Field studies
- Experimental realism

Business Research C6

Uploaded by

Copyright:

Available Formats

Business Research C6

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Business Research C6

Uploaded by

Copyright:

Available Formats

RESEARCH METHODOLOGY

Description • Labels that categorize variables: essentially a noun

Description • A ranking scale

Description • Numerical data measured using continuous intervals

Description • Numerical data with absolute zero

Comparative Rank Order

N– Total number of items in a scale

- Pre and post test effects - Clear inclusion exclusion criteria

You might also like