Attitudes, beliefs and behaviour

1
Attitudes, beliefs and behaviour

• Week 4 21 May 2003

2
Predicting behaviour

• Measurement of
• Attitudes
• Beliefs
• Intentions
• Objectives
• Predicting sales
• Influencing consumers

3
Attitudes

• the strength of the consumers' belief with
  regard to the company image or brand
• positive
• negative
• neutral
• 3 components
• cognition (knowledge)
• affectivity (emotions and feelings)
• conation intention (behaviour).

4
Measuring attitudes

• E.g. semantic differential scale question
• I find that eating at the Blue Room is

Bad
Good
5
Attitudes and beliefs

• An attitude (to actions) can be seen as a set of
  beliefs
• Beliefs mental and verbal ideas and assessments
  we have and we make about the world we inhabit
  they will be of varying strengths

6
Attitudes and beliefs

• The walk to the Blue Room is nice
• The food at the blue room is warm
• The price is reasonably low
• Eating at the Blue Room is good

7
Measuring beliefs, evaluations and outcomes

• Contracting salmonella from eggs is
• I think that the consequences of salmonella are

Unlikely
Likely
(b)
-2 -1 0 1 2
(e)
Mild
Strong
-2 -1 0 1 2
8
Outcomes

• Outcome (b) (e)
• Recoding from (-2 to 2) to (1 to 5)
• E.g.
• Salmonella is likely and with strong
  consequences 25 (2)
• Salmonella is unlikely and mild 1 (-2)
• To go back to the original scale
• Square root
• Subtract 3

9
The properties of attitudes

• Valence (positive, negative, neutral)
• Extremity (intensity)
• Resistance (immunity to change)
• Persistence (erosion through time)
• Confidence
• I like a lot eating at McDonalds, because it is
  fashionable

10
Multiattribute analysis and the Expected-value
model
Utility measurement?

• A attitude towards the product
• bi strength (likelihood) of the belief that the
  product has attribute i
• ei evaluation of the attribute i
• n number of salient attributes

11
Does it work?

• Fishbein interviewed 50 subjects and asked them
  attributes evaluation and a final overall
  (global) assessment
• The correlation between the A score and the
  global assessment was 0.8

12
Salient beliefs

• Salience importance assigned to an attribute
• Problem salient attributes differ across
  consumers, can we submit a single questionnaire
  to many consumer?

13
Eliciting salient beliefs

• Define the action and target group
• Elicit salient beliefs, asking Is there anything
  else?
• Consider negative actions (beliefs for not
  acting)
• Who should (not) do the action?
• What are the control factors?
• Combine similar beliefs
• Refine the list of beliefs

14
Exercise

• Answer to the following questions with a value
  between 2 (least) to 2 (most), with 0 as the
  indifference/zero value
• (b1) Do you think there are vitamins in orange
  marmalade?
• (e1) What do you think is the effect of the
  vitamins in orange marmalade?
• (b2) What colour do you prefer in orange
  marmalade?
• (e2) How important is the colour for you?
• (b3) What is the orange cut you prefer in orange
  marmalade?
• (e3) How important is the cut for you?

b1e1
-2
-1
0
1
2
b2e2
-2
-1
0
1
2
-2
-1
0
1
2
-2
-1
0
1
2
b3e3
-2
-1
0
1
2
A
15
Salience

• Divide your score by three
• What would be your score if you just consider the
  attribute which is most important for you?

16
The Ideal-Point model

• A attitude towards the product
• Wi weight (importance) of attribute i
• Ii ideal performance on attribute i
• Xi belief about actual performance on attribute
  i
• n number of salient attributes

17
Attitude and behaviour

• Behaviourism (reinforcement paradigm)
• Thought and feelings are effects not causes of
  behaviour
• Attitude data allow to predict behaviour, not to
  explain it
• Cognitivism (cognitive paradigm)
• Attitudes and knowledge control behaviour
• Experience changes attitudes and knowledge
• Communication may modify behaviour

18
Changing attitudes through advertising

• Change beliefs
• Correct misperception (e.g. price)
• Comparative advertising
• Change attribute importance
• Reduce the importance of poor attributes
• Increase the importance of good ones.
• Change ideal point
• Change the target ideal good

19
(No Transcript)
20
(No Transcript)
21
Frequently purchased goods

• Usage precede intention to buy
• Intentions depend on past usage
• Consumers tend to associate positive attributes
  with a brand if they are currently using it

22
Experience, information attitudes

• It is more likely that attitudes change after
  trial (direct experience) rather than because
  of advertising (indirect experience)
• Direct experience provides a stronger link
  between attitude and behaviour
• Attitudes learned by experience are more
  accessible (strong)
• Speed of response
• Confidence in the evaluative judgment
• Attitude stability
• Repetition strengths attitudes

23
Predicting behaviour

• Attitudes could also be unrelated or just
  slightly related to actual behaviours
• Other variables may interact
• Personal (in)ability
• Social constraints
• Uncertainties about outcomes
• Discrepancies when measures are taken at
  different times (information changes)

24
Measuring attitudes

• Three components model

Observable independent variables
Inferred variables
Observable dependent variables
25
Problems with the3 components model

• The attitude concept includes evidence of actual
  behaviour
• Inferring attitudes from behaviour might be wrong
• Attitude towards the product
• Attitude towards purchasing the product (higher
  correlation with behaviour)

26
The compatibility principle

• Attitudes to the purchase of the product must be
  measured if purchase is the object of prediction
• Compatibility between measures of attitude and
  behaviour

27
Examples of the compatibility rule

• Attitudes towards (adapted from East)
• Pizza
• Mozzarella tomato pizza
• Buying takeaway MT pizza
• Buying takeaway MT pizza from Pizza Hut
• Buying takeaway MT pizza from Pizza Hut tonight
• You buying takeaway MT pizza from Pizza Hut
  tonight

Target
Action
Context
Timing
Personal aspect
28
Purchase intentions and market research

• Predicting new product sales
• First purchase can be correctly predicted through
  likelihood of purchase
• Further purchases depend on the experience
  associated with first purchase
• Distinguishing between user and non-users
• Works better for frequent purchases

29
Theories for predicting purchasing behaviour

• Fishbein (1963), Ajzen and Fishbein (1980), Ajzen
  (1985, 1988,1991)
• Expected value theory of attitude
• Compatibility
• Other variables
• Theory of reasoned action
• Subjective norm
• Theory of planned behaviour
• Perceived behavioural control

30
Theory of Reasoned Action

• Economic rationality
• Perfect knowledge
• Selfishness
• Optimality
• Reasoned action
• Limited knowledge of outcomes
• Accessible outcomes kept into account
• Normative influence of other people
• Limited power to actually act like intended
• Intentions rather than actions are predicted

31
Theory of reasoned action
Outcome beliefs
Referent beliefs
Attitude to behaviour (AB)
Subjective norm (SN)
W2
W1
Structural equation modelling
Intention
Behaviour
32
Sufficiency

• All change in attitude and behaviour derive from
  new beliefs or modified beliefs
• External factors act only indirectly (through
  attitudes and subjective norms) on behaviour
• Problem past experience usually has a direct
  effect on behaviour

33
Theory of Planned Behaviour (TPB)

• Inclusion of other variables
• Moral norm
• Own personal values
• Perceived behavioural control
• Self-perceived ability of taking the desired
  action (confidence)
• Prediction of behaviour rather than intention

34
Examples

• Attitudes towards GM foods (Cook et al.)
• Gene technology in tomato production (Saba and
  Vassallo)
• Organic food consumption (Shepherd and Raats)

35
Measurement

• Seven-point semantic differential scale

My eating of tomatoes produced by gene technology
in the future will be . .
Extremely harmful
Extremely beneficial

36
Measurement

• Outcome beliefs (expected-value model)
• Referent beliefs (similar to the ev model)
• Normative belief (My friends think that)
• Motivation to comply (likelihood to comply)
• Control beliefs
• Controlling factor (e.g. having the money)
• Access to the control factor (probability to have
  it)
• Global variables
• AB (two or more scales) For me
• SN Most people who are important to me think
• PBC For me doing this is (semantic scale)
  difficult - easy

37
Sum variables
Outcome beliefs
Referent beliefs
Control beliefs
Global variables
Attitude to behaviour (AB)
Subjective norm (SN)
Perceived behavioural control (PC)
W2
W1
W3
Intention
Behaviour

• PC is a good predictor of intention, its
  inclusion significantly improves the model
• SN is the weakest predictor

38
Explanation Behaviour (level 1)

• Intention
• Perceived behavioural control (weaker)

39
Explanation Intention (level 2)

• Relative weight of AB, SN and PC
• Changes according to product / situation

40
Explanation Specific factors (level 3)

• Specific outcome
• Referent factors
• Control factors

Global variables Intention
E.g. Complaining about a product Is it most
related to AB, SN or PC? Identifying a limited
number of factors can help marketing strategies
41
Limits

• Difficult to select salient belief
• Low correlation between sum and global variables