BUS 362 - Fall 2011 AUBG, Department of Business BUS 362 LECTURE 12: Understanding Measurement Prof. Milena S. Nikolova Department of Business OUTLINE Question Response Format Basics of Measurement Scales 2 1 BUS 362 - Fall 2011 AUBG, Department of Business 1. Establish need for marketing research 2 Define the problem 2. 3. Establish research objectives 4. Determine research design 3 5. Identify information types and sources 6 Determine methods of accessing data 6. 4 2 BUS 362 - Fall 2011 AUBG, Department of Business QUESTION RESPONSE FORMATS INGREDIENTS OF QUESTION DEVELOPMENT Reasoning Creativity 6 3 BUS 362 - Fall 2011 AUBG, Department of Business BASIC QUESTION-RESPONSE FORMATS Open-ended Categorical Scale-response 7 BASIC QUESTION-RESPONSE FORMATS OPEN-ENDED | Respondent is not presented with a response options | Could be: y Unprobed format -- seeking no additional information | Advantage: -- Allows respondent to use his or her own words | Disadvantages: g -- Difficult to code and interpret -- Respondents may not give complete answers 8 4 BUS 362 - Fall 2011 AUBG, Department of Business BASIC QUESTION-RESPONSE FORMATS OPEN-ENDED EXAMPLE: What is your favorite brand of toothpaste? 9 BASIC QUESTION-RESPONSE FORMATS OPEN-ENDED | Respondent is not presented with a response options | Could be: y Probed format -- includes a response probe instructing the interviewer to ask for additional information | Advantage: -- Elicit complete answers | Disadvantages: -- Difficult to code and interpret 10 5 BUS 362 - Fall 2011 AUBG, Department of Business BASIC QUESTION-RESPONSE FORMATS OPEN-ENDED EXAMPLE: What is your favorite brand of toothpaste and why? Anyy other brand you y like? 11 BASIC QUESTION-RESPONSE FORMATS CLOSED-ENDED (CATEGORICAL) | Respondent is provided with options on the questionnaire that can be answered qquickly y and easily y | Could be: y Dichotomous -- has only two response options (“yes” or “no”) | Advantage: -- Simple to administer and code | Disadvantages: -- May oversimplify response options 12 6 BUS 362 - Fall 2011 AUBG, Department of Business BASIC QUESTION-RESPONSE FORMATS CLOSED-ENDED (CATEGORICAL) EXAMPLE: Are you interested in trying the new toothpaste “Shining Smile”? Yes No 13 BASIC QUESTION-RESPONSE FORMATS CLOSED-ENDED (CATEGORICAL) | Respondent is provided with options on the questionnaire that can be answered qquickly y and easily y | Could be: y Multiple response has more than two options for the response Advantage: g off possible p responses p -- Allows ffor broad range -- Is simple to administer and code | Disadvantages: -- Must distinguish “pick one” from “pick all that apply” -- May alert respondents to response options, which they were unaware of | 14 7 BUS 362 - Fall 2011 AUBG, Department of Business BASIC QUESTION-RESPONSE FORMATS CLOSED-ENDED (CATEGORICAL) EXAMPLE: Which of the following toothpaste brands are you most likely to buy? Whiteness “Whiteness” “Shining Smile” “Good morning” 15 BASIC QUESTION-RESPONSE FORMATS SCALED RESPONSE | Utilizing a scale developed by the researcher to measure the attributes of some construct under studyy | Could be: y Natural -- using a scale that may be purely numerical or readily understood measure (years, times, etc.) | Advantages: g off intensity/feelings yf g to be -- Allows ffor degree expressed -- Is simple to administer and code | Disadvantage: -- Respondents may not relate well to the scale 16 8 BUS 362 - Fall 2011 AUBG, Department of Business BASIC QUESTION-RESPONSE FORMATS SCALED RESPONSE EXAMPLE: How many times do you go shopping per month? One Two Three Four or more 17 BASIC QUESTION-RESPONSE FORMATS SCALED RESPONSE | Utilizing a scale developed by the researcher to measure the attributes of some construct under studyy | Could be: y Synthetic -- using a made-up continuum (agreement/ disagreement, intention, etc.) Advantages: g of intensity/feelings y g to be -- Allows for degree expressed -- Is simple to administer and code -- Respondents can relate to scale | Disadvantage: -- Scale may be “forced” or overly detailed | 18 9 BUS 362 - Fall 2011 AUBG, Department of Business BASIC QUESTION-RESPONSE FORMATS SCALED RESPONSE EXAMPLE: How likely are you to buy the new “Whiteness” toothpaste instead of your favorite brand next time you go shopping (scale of 1 to 5)? 1 – Highly unlikely 2 – Somewhat unlikely 3 – Neutral 4 – Somewhat likely 5 – Very likely 19 BASIC QUESTION-RESPONSE FORMATS 20 10 BUS 362 - Fall 2011 AUBG, Department of Business CONSIDERATIONS IN CHOOSING A QUESTIONRESPONSE FORMAT | The nature of the property being measured y | Previous research studies y | Use format in previous study if desire to compare The ability of the respondent y | Gender = dichotomous; liking for chocolate = scale Children can’t relate to scaled response The scale level desired y Yes/ no (% of shares) but How many times? (average) 21 BASICS OF MEASUREMENT 11 BUS 362 - Fall 2011 AUBG, Department of Business BASIC CONCEPTS IN MEASUREMENT | Questionnaires – tools for collecting of information via measurement | Measurement -- determining how much of a property (attributes or qualities) is possessed by an object | Properties: specific features or characteristics of an object that can be used to distinguish it from another object Objective properties are physically verifiable y Subjective properties are mental constructs y 23 MEASUREMENT LEVELS | Objective properties y Physically verifiable characteristics age, gender, number of bottles purchased, etc. | | | Observable and tangible Often predetermined as response options Subject ve Subjective y Cannot be directly observed because they are mental constructs person’s attitudes, opinions, or intentions | | Unobservable and intangible Relying on respondents to translate mental constructs onto an intensity continuum (using synthetic scales) 24 12 BUS 362 - Fall 2011 AUBG, Department of Business MEASUREMENT PROCESS THE OBJECT Person Product Brand C Company PROPERTIES MEASUREMENT RESULT Age How old are you? _______ years 35 years Gender What is your gender? ___Male ___Female Female Preferred Brand Which is your favorite brand? __Coke __Pepsi __Fanta Fanta Opinion How do you rate our brand? __Poor __Good __Excellent Good SCALES: CHARACTERISTICS & LEVELS 13 BUS 362 - Fall 2011 AUBG, Department of Business SCALE CHARACTERISTICS AND LEVEL OF MEASUREMENT Scale development designing questions and response formats to measure the subjective properties of a subject | Scale characteristics determine the scale level of measurement | 27 SCALE CHARACTERISTICS | Characteristics of scales 1) Description: the use of a descriptor descriptor, or label label, to stand for each “unit” unit on the scale (all scales include description): “yes,” “no,” “male,” “female,” etc. Order: the relative sizes of the descriptors are known, allowing us to say one is “greater/less than” the other: less preferred brand, more preferred brand Distance: the differences between the descriptors are known: there is a $1-difference between $4 and $5. $5 There is a 10-degree difference between 90 and 100 degrees. Origin: there is a true, natural zero: there is a zero level of dollars, market share, sales 2) 3) 4) 28 14 BUS 362 - Fall 2011 AUBG, Department of Business SCALE CHARACTERISTICS Each scale characteristic builds on the pprevious one If a scale has order, it also has description… if it has distance, it also has order and description… If a scale has a higher-level property, it has all lower-level properties 29 LEVELS OF MEASUREMENT SCALES Scale characteristics determine the level of measurement Nominal scales: -- using only labels, possess only description | Ordinal scales: -- with which the researcher can rank/order the respondents or responses | Interval scales: -- in which the distance between each descriptor is equal | Ratio scales: -- ones in which a true zero exists | 30 15 BUS 362 - Fall 2011 AUBG, Department of Business LEVELS OF MEASUREMENT SCALES Scale characteristics determine the level of measurement | Nominal scales: Gender, Race, Religion, Buyer/Non-buyer, etc. | Ordinal scales: Rank brand in order of preference; Use dry cleaning once a month, twice a month, more than twice | Interval scales: | Ratio scales: Car brand rated 4 – car brand rated 5; Representative rates are very friendly – rated as generally friendly How many car purchases are registered today; How many dollars do you spend per day (can compare and construct ratios) 31 LEVELS OF MEASUREMENT SCALES 32 16 BUS 362 - Fall 2011 AUBG, Department of Business IMPORTANCE OF LEVEL OF A MEASUREMENT SCALE | Importance: The scale affects what may or may not be said about the property being measured y The scale determines which statistical analysis you may use y 33 EXERCISE | | | | Group 1: y You are marketing experts for a small university deciding on marketing plan for next five years; Your survey tries to determine which marketing channels to use: internet, radio, other? Group 2: y You are marketing experts for a telecommunications company deciding how to launch promotion for new services for teenagers; Your survey tries to determine whether the promotion of new products needs to be targeted at parents or the teenagers. Group 3: y You Y are marketing k ti experts t for f a TV channel h l launching l hi a new show h targeted t t d att young professionals; Your survey tries to determine whether to air show on weeknights or during weekend Case 4: y You are marketing experts for a luxury sports gear company deciding whether to launch a big sale to compensate low sales level; Your survey tries to determine whether price promotions will impact the luxury brand in a negative way 34 17 BUS 362 - Fall 2011 AUBG, Department of Business WORKHORSE SCALES | Frequent need to measure psychological aspects of consumer behavior such as attitudes, opinions, evaluations, beliefs, impressions, perceptions, feelings, and intentions. | Need for means that allow the expression of direction and intensity of subjective properties – synthetic scale scale-response response questions 36 18 BUS 362 - Fall 2011 AUBG, Department of Business WORKHORSE SCALES USED IN MARKETING RESEARCH | Workhorse Scales: y y y Scaled response questions with an interval scale format (metric) Sometimes numbers indicate a single unit of distance between the positions on the scale Usually (not always!) the scale ranger from extreme negative, through neutral, and to extreme positive | Neutral is not a zero or origin but a point along the continuum Negative Neutral Positive Extremely dissatisfied No opinion Extremely satisfied Strongly disagree Neither agree nor disagree Strongly agree 37 WORKHORSE SCALES USED IN MARKETING RESEARCH | Workhorse Scales: The Likert Scale The Life-Style Inventory y The Semantic Differential Scale y y | y Halo effect Other Scaled-Response Question Formats 38 19 BUS 362 - Fall 2011 AUBG, Department of Business LIKERT SCALE | Respondents are asked to indicate their degree of agreement or disagreement on a symmetric scale for each statement | Statements need to be simple and plain | Modified depending on needs (statements, intervals, options, etc.) LIKERT SCALE 20 BUS 362 - Fall 2011 AUBG, Department of Business THE LIFESTYLE INVENTORY Register values and personality traits through evaluation of unique activities, interests and opinions of respondents – measure consumers’ unique way of living | Could include questions on: -- price awareness and attitude, fashion awareness and attitude, activeness in expressing opinion, family orientation, social preferences, etc. | | Example: the VALS Survey THE SEMANTIC DIFFERENTIAL SCALE Features a series of bipolar adjectives for the various properties of the property at study; respondents evaluate through points along continuum | Focus: measure meaning of object, concept, person | Phrases are flipped to avoid grouping of positive or negative only on one of the sides | y | Avoiding the halo effect (bias associated with overall feeling towards object at study) Result – plot profile of object image 21 BUS 362 - Fall 2011 AUBG, Department of Business THE SEMANTIC DIFFERENTIAL SCALE THE SEMANTIC DIFFERENTIAL SCALE 22 BUS 362 - Fall 2011 AUBG, Department of Business OTHER SCALE TYPES | Summated scales: – includes a list and respondents p select as manyy answers as apply pp y to them producing a metric measure (total number) y y | Example: Select each of the appliances that you possess: Example: Mark each of the countries you have visited for leisure purposes: Anchored scales: -- endpoints are identified as by the opposite ends of the measurement continuum and are associated with the beginning and ending numbers Example: Please rate the quality of the customer service representative you just interacted with: 1-unfriendly and 5-very friendly | Unanchored scales: -- endpoints are not identified Example: Please rate the friendliness of the customer service representative you just interacted with a number from 1 to 5 45 ISSUES WITH SYNTHETIC SCALES | Including a neutral point y | Using symmetrical scales y | Some respondents may have not formed an opinion but some respondents may be using neutral as a dodge or a way to hide opinion Fort certain categories symmetrical does not make sense (important/ unimportant) and some groups may respond better to an unsymmetrical scale Cultural specifics y Some cultures tolerate extremism in opinion, others value “going with the crowd” 23 BUS 362 - Fall 2011 AUBG, Department of Business RELIABILITY AND VALIDITY | Reliability: -- respondent responds in the same or a similar manner to an identical or nearly identical measure | Validity: -- accuracy of responses to a measure y Face validity 47 EXAMPLE Travel Experience p & Patterns ACT DUR H1 ; H2 Retrospective Global Evaluation B h i Outcomes Behavior O t H4 PEAK GVAL WOM REV TRN H3 Post-Travel Brand Experience BVAL ACT = Activities DUR = Trip Duration TRN = Overall Trend PEAK = Peak Intensity GVAL = Global Value BVAL = Brand Value Control Variables Demographics Negative Events 48 WOM = Word of Mouth REV = Repeat Visit 24 BUS 362 - Fall 2011 AUBG, Department of Business EXAMPLE Hedonic Dimensions Scale Voss, K. E., Spangennerg, E. R., & Grohmann, B. (2003). Measuring the hedonic and utalitarian dimensions of consumer attitude. Journal of Marketing Research, XL(August), 310-320. Not fun Dull Not Delightful Not Thrilling Not Enjoyable Fun Exciting Delightful Thrilling Enjoyable 49 EXAMPLE Behavior Outcome Scale Zeithaml, V Zeithaml V. A A., Berry Berry, L L. L L., & Parasuraman Parasuraman, A. A (1996). (1996) The behavioral consequences of service quality. Journal of Marketing, 60(2), 31-46. - Likelihood to say positive things. - Likelihood to recommend . - Likelihood to encourage friends and relatives. relatives - Likelihood to go back to the destination and engage in the same activities and/or visit the same regions you visited during past trip(s)? - Likelihood to go back to the destination and engage in different activities and/or visit different regions from the ones included in your past trip(s)? 50 25 BUS 362 - Fall 2011 AUBG, Department of Business EXAMPLE 51 Thank You! Milena S. Nikolova, PhD mnikolova@aubg.bg 26