Please enable JavaScript.
Coggle requires JavaScript to display documents.
7.3 Consumer Decision Making, Research Methods for Jedetski et al - Coggle…
7.3 Consumer Decision Making
Models of Decision Making
Utility Theory
- You make a rational decision that focuses on the likely outcomes. For example, when buying a
car
, you look at the mileage, price, size, etc
Satisficing theory
- Consumers make decision where they get approximately what they want. For example, they get something that is 'good enough' for them. For example buying a simple
kettle
.
Prospect theory
- It is an item that is more precious when you own it and gains and loses are considered differently. For example,
dada's coin collection
Strategies of Consumer Decision Making
Compensatory Strategies
- Willing to compensate on certain attributes
Equal weight strategies
- All attributes are equally important
Weighted Additive strategies
- Some are more important than others
Non Compensatory Strategies
- each value is assessed individually
Satisficing
- Good enough product - For example the kettle
Lexigraphic
- When the most important attribute to a
product exceeds the cut off
value and you get that product. And that's where the decision making stops
Elimination by aspects
- This theory assesses each attribute individually where each product has a cut off value and once that attribute reaches that cut off value, they move on to the next attribute. Basically is the product attribute meets minimum requirement, then they move on to the next attribute
Partially compensatory
- Middle ground between the 2
Majority of Conforming Dimensions
- The person will compare 2 items and select the one that does the best. Then they compare that to next one and so on until they find the one that does the best.
Frequency of Good and Bad Features
- All products are compared to the appropriate cut off value and the one with the most positive features are shown
Example Study - Jedetski et al
The researchers were interested to see the decision making strategies that people use on an online platform. The researchers used
compensatory & non compensatory
. The researchers predicted that the
number of alternatives
would determine what strategy they use. They also predicted that p's would use compensatory strategies for websites that allow comparison and non compensatory for the other
There were a total of
24 p's
and they were asked to read a report on the different types of decision making strategies (3 comp and 3 non comp) and had to complete a quiz. Later, the p's had to visit 1 out of 2 websites where they would have to make a decision about buying a product and need to speak their process out loud.
P's also had to complete a questionnaire about how satisfied, frustrated, confident they were about their decision, while the experimenter recorded the time taken to make decision and the item they chose.
The results showed that the p's were more likely to choose
compensatory strategies
for website that allow comparison like
CompareNet
and
non compensatory
for websites that didn't allow like
Jango
. The number of alternative also affected their decision making process as p's were more likely to choose
compensatory strategies for less than 30
alternative and non compensatory for less than 100. Lastly, there was not a significant difference between the 2 strategies in term of confidence, satisfaction, frustration on the questionnaire. P's wer more satisfied using CompareNet than Jango
Choice Heuristics
Heuristics are mental shortcuts that help us to make decisions quickly but this can often be inaccurate and lead to errors of judgement
Availability Heuristic
It is the shortcut that is based on the how easily examples come to mind but this can often be based on inaccurate information and faulty thinking. For example, if you buy a product that keeps working badly, you might thinking that company isn't that good
Communication/ word of mouth can also lead to this heuristic.
Representative Heuristic
When you judge something by comparing a product to the an already
established
brand. For example, if a person compares the latest brand of phone to an already established company. And if the features are similar, they might think the quality is similar and as good as the established brand
Recognition Heuristic
It is a heuristic where when you compare 2 products, you are more likely to choose the product that you recognize because you are familiar with it. For example, when choosing between Lays and Bingo, you are more likely to choose Lays because you might recognise that
Take the Best Heuristic
It is a heuristic that is based on
only one factor or attribute
. That one attribute is the most important and ignores all the other information. For example when buying a phone, a person with this heuristic might only focus on one aspect like the camera quality. And if they find the one that has the best, their decision making process stops.
Anchoring
It is a heuristic that is based on the first bit of information that we receive and the decision you make will result based on the info you recieve. It's basically like the first impression of the product. For example, if you are buying a car and the price is really high on the website, and if you go to the car showroom and the price is lower you might get that.
POINT OF PURCHASE DECISIONS
Multiple unit price
is when the price is reduced for when you are buying several items. This can increase sales and can attract customers
Suggestive selling
is when the customer is 'suggested' to get something else. For example, Amazon's 'Customers also buy option'.
EXAMPLE STUDY - Wansink et al (1998)
They wanted to see the factors that influence how many units a customer buys. To test this, they conducted 2 field and 2 lab experiments, (we will look at one of each) that investigated the effects of
suggestive selling and Multiple unit prices
For the
field experiment
, it was conducted in
86 stores
with the original price displayed (99 cents) and the Single unit price (75 cents) or Multiple unit price (2 for 1.50). The shelf size was the same and there were
13 common food items
. The results showed that there was a
32% increase
in sales in
Multiple unit promotional pricing
.
12 of 13 products had their sales increase
and 9 of them were showed to be statistically significant.
However, no self report data was gathered and they only showed the increase in sales. We do not know how many items they got and so the sales could indicate that either the same customer bought many items or many people bought many things and thats why sales increased.
The other study conducted was to test the
effectiveness of suggestive selling
. In this case it was a new and innovative way to enjoy items, such as 'Snickers bars - buy them for your freezer'. Effects of suggestive selling when
accompanied with or without prize discounts.
The study involved a shopping
scenario
of
120 UG students
where they were presented with 6 well known products with one of 3 prize levels: actual convenience store price, 20% discount, 40% discount. P's were given suggestive selling
slogans with no product quantity anchor
(Snickers- buy them for your freezer)
and explicit product quantity anchor
( Snickers - buy 18 for your freezer). The p's were not given an indication on whether it was a discount or not and were asked to give purchase quantity intention for all products.
The results showed that
suggestive selling and discount increased purchase quantity intentions
. Suggestive selling even increased the intended purchase quantities without a discount.
Research Methods
Field Experiment - high ecological validity, so will behave in a natural way.
No self report - there is no qualitative data, so no way of knowing the reasons for the purchase nor how they were feeling.
Second experiment included only UG students which means that there is less representation and less generalisability. This is important area for consumer behaviour because there will be a wide range of customers.
Applicable to real life because it shows the effectiveness of promotion techniques and suggestive selling. This info can be useful to sales people and shop owners to increase their sales. It can also help the consumers as they understand the psychology behind the sales.
Decision Making styles
EXAMPLE STUDY - del Campo et al (2016)
The aim of the study was to see the relationship between
heuristics and decision making styles
. There were 2 experiments conducted - one in
Vienna
and one in
Madrid
.
The sample consists of
320
p's and an approximate split of men and women.
P's had to complete a
computer based task
and they were asked to choose between 5 different options of egg. The 5 options were based on different heuristics such as - take the best heuristic (emphasized the price only
2.99
), recognition heuristic (had the brand name), emotional (contained a chicken picture), cognitive (lot of info) and a filler task. The p's were either
under time pressure (40 seconds) or not
.
After the task, the p's were given a questionnaire where the researchers measured the
reasons
for their purchase,
buying
behavior
, and their
attitudes
towards the product. The researchers used Scott and Bruce's instrument to see the decision making style
The results showed that there were a lot of difference culturally between the samples. Firstly, the Vienna Experiment,
time pressure
made the p's shift their decision from cognitive to take the best heuristic, whereas this was not the case with the Madrid experiment where there was little evidence to suggest that time pressure increased the likelihood of using a heuristic. Secondly, in the Vienna experiment, those who scored
high on the spontaneous
decision making styles where more likely to choose the
recognition heuristic
more than cognitive. However, those who scored high on the
rational
score, chose the
take the best option
rather than the cognitive one regardless of time pressure. In the
Madrid experiment
, those who score high on the
rational
were likely to chose the
cognitive
without the time pressure but this was reversed with time pressure.
According to
Scott and Bruce
there are 5 decision making styles.
Rational
- Making decisions in a logical way and keeping the option open for a good purchase
Intuitive
- Purchasing something that 'feels right', making a decision based on intuition
Spontaneous
- Making an quick and impulsive decision
Dependent
- Relying on others for information and assistance for a decision.
Avoiding
- Putting off decisions or making decisions last minute
The instrument devised by Scoot and Bruce gives each a person a score on each of the above dimensions.
The use of a heuristics can be influenced by the task, context and personal factors such as personality and traits. Time pressure also a factor that can influence heuristics.
Mistakes in Decision Making
Thinking Fast and Slow
According to Kahneman, there are 2 different styles of thinking -
System 1 and System 2
.
System 1 refers to "
thinking fast"
, which is making decisions in an intuitive, unconscious and effortless way, often involving heuristics and associations. This can be useful to use as decisions are made quicker and easier so our lives can be easier.
System 2 refers to "
thinking slow"
, which is making decisions in a slow, conscious and controlled manner, often relying on statistics and requires time and attention.
Kahneman and Tversky claim that system 1 describes "normal" decision making and its easier to use. Its not only used for insignificant decisions but for big decisions as well. However, since it relies on heuristics it can lead to errors as well. Whereas for system 2, we do not use it that often.
There are a lot of difficulties when considering System 1 and 2 thinking. Firstly, these
parameters of making decisions can vary between people
. For example, when calculating 20 X 20, for a mathematician this is a lot easier and can use System 1 thinking. However, for some, it requires them to calculate and it may take some time, thus relying on System 2 thinking.
Secondly, we focus a lot on how System 1 can cause errors, however,
System 2 can also can errors
. For example, when calculating the 20 X 20, some people despite taking their time and calculating can still make mistakes. Thus, when we see an error, we are not sure whether it is cause by system 1 or 2 or both.
Lastly, most information that is considered during system 2 comes from system 1. For Example, what might actually trigger someone to engage in system 2?
These question remain unanswered and can make it difficult to predict decisions.
KEY STIUDY - HALL ET AL - CHOICE BLINDNESS
Context
Choice blindness is an error in decision making where
someone does not notice the difference between what they have chosen and the outcome
they got. For example, Johansson et al, showed participants photos of 2 women and asked them which one was more attractive. After picking, the researchers swapped the photos and that was given to the participant. They were then asked to justify their reason for their decision. Only in 26% of the did the p's noticed that it was not the one they chose
Hall et al want to see whether choice blindness went beyond visual processing and human faces. They were interested to see how this can be
applied to real life
.
Main theories and Explanations
Our decision making process can either be due to
implicit influences
or
explicit influences
. Implicit influences includes preference fluency (how easy or difficult it is to make the decision) and placebo effect (where people tend to believe that the product will be less effective at a reduced price).
Explicit influences is when your decisions are made based on what you actually see and hear about the product
Thus, choice blindness is interesting to investigate because it considers the interaction between explicit choices and implicit choices
AIM
The aim of the study is to see whether
choice blindness
would be
shown with consumer decision making in a natural setting
.
Design
It was a
field experiment
where the tasting venue was set up in a Swedish supermarket. The sample consisted of
180 p's
out of which 118 were females. The age range was from 16-80 years old and it was an
opportunity sample
where p's were asked if they would like to be involved in a
quality control test for pairs of tea or jam.
The p's were asked to taste the jam and smell the tea and say which did they prefer. Half of the p's were told that they would receive the chosen tea or jam as a gift. For each of the p's, the
order of presentation
, the
type of manipulation
, and
which tea or jam was used
was randomly allocated.
There were
2 experimenters - Exp 1
managed the recording device, asked questions and took notes whereas
Exp 2
conducted the preference test. For each p's either the tea or the jam condition was manipulated. In the
manipulated condition
, the p's were asked to taste the jam or smell the tea and
rate it
on a
scale of 1 -10
(not at all very good, very good). While rating, Exp 2 flipped the jar and the p's were asked to try the second option and rate on a scale of 1 - 10 and the Exp 2 again flipped the jar. Now the p's were asked to taste the
preferred jar
again, this time
defending
their choice, rating on a
10 point scale
,how difficult they found the decision and how confident they were of the decision. They were also asked if they found anything odd or unusual.
Then the p's were fully debriefed, with the aims explained and asked if they suspected any manipulation.
There were
3 criterias for detection of manipulation
- one was
concurrent detection
(where they noticed a change after tasting or smelling),
retrospective detection
( where they noticed the deception after the experiment, before or after the deception), and
sensory change
( where the p's did not report anything but made comments about the change). Those with the concurrent detection, their results were excluded in the report analysis.
Results
Concurrent detection
- 14.4% for jams, and 13.8% for tea
Retrospective detection
- 6.2% for jams, and 6.9% for tea
Sensory change
- 12.4% of jams and 11.5% of tea
Overall, 33.3% of manipulated trial in the jam condition and 32% of MT in the tea condition were detected.
Those who were offered tea as gifts were less likely to detect any manipulation compared to the jam condition
.
No difference in the perceived ease of distinguishing the 2 samples in the MT & NMT. The same goes for the confidence in decision making.
Results supported the main prediction - that no more than
1/3rd of the manipulated trials were detected
. This shows that in most of the conditions p's were blind to the mismatch between their intended and their actual decision. Moreover, in
2/3rds of detected trials
, p's noticed a change only after tasting through
sensory change
. Even for very different smells like Mango and Pernod, p's made detections in less than half of the times and 1/5th of them were concurrent
Furthermore, in the
NMT and the NDMT the ease of distinguishing between the 2 samples was high with
7.3 for jams and 8 for tea
. This shows that p's could tell the sample apart but could not detect any manipulation.
This suggests that choice blindness remains at a significant level in a real life setting.
Evaluation
Opportunity sampling
- it is quick, easy and convenient to gather the samples as it saves time and energy. There is also a reasonable spread of people in an opportunity sample (16-80). However, we cannot control what type of sample it is so it could be less representative.
Use of Pilot study
- this was done to see if the samples were distinguishable or not. This increases validity to check for choice blindness
Consumer memory for advertising
Research shows that people are more likely to be able to remember information for a period of time is learning takes place. There are 2 types of interferences that affect the consumers memory -
Retroactive interference
(when info is hard to recall because of new info that is being learnt) and
Proactive interference
(when prior learning affects the person's ability to recall and learn new info).
Retrieval failure
is when forgetting occurs due to retroactive and proactive interference.
McKinney investigated the role of the
retroactive interference
in advertising by asking p's to learn a one full page magazine advert. Half of the p's were given a rest period whereas the other half had to learn up 2 more adverts (work period). The p's were then given a
cued recall test
and the results showed that p's in the rest condition were more likely to remember the advert (94%) than the work condition (72%). This suggests that subsequent (after) learning affected their ability to recall the original advert.
Blankenship and Whitely investigated the effects of
proactive interference
in advertising by asking p's either to learn one list of 18 store items and prices or to learn this list and learn a previous list. The results showed that after 48 hours, those who had learned one list were more likely to recall the items that compared to those who learned a previous list, suggesting that previous info interfered with recall information
Example Study - Burke and Srull
They wanted to see whether a consumer's ability to recall distinctive brand info would be hindered if they viewed another similar product by a different brand, or another product of the same brand. They wanted to see how the p's memory was affected by the person's
information processing objective
(reason why they are looking at the product).
There were 2 experiments conducted. In
Exp 1
the p's were shown 12 adverts (3 of which were target one shown at the beginning) and their information processing objective was manipulated. They were asked to look at the advert and see which one they are more likely to buy or how interested they thought the advert was.
The 12 adverts were either similar products by different brands, different products of the same brand or a range of different products and brands. After viewing the p's were given a
distraction task
where they were given a questionnaire about magazine and TV viewing habits. After the distraction task, the p's were given a surprise recall task were they were given the target brand name and product were given 2 minutes to recall any information.
The results showed that the p's were able to
recall twice as much when they were asked about their intention
rather than rating the advert on interest. Results also showed that the
recall was significantly lower
for those who viewed the adverts for
similar products or same brand
.
This shows
retroactive interference
because new learning affected their ability to recall the target adverts. However, since the Info processing objective was manipulated, it
mediated this effect
and showed that p's viewing the products with a purchase intent were less likely to get affected from the interference.
Exp 2
was conducted a similar way to investigate the effects on
proactive learning
and the target adverts were shown after the filler products. The result were similar to Exp 1 - those rating on purchase intention recalled more . However, unlike Exp 1, they did not find that IPO had a mediating effect on the adverts
Research method and design
Lab experiment - highly controlled as they control extraneous variables that can affect the study. This makes it highly reliable and valid. It was also standardized
Research Methods for Jedetski et al
Use IMD - No order effects which means that they are less likely to figure out the aim of the study, which means that it increases validity. However, they compare the groups, so there is a risk of p's bias
Consumer decision making takes a nomothetic approach to establish universal laws and rules to explain consumer behaviour. This is useful as it enables a good understanding about how people are likely to behave. Nomothetic approach requires large amounts of data so this is more likely to be more accurate
P's were informed about the decision making strategies so that they could describe the process. This reduces ecological validity, meaning that they are less likely to see it in a natural way