COMPARATIVE SCALES

• 
  

• SCALING TECHNIQUES

• COMPARATIVE SCALES

• NON COMPARATIVE SCALES

• NET PAPER 1
• MARKETING RESEARCH/RESEARCH METHODOLOGY
• CLASSIFICATION O SCALING TECHNIQUES
• 
  
• MEANING OF COMPARATIVE SCALES

1. MUST BE INTERPRETED IN RELATIVE TERMS AND HAVE ONLY ORDINAL AND RANK ORDER PROPERTIES
2. INVOLVE THE DIRECT COMPARISON OF STIMULUS OBJECT
3. LIKE WHETHER YOU PREFER PEPSI OR COKE
4. ALSO REFEREED TO NON METRIC SCALING

• THE BENEFITS :
• SMALL DIFFERENCE BETWEEN STIMULUS OBJECT CAN BE DETECTED
• EASILY UNDERSTOOD
• CAN BE APPLIED EASILY
• INVOLVE FEWER THEORETICAL ASSUMPTIONS
• DISADVANTAGES
• COMPARATIVE SCALES INCLUDE THE ORDINAL NATURE OF THE DATA AND INABILITY TO GENERALIZE BEYOND THE STIMULUS OBJECT
• CLASSIFICATION OF COMPARATIVE SCALES :
• 
  

1. PAIRED COMPARISON
2. RANK ORDER
3. CONSTANT SUM
4. Q SORT AND OTHER PROCEDURES

• NON COMPARATIVE SCALES

1. ALSO REFERRED TO MONADIC OR METRIC SCALE
2. EACH OBJECT IS SCALED INDEPENDENTLY OF THE OTHERS IN STIMULUS
3. DO NOT COMPARE THE OBJECT BEING RATED EITHER TO ANOTHER OBJECT OR TO SOME SPECIFIED STANDARD
4. EVALUATE ONLY ONE OBJECT AT TIME
5. RESULTING DATA ARE GENERALLY ASSUMED TO BE INTERVAL OR RATIO
6. LIKE RESPONDENT MAY BE ASKED TO EVALUATE COKE ON 1 TO 6 SCALE

• CLASSIFICATION OF NON COMPARATIVE SCALES

1. CONTINUOUS RATING SCALES
2. ITEMIZED RATING SCALES
3. LIKERT
4. SEMANTIC
5. STAPEL

• COMPARATIVE SCALES TECHNIQUES
  PAIRED COMPARISON

1. RESPONDENT IS PRESENTED  TWO OBJECTS AT TIME
2. ASKED TO SELECT ONE OBJECT IN THE PAIR ACCORDING TO SOME CRITERION
3. DATA OBTAINED ARE ORDINAL IN NATURE
4. N(N-1)/2 PAIRED COMPARISON

• ANALYSIS :IN SEVERAL WAYS:

1. CALCULATE THE PERCENTAGE
2. EVALUATION OF ALL THE STIMULUS SUBJECT IS ALSO POSSIBLE
3. POSSIBLE TO CONVERT PAIRED COMPARISON DATA TO RANK ORDER
4. ASSUMPTION OF TRANSITIVITY OF PREFERENCE

• EXAMPLE
• 
• RANK ORDER SCALING

1. RESPONDENTS ARE PRESENTED WITH SEVERAL OBJECTS SIMULTANEOUSLY AND ASKED TO ORDER OR RANK THEM ACCORDING TO SOME CRITERION
2. USED TO MEASURE PREFERENCE FOR BRANDS AS WELL AS ATTRIBUTES
3. DATA IS OBTAINED FROM RESPONDENTS IN CONJOINT ANALYSIS
4. IF THERE ARE N STIMULUS OBJECT ONLY N-1 DECISIONS WOULD BE REQUIRED
5. THE MAJOR LIMITATION IS THAT PRODUCES DATA ONLY ORDINAL DATA

• EXAMPLE
  
• CONSTANT SUM SCALING

1. RESPONDENTS ARE REQUIRED TO ALLOCATE A SUM OF UNITS,SUCH AS POINTS,DOLLARS,OR CHIPS ,AMONG A SET OF STIMULUS OBJECTS WITH RESPECT TO SOME CRITERION. EXPLAINED WITH THE EXAMPLE ON NEXT SLIDE
2. BENEFITS: ALLOWS FINE DISCRIMINATION WITHOUT USING MUCH TIME

• DISADVANTAGE :

1. MAY ALLOCATE OR OR FEWER UNITS THAN THOSE SPECIFIED
2.  ROUNDING ERROR

• CONSTANT SUM SCALING
• 
  
• Q-SORT AND OTHER PROCEDURE

1. Q-SORT SCALING WAS DEVELOPED TO DISCRIMINATE AMONG A RELATIVELY LARGE NUMBER OF OBJECTS
2. USES A RANK ORDER PROCEDURES IN WHICH OBJECTS ARE SORTED INTO PILES BASED ON SIMILARITY WITH RESPECT TO SOME CRITERION
3. RESPONDENTS ARE GIVEN 100 ATTITUDE STATEMENTS ON INDIVIDUAL CARDS AND ASKED TO PLACE THEM INTO 11 PILES RANGING FROM MOSTLY HIGHLY AGREED WITH TO LEAST HIGHLY AGREED WITH
4. THE NUMBER OF OBJECTS TO BE PLACED IN EACH PILE IS  PRESPECIFIED