Lecture 5 measures Flashcards
measures for attitudes
- verbal versus nonverbal
- self-report versus observational
- direct versus indirect
- reactive versus nonreactive
- explicit versus implicit
verbal versus nonverbal measures
with verbal measures people can respond to questions and express their opinion. you can also measure non-verbal responses because people don’t always want to share some things.
self-report versus observational
there are different ways to self report
- thurstone scale
- likert scale
- osgood’s semantic scale
and different observational measures
thurstone scale
procedure developed by thurstone in “attitudes can be measured”:
- generate 100-150 evaluative statements
- reduce them to the 80 best ones
- ask about 300 people to place them on 11-point scale
- cut out statements with high variance
- select two or three statements
- participants say with wich statement they agree
- attitude score = mean of values of endorsed attitudes.
very complex procedure, but structured.
Likert
came up with an easier way
- prepare 100 statements toward attitude object
- pre-test on a likert scale (agree or disagree)
- drop items with low item-to-total correlations and retain about 20
- administer final set of items to other samples
- attitude is total score
we use averages, because it is an easier score to interpret
Osgood’s semantic differential
identified three dimensions of meaning to measure attitude
- evaluation
- potency
- activity
are self-reports actually measures?
a feeling on 1-10 scale is a made up number, it has no objective base.
but this was studied and it showed that giving just a single number has a greater value than a whole set of variables.
self report satisfaction was in this study a good predicter of behavior which suggests self-report is a useful measure.
observational measures
- observe behavior
- facial muscular activity
- physiological measures
with this you can find out things people don’t want to say.
direct versus indirect measures
- indirect = a method in which a researcher gathers data about one variable as a means of representing a second variable of interest that cannot be measured in a more straightforward manner
- direct might not even be possible
direct versus indirect attitude measurements
- direct = any procedure for assessing attitudes that require a person to provide a report of their attitude
- indirect = any procedure for assessing attitudes doesn’t require a person to provide a report of their attitude
indirect measures methods
- error-choice
- projective techniques
- list experiment
error choice
ask questions, but only give incorrect answer options. which is given reveals something about attitude
projective techniques
for example give shopping list with different type of things and ask opinion about the person that wrote the shopping list.
list experiment
give people a list of things that people would like or dislike and ask them how many they dislike. Then give them a second list with one extra thing and ask them how much they dislike
are indirect measures better?
even though they have been developed a lot, you might sometimes be better of just asking. some people won’t be honest, but this error is smaller than when you do all the indirect measures.
reactive versus non-reactive measures
- reactive = present attitude object and record response
- non-reactive = observe traces.
explicit versus implicit measures
we have explicit and implicit attitudes.
- implicit attitude = no awareness, no control of response
- implicit attitude measure = infer attitude from automatic responses without awareness that attitude is being measured.
evidence for implicit attitudes?
a large number of studies about if changing implicit attitudes leads to change in behavior have been analyzed and this was not the case.
explicit attitude change leads to behavioral change, but implicit does not. (except when procedures where used that involved threat).
also not the case for behavior that is mediated by implicit attitudes.
why are implicit attitude measures not valid?
- most known measures aren’t
- low correlation between measures of same attitude
- measures aren’t accurate
common measures used in marketing practice
- semantic differential/evaluative dimension
- net promoter score
- willingness to pay
semantic differential/evaluative dimension
using a scale with different aspects of a product to evaluate it. gives good idea about consumer attitude and easy to use.
net promoter score
if you would recommend a product or service to someone else. this is intuitive
willingness to pay
- direct self-report = how much are you willing to pay?
- willingness to buy = would you buy X for €?
- BDM
- Van-Westendorp
- choice based conjoint analysis
BDM
kind of auction with price randomly determined. if bid>price people have to buy it. if bid< price individual can not buy.
Van-Westendorp
asks you to give very specific rating
choice based conjoint analysis
participants make several choices between products with specified attributes. these attributes vary across choices. willingness to pay can be estimated with conjoint analysis.