Transcript The Myth of the Multiple: Exploding an Urban Legend

Exploring The Comparative
Communications Effectiveness of
Advertising and Media Placement
Initial Findings
An Experimental Study Conducted By:
Dr. David Michaelson
Dr. Don Winslow Stacks
The Assumption


Editorial coverage through public
relations activities has a value that is
greater than an equivalent
advertisement
Common assumption is that value* of
public relations placements is upwards
to three time greater than advertising if
messages are the same
* Often expressed as advertising value equivalence or AVE
The Reality


To date, this increased value of public
relations activities resulting in editorial
coverage remains an assumption
Other than Michaelson & Stacks (2004),
there has never been a definitive test or
experiment to determine if an increased
value of public relations placements
actually exists
The Research Objective

Determine if editorial coverage and
advertising perform differently on key
measures





Credibility of and homophily of message
Brand knowledge
Brand image and attributes
Brand purchase intent or interest
If differences do exist, do they support the
concept that the performance of public
relations efforts is greater than advertising?
Definitions

Credibility – “ethos” the
believability or trust in a
source

Authoritativeness, e.g.,






Homophily – the
similarity between a
source and individual

Respect
Intelligence
Information
Character, e.g.,


Honesty
Reputation
Pleasant or goodness
Attitudinal


Reflects how people
think about others as
similar to themselves
Behavioral

Reflects how people
expect to behave as
similar to themselves
Definitions,

continued
Awareness


Knowledge that a
product or brand exists
However, depth of
information may be quite
limited beyond the basic
level of product or brand
recognition

Purchase Intent


The stated likelihood to
buy a particular product
or brand
This likelihood is often
based on exposure to
communications that
describe the product or
brand and its benefits
History of This Project


Initial study conducted in 2004 among
students at University of Miami
The hypothesis was that advertising and
editorial coverage have equivalent
performance on key measures was not
supported


In essence the editorial coverage may not have
greater impact or effectiveness
Findings, however not definitive


Limited geography
Small sample consisting only of students
History of This Project

Outside funding provided resources for
first phase of study that tests for the
comparative effectiveness of editorial
coverage and advertising among
national sample of newspaper readers
Research Design


Sample of 351 adults who read a
newspaper at least once a week
Sample divided into three parts:



150 respondents exposed to advertising
message only
150 respondents exposed to news article
only
51 respondents not exposed to any test
materials and function as control group
Research Design



Each respondent asked to complete selfadministered questionnaire after exposure to
test materials
Sample weighted to reflect demographic
profile of actual newspaper readers
Each test cell statistically matched to
eliminate biases in responses that may be
associated with demographic differences
Research Design


Data collected in five locations throughout the
continental U.S.
Interviews conducted in March 2006 at malls
in the following locations:





1)
2)
3)
4)
5)
Baltimore, MD
Duluth, GA (Suburb of Atlanta)
West Dundee, IL (Suburb of Chicago)
Fort Worth, TX
Santa Ana, CA (South of Los Angeles)
Test Concept



“Created” product specifically created
for this experiment to eliminate bias
that may be associated with specific
brand preferences
Product is snack food called “Zip Chips”
that has no sodium or fats
Advertising and news article created
with parallel messages
What We Learned


There are differences between the
impact of advertising and the impact of
a news article
But, these are not differences you
might have expected
What We Found


Zip Chips brand recognition
was significantly higher than
five competitors (all major
national brands)
No statistically significant
difference between ad and
editorial
 But, significantly greater
recognition than control
group
 Therefore experimental
manipulation confirmed
Zip Chip Brand Awareness
100%
92%
84%
80%
60%
40%
20%
12%
0%
Ad
Editorial
Control
What We Found

No statistically
significant difference
between ad and
editorial groups on
believability
Zip Chip Brand Believability
100%
Somewhat Believable
Very Believable
80%
60%
40%
54%
42%
20%
17%
17%
Ad
Editorial
0%
What We Found
Zip Chip Brand Information

Increased information
does not translate into
increased believability
100%
Good
Excellent
80%
60%
54%
40%
41%
20%
16%
0%
5%
Ad
Editorial
What We Found


No statistically
significant difference in 100%
purchase interest
80%
between ad and
60%
editorial
However, those reading 40%
editorial showed less 20%
variance in overall
0%
interest
Zip Chip Purchase Intent
63%
73%
Interested
Neutral
Uninterested
15%
17%
23%
10%
Ad
Editorial
What We Found

Ad and editorial
contributed equally to
perceptions of brand
attributes

On composite and
individual attribute
questions
Zip Chip Brand Attribute Ratings
4.50
4.40
Ad
Editorial
4.30
4.20
4.10
4.00
3.90
Tastes great
Friends will like
Tastes
great/healthy
Fun to eat
Go with
everything
What We Found

Preference for Zip Chips
identical regardless of
source of information
Zip Chip Brand Preference
20%
16%
12%
8%
7%
7%
Ad
Editorial
4%
0%
What We Found

High level of uncertainty
about Zip Chips likely due to
single exposure to a new
brand


Makes composite analysis of
product credibility and
product homophily
challenging
However, analysis of 14
credibility or homophily
statements commonly used
in research provided some
insight when “don’t knows”
were examined by medium

Of the 14 statements tested,
“don’t know” frequencies
were lower for editorial in 12
cases


Equivalent levels of “don’t
know” responses not found
on other measures
In addition, there was a lack
of significant differences
between ad and editorial in
12 of 14 statements tested
Homophily Statements


Credibility of Character

The product has been presented
honestly.

Based on what I know of it, this
product is very good.

This product is very consumer
unfriendly.

Based on what I know of it, I find
this product quite pleasant to use.

This product is awful.
Credibility of Authority

Based on what I know of it, this
product is an excellent choice for
me.

This product is a value for its price.

I think this product is very reliable.


Attitude Homophily

This product is something that is like
me.

People who buy this product are
very much like me.

I would purchase this product
because it reflects my lifestyle.
Behavior Homophily

This product is used by people in
my economic class.

This product reflects my social
background.

People who use this product are
culturally similar to me.
Homophily Statements
60%
56%
65%
62%
61%
61%
50%
41%
57%
50%
56%
64%
55%
57%
This Product is a v alue for its price (CA)
This product has been presented honestly (CC)
Based on what I know of it, this product is v ery good (CC)
This product is something that is like me (AH)
Based on what I know of it, this product is an excellent choice for me (CA)
Based on what I know of it, I find this product quite pleasant to use (CC)
This product is used by people in my economic class (BH)
I think the product is v ery consumer unfriendly (CC)
People who buy this product are v ery much like me (AH)
I think this product is v ery reliable (CA)
This product reflects my social background (BH)
I would purchase this product because it reflects my lifestyle (AH)
This product is awful (CC)
People who buy this product are culturally similar to me (BH)
0
20
40%
44%
35%
38%
39%
39%
50%
59%
43%
50%
44%
36%
45%
43%
40
60
Percent of Don't Knows by Version
Ad
Story
80
100
What We Found
Homophily Analysis
CredCharacter
CredAuth
AttHomoph
BehHomoph
3.70
3.60
• No significant differences
between ad/story for either
credibility measure (p>.05)
Mean
3.50
3.40
•Significant differences
between ad/story for both
homophily measures
(p<.05)
3.30
3.20
3.10
Test Cell #1 - Ad
Test Cell #2 - Story
Version
What Are The Implications?

Unwarranted Assumption or Fact? – Is
editorial coverage more effective than
advertising?



If assumption, what does this mean for the
public relations profession?
If assumption, what does this mean for
systematic public relations measurement?
If assumption, what what does this mean for
the role of public relations in the broader
communications world?
What Are The Implications?

If fact,


What is the relative value?
How does the it work?



Is it positive, neutral or negative?
Are there multiple “multipliers”?
Is the it linear?

What is the role of frequency for each medium
as it relates to relative value?
Where Do We Go From Here?



Take into account brand comparisons to
better understand the relationship
between media and outcome
Create a more complex study allowing
for multiple comparisons while
extending the two studies conducted
Refine brand credibility and homophily
measures as related to medium
Exploring The Comparative
Communications Effectiveness of
Advertising and Media Placement
Initial Findings
An Experimental Study Conducted By:
Dr. David Michaelson
Dr. Don Winslow Stacks