Taking Measurements

Measurement scales • Nominal scale – Bus #; Gender? • Ordinal scale – House numbers; Rank in class • Interval scale – Complex, maybe Grades • Ratio scale – Heart rate; any measure in % Scales & Statistics • Can you take an average of A nominal scale? – Gender – • Think about the meaning of the measurements! Scales • In most cases when asking people for their opinions researchers behave as if they have ratio scales, wishing they had interval scales but in fact they have an ordinal scale. • It is also important to use measurement scales that people find meaningful – If people can distinguish only 3 levels, don’t use an 100 point scale Two types of measures • “Subjective” – Introspection – Intuitions – Opinions • Objective – Behavioral – What SS actually do • Examples: • How much do you enjoy this class? • Who came to class today? • Which is better? Why? Objective measures • • • • • • • Observing behavior (parking, men's room) Choice Payments, bids, etc Time in task Limited to things we can observe Statistically expensive in many cases (choice) • usually harder to get Subjective measures • Intentions – What would you do – What would others do • Attitudes – How much do you like X? • Verbal reports – On-line introspections Quantitative subjective measures • Likert scale – 1-7; 1-10; 1-100 • Guttman scale – A yes no response to a continuum of questions that are ordinal • Hypothetical choice • Ranking • • Quantitative ≠ objective Fresh-Samantha Paradox Fresh Samantha Flavors Average Rating Carrot 2.7 33% Orange Carrot 3.0 0% Orange 2.3 33% Soy Shake 2.0 33% 1 worst • Market Share 2 3 4 best Can anyone come up with an explanation for this data? Highest average ranking, but 0 market share (4 points) 1st Choice (3 points) 2nd Choice (2 points) (1 point) 3rd Choice 4th Choice (A) Carrot Orange-Carrot Orange Soy Shake (B) Orange Orange-Carrot Carrot Soy Shake (C) Soy Shake Orange-Carrot Carrot Orange What is the origin of this paradox? • Ratings do not translate directly to behavior Subjective measures • Very popular • Cheap • Can tell us things we cannot observe (why..) • • Social desirability demands • Demand characteristics (effects) • • Depend on interpretation • Intuitions & not behavior • Depends on the extend to which people can introspect Subjective measures I • Social desirability demands – How often do you cheat? – Would you steal if you were sure no one will catch you? – Would you rather watch “the piano” or Rambo? – • Demand characteristics (effects) – Do you agree that red is a nice color? Subjective measures II • Individuals have problems with: – Reporting their intentions – Predicting their own emotions in a different state – Interpreting own emotions Reporting intentions • Hypnosis interpretation • ....... • – People can easily give reasons for their behavior but it is not clear that the reasons articulated have anything to do with reality Predicting I • Predicting emotions & actions in a different state (VanBoven & Loewenstein) before exercising after exercising Thirst before hunger in essay 19% 50% Thirst more unpleasant 61% 92% Would regret more not packing water 61% 88% Thirst more unpleasant for hikers 57% 88% Hikers would regret more no packing water 52% 92% Predicting II Interpreting own emotions • • • • • Run because you are afraid? Afraid because you run? Festinger -- Cognitive dissonance Running up the stairs makes people believe the experimenter is attractive. • Crossing a bridge… • The movies, an ideal place for a date… 3 cognitive stages in answering questions interpret question • retrieve information & form judgment map judgment onto response scale "File drawer" view: – Attitudes are stable dispositions that are stored in long-term memory, and you just retrieve them • "Construal" view: – Attitudes are constructed on the fly, computed when necessary from whatever information comes to mind Do people interpret the question as intended? • What brand of soft drink do you typically buy? – – – – – interpret question What is the time period involved? (How far back do I go … I bought Gatorade in the summer; Hi C when I was a kid) Do purchases at movies, in restaurants, & sporting events count? Does "you" refer to me or to my household? What is a "soft drink" exactly? • do lemonade, iced tea, fruit punch, & mineral water count? Are Coke and Cherry Coke the same brand? retrieve information & form judgment map judgment onto response scale Can people retrieve the relevant information? • How many ounces of alcohol did you consume last year? – What would be a better question to ask? interpret question retrieve information & form judgment map judgment onto response scale How do people select responses? • How successful are you? – on a scale from 1 to 10? – • Is 10 the valedictorian of your high school? Ted Turner? Michael Jordan? interpret question retrieve information & form judgment map judgment onto response scale Effects of response scale format How successful have you been in life ? 0 1 2 3 4 5 6 7 8 9 very not at all -5 not at all -4 10 -3 -2 1 0 1 2 3 4 5 very Response scale provide cues for the "normal" levels of a behavior How many hours a day do you spend watching TV ? 7% 18% _____ _____ 26% 15% _____ _____ 18% _____ 16% _____ up to .5 1 to 1.5 2 to 2.5 more than 2.5 .5 to 1 1.5 to 2 How many hours a day do you spend watching TV ? 62% 23% _____ _____ 8% 5% _____ _____ 2% 0% _____ _____ up to 2.5 2.5 to 3 3 to 3.5 4 to 4.5 more than 4.5 3.5 to 4 Effects of scales: • In many cases the scales we give people are not ones they are to think about (even hours of watching TV) • In such cases people use the scales to determine the acceptable range and their subjective fit within the range to determine their response Response scale provide cues for the "normal" levels of a behavior How many times do you floss in a typical day ? 0 1 2 3 4 5 6 7 8 9 10 >10 How many times do you floss in a typical year ? 0 1 2 3 4 5 6 7 8 9 10 >10 How likely are you to develop gum disease? How much would you pay for a bottle of Scope mouth wash? Another example…. The importance of context 1. How bad is it to jaywalk? 2. 0 1 2 3 4 5 6 7 8 9 10 not at very all bad bad 1. How bad is it to murder your professor? 2. 0 1 2 3 4 5 6 7 8 9 10 not at very all bad bad 3. 4. How bad is it to spray paint the side of a building? 3. How bad is it to spray paint the side of a building? 4. 5. 0 1 2 3 4 5 6 7 8 9 10 not at very all bad bad 0 1 2 3 4 5 6 7 8 9 10 not at very all bad bad – – Ease of recall affects liking Condition 1 • 2) 3) 4) 5) 6) 7) 8) 9) 10) 11) List 10 reasons to buy a BMW: __________________ __________________ __________________ __________________ __________________ __________________ __________________ __________________ __________________ __________________ Condition 2 List 1 reason to buy a BMW: 1) __________________ • – Effects of scales: • The constructive view of preferences: – People do not know but if you ask them they will give you an answer – And they will be sure about it – But ….. • In some cases people”learn” about themselves from their own answers Extracting sensitive information • • Do you happen to have done X in the last year? As you know, many people have been doing X these days. Do you happen to have done X? Do you know anyone who has done X? How about yourself? • • • Pull a coin from your pocket. If it lands Heads, answer Question 1. If it lands Tails answer Question 2: – 1. Is your birthday between January 1st and June 1st ? 2. Have you ever done X? 0.5 "YES" 25% H 0.5 65% 0.5 "NO" 25% "YES" 40% "NO" 10% 0.5 T Have done "X" 0.8 0.2 Have not done "X" Question Order I • • • • How satisfied are you with your life overall? How is the weather? How satisfied are you with your life overall? – – In one of these conditions, was there a strong correlation between current weather (whether it was raining outside or sunny) and people's appraisal of their life as a whole. – – In which of these two conditions was the correlation observed. Question Order II • How many dates did you have last • How many dates did you have last month? • How satisfied are you with your life, in general? – R = 0.66 – • How satisfied are you with your life, in general? month? – R = 0.0 Question Wording • The government is spending too little on "welfare" » Yes • No 25% 75% • • The government is spending too little on "assistance to the poor" • • • Yes 65% No 35% Response language • Asking questions is like a discussion • People make assumptions about the intended meaning • Summary • Subjective vs. objective (behavioral measures) – The main question is when are people able to report – Sensitivity to scales, order, etc. – Response language • The constructive view – Do people have preferences? – Inferences of attitudes from questions / answers – Awarness based on questions •