Instructiontoevaluators(lnthesurvey,thefollowinginstructionisgiventoevaluators.)1.This experiment uses scales by adjectives and examines what kinds of meaningare given to a certain object.Please evaluate objects using yourKansei.2.Donotreviewthepreviousevaluation.Evaluatetheobjectindependentlyaccording to each scale.Please carry out the evaluation at a fast speed.3.Thinkabouthowwelltheobiectfitsintotwoevaluationwords.Ifneither.checkthecenter.Pleasenoticethattheintervalsareevenlyspaced.Ifthisdoesnothold,youcannotusethedataforsomestatisticalanalysis.6712345口口口口口口口CoolWarmCooeutraWaWarVerySlightlySlighth11
Instruction to evaluators (In the survey, the following instruction is given to evaluators. ) 1. This experiment uses scales by adjectives and examines what kinds of meaning are given to a certain object. Please evaluate objects using your Kansei. 2. Do not review the previous evaluation. Evaluate the object independently according to each scale. Please carry out the evaluation at a fast speed. 3. Think about how well the object fits into two evaluation words. If neither, check the center. Please notice that the intervals are evenly spaced. If this does not hold, you cannot use the data for some statistical analysis. 11
Fourtypesofscales (Reasonwhyequidistanceis emphasized)(1).Nominal scale:Scalefordistinguishingthings:Bloodtype:StudentID number (there is no superiority orinferiority in alphabetical order)Ordinalscale:Scaleof order (wheninterval cannotbequantified):Orderofmedicine'seffect.Order offavorite cake+一×-Fourarithmeticoperationscannotbeperformed.12
Four types of scales (Reason why equidistance is emphasized)(1) • Nominal scale: Scale for distinguishing things • Blood type • Student ID number (there is no superiority or inferiority in alphabetical order) • Ordinal scale: Scale of order (when interval cannot be quantified) • Order of medicine’s effect • Order of favorite cake Four arithmetic operations cannot be performed. 12 + − × ÷ > >
Fourtypes of scales (Reason why equidistance is emphasized)(2)·Intervalscale:Scalewiththeinterval having meaning (theratiohasno meaning): The difference of 10 degrees between the temperature 10 °C and 20 C has meaning,butthelatterisnottwiceaswarmastheformer.Deliciousnessofcake (verydelicious=4,delicious=3,bad=2,verybad=1)Equidistance is usuallyassumedinKansei·Proportional scale:Scalewiththe ratio having meaningevaluation experiment.Bodyweight(60kg istwiceasheavyas30kg.)·Distance(30kmisthreetimesfartherthan10km.):Absolute temperature (Temperature is alsoa proportional scale if-273.15 isusedas the standard.)Absolute zerodegree13
Four types of scales (Reason why equidistance is emphasized)(2) • Interval scale: Scale with the interval having meaning (the ratio has no meaning) • The difference of 10 degrees between the temperature 10 ℃ and 20 ℃ has meaning, but the latter is not twice as warm as the former. • Deliciousness of cake (very delicious = 4, delicious = 3, bad = 2, very bad = 1) • Proportional scale: Scale with the ratio having meaning • Body weight (60 kg is twice as heavy as 30 kg.) • Distance (30 km is three times farther than 10 km.) • Absolute temperature (Temperature is also a proportional scale if -273.15 ℃ is used as the standard.) 13 Absolute zero degree Equidistance is usually assumed in Kansei evaluation experiment
Why bipolar scales?.Warm←→NotwarmTherearevariousconceptsof"notwarm."Warm←>ColdIf you do notdeterminetheWarm←>Lukewarmcounter word,"what youareevaluating"isambiguous.Warm←→CoolWarm←Chilly14
Why bipolar scales? • Warm Not warm There are various concepts of “not warm.” If you do not determine the counter word, “what you are evaluating” is ambiguous. 14 • Warm Cold • Warm Lukewarm • Warm Cool • Warm Chilly
Data screeningExpertsConsumers10.2Traditional<>Modern12345671234567Traditional<>ModernTraditionalModern.We seefromtheabovefigurethatgeneral consumers do notknowpreciselythedifferencebetweentraditionalcupsandmoderncups.Insuchacasewehavetodecidewhetherwedonotusethepair<traditionalmodern> at all, or whether we use thedata reflecting the experts'opinions only.15
Data screening • We see from the above figure that general consumers do not know precisely the difference between traditional cups and modern cups. • In such a case we have to decide whether we do not use the pair <traditional, modern> at all, or whether we use the data reflecting the experts’ opinions only. Experts Consumers Traditional Modern Traditional Modern -3 -2 -1 0 1 2 3 -3 -2 -1 0 1 2 3 Traditional Modern 15 1 2 3 4 5 6 7 1 2 3 4 5 6 7