Green With Guilt - East Carolina University€¦ ·  · 2013-04-16Green With Guilt: Navigating the ... we serve only locally caught sustainable seafood paired with fresh organic

Green With Guilt:

Navigating the Intersection of

Morality and Marketing in Sustainable Services

Michael Giebelhausen, Cornell University

It’s good to be green

• “In contrast to descriptive norms which specify

what is done, injunctive norms specify what

ought to be done” (Cialdini et al. 1990).

Opportunity for pro-social behavior:

Opportunities for sustainable behavior:

Sustainability as religion?

• “Sustainability is a religion in need of a catechism” – McCall (2013)

Sustainability as status

• Sexton (2011), Conspicuous Conservation: The Prius Effect and

Willingness to Pay for Environmental Bona Fides

http://ww

w.tubech

op.com/

watch/10

79201

Tourism further complicates the issue

• Of Hotels: “Any establishment that sold liquor and

contained so many beds had to be sinful.” – Hotel: An American History, Sandoval-Strausz (2007)

Self-Service Sustainability:

Rewards and Risks of Green Programs

Requiring Customer Co-production

Michael Giebelhausen, Cornell University

Helen Chun, Cornell University

Joe Cronin, Florida State University

The Choice

Self-signaling

• People observe their actions in order to make

inferences about their own attitudes

(Bem 1972, Festinger 1957).

• People's behavior is, in part,

motivated by what it tells

them about themselves

(Bodner and Prelec 2002).

Self-signaling and Satisfaction

• Consumers derive greater utility when choosing a

virtuous food in the presence of vice foods. The

opposite is true when consumers select a vice food

(Dhar and Wertenbroch 2012).

=

Status seeking or self-signaling?

H1: Effect of Participation on Satisfaction

– Optional participation in a green program affects

satisfaction with the co-producing service provider

such that compared to those who are not offered

opportunities to participate in the service provider’s

green programs: a) individuals who choose to

participate are more satisfied and b) individuals who

choose not to participate are less satisfied.

< <

STUDY 1:

J D Power GSI Data

(establishing the phenomenon)

Study Design

1. No Opportunity, n=1076

• guests at hotels where green programs were present, but

those programs did not require consumer co-creation

(e.g. “Uses energy efficient light bulbs”)

2. Participated, n=3316

• guests who indicated they co-created a green outcome

(e.g. “Linen and towels changed only by request”)

3. Did not participate, n=16455

• guests who did not participate despite their awareness of

green co-creation opportunities

Study 1 Results - JD Power GSI Overall Satisfaction

F(2,6034) = 17.646, p<.001, Mno_opp = 7.836, Mparticipated = 7.966, p < .023Mno_opp= 7.836, Mdid_not_participate= 7.574, p <.001

STUDY 2:

Hotel Towel Reuse Simulation

(uncovering the mechanism)

“Injunctive pride”™

• Pride is associated with

reinforcing the pro-social

behaviors that help maintain

a positive self-concept

(Tracy and Robins 2007).

• Pride mediates the effect of

customization on willingness

to pay (Franke, Schreier and

Kaiser 2010).

H2: Injunctive Pride as Mediator

– The effect of optional participation described in H1

will be mediated by injunctive pride - a self-conscious

emotion that arises following compliance/non-

compliance with an injuctive social norm (green

program participation in this case).

Incentives

• Incentives spoil the reputational value of good

deeds, creating doubt about whether they were

performed for the incentives rather than for

themselves (Benabou and Tirole 2006 pg. 1645)

=

H3: Incentives as moderator

– There will be an interaction of rewards and optional

participation such that: a) individuals who choose to

participate will be less satisfied when a reward is

offered vs. when no reward is offered and b)

individuals who decline to participate will be more

satisfied when participation is rewarded vs. when no

reward is offered.

+ = b) a)

Study Design

• Three Conditions: 1. participation was not available

2. participant selected whether or not to participate, but a reward

was not offered (27 opted out)

3. invitation to participate was paired with a reward

• Amazon Mechaical Turk (mturk.com) – Mean age 34, 58% female, 39.3% bachelors degree

• n = 290

Study 2 Results - Hotel Satisfaction and Injunctive Pride

• 2(participation)x2(reward) Interaction (F(1,189) = 4.684, p = .032)

• Main effect of participation (F(1,189) = 10.494, p = .001)

Mediation

• Indirect Only (i.e. “full” mediation) – Interaction of participation and rewards in determining

injunctive pride (t = -1.9780, p = .0494)

– In determining satisfaction, the effect of injunctive pride was

significant (t = 14.0005, p =< .001)

– While the interaction of participation and rewards on

satisfaction became insignificant (t=-1.0596, p=.2907)

PROCESS Model 8

(Hayes 2012)

STUDY3:

Reusable Grocery Bag Simulation

(generalizabilty, incentive characteristics)

Incentive Characteristics

• It’s easier to justify the choice of a utilitarian or

virtuous product than a hedonic or vice product

(Sela, Berger, Liu, 2009).

• Cash rewards are shown to make people adopt a

clear market norm (Heyman and Ariely 2004).

H4-5: Incentive Type as Moderator

– There is an interaction between the virtue/vice characteristics of the reward and

optional participation such that individuals who choose to participate are less satisfied

when the reward represents a vice as compared to when the reward represents a virtue.

The opposite is true for individuals who choose not to participate.

– Compared to non-cash rewards, cash rewards decrease satisfaction with the co-

producing service provider for participating consumers and increase satisfaction for

non-participating consumers.

< <

< <

Study Design

• 2 (participation: did not participate, participated)

x 4 (incentive type: none, virtue, vice, cash)

– Forced choice design

• Amazon Mechanical Turk

– Mean age 32, 53% female

• n =385

Study 3 Results - Grocery Store Satisfaction and

Injuctive Pride (incentive types combined)

• 2(participation)x2(incentive) interaction - (F(1,339) = 5.508, p = .020)

• Main effect of participation - (F(1,339) = 19.903, p < .001)

Study 3 Results - Satisfaction and Grocery Store

Incentive Types

(3,2,1,-3,-2,-1) (t(5,254) = 2.284, p = .023)

Takeaway

• Create opportunities for guests to downgrade their

service (for the good of the environment).

• Don’t incentivize participation in sustainability

programs, but do encourage it.

– Make it easy

– Employ social norms

– Provide rationalizations

for the guilty

– Future research needed

Sustainable Packaging:

The effect of the “Green is Good” Prime

on food consumption behavior

Helen Chun, Cornell University

Michael Giebelhausen, Cornell University

Brian Wansink, Cornell University

vs.

Priming

• Apple logos make people creative

– Chartrand and Fitzsimons (2008)

• Fancy restaurant pictures make people polite

– Aarts and Dijksterhuis (2003)

• What might sustainable packaging make people do?

Green Packaging

Green primes goodness?

STUDY 1:

Ordering at the drive-through

(do we need an audience)

Study Design

• 2(packaging: no mention, sustainable)

x 2(consumption context: private vs. public)

• Amazon Mechanical Turk

– 58% female , mean age = 37

• N=288

"Welcome to Charlie's. Before taking your order, I want to let you know that we

now use 100% recycled paper packaging for all orders."

Study 1 Results – effect of sustainable

package on consumption

• Only a main effect of packaging type:

F(1,283)=4.34, p < .04

• Only a main effect of packaging type:

F(1,283)=5.35, p < .03

STUDY 2:

Movie Popcorn

(real eating of different food catagories)

Study Design

• 2(packaging: regular, sustainable)

x 2(food type: unhealthy, healthy)

• Lab study with student participants

• N = 158

• The movie theater is considering switching to a new butter-

flavored topping for its popcorn and wants to see if people

would prefer it to the original butter-flavored topping.

• The movie theater is considering adding a healthy popcorn

option and wants to see which one of two different types of

healthy popcorn people would prefer.

Measures

• DV: Amount of popcorn consumed

– More on that later

• Manipulation checks

“healthy” “butter”

Study 2 results: popcorn consumption

• Main effect of popcorn type (F(1,151) = 6.88, p < .05)

• Popcorn type X packaging interaction effect (F(1,151) = 3.77, p= .05)

Post-hoc Study on Perception of Healthiness

STUDY 3:

morning snack

(replication of popcorn study)

• 2 (package type: sustainable vs. control) X

2 (food type: healthy vs. unhealthy)

• Online study for course credit

• N = 139

Study Design

Study 2 Results: Order size

• Controlled for health consciousness

• Food type X packaging interaction effect (F(1,134) = 6.65, p < .01)

STUDY 4:

process

(priming vs. self signaling)

Green primes goodness?

Takeaway

• Seemingly insignificant aspects of the servicescape

(related to sustainability) can change behavior.

• Sustainability messaging makes it less likely that

people will indulge

– For some services this might

be a good thing

– For others, decidedly bad

– Future research needed

• What makes them happier?

Green With Guilt:

How feeling Guilty Reduces Receptiveness

to Sustainable Services

Michael Giebelhausen, Cornell University

Stacey Robinson, East Carolina University

Baggage (not the suitcase kind)

• We all have it

Will guilt repel or attract people to sustainability?

• Negative State Relief Model

– Baumann, Cialdini, and Kenrick 1981

• If you save a tree in the forest and there is no one

around to see it, does it improve your mood?

STUDY 1:

Lawn Service

(sloth vs. social norms)

Study 1: Virtue vs. Vice MOTIVES

• 2(green messaging: no, yes) x 2(motive: vice, virtue)

• 120 participants

• Hypothesis: There will be an interaction of green messaging and service consumption motive such that green messaging will decrease consumption of vice-motivated services products and increase consumption of virtue-motivated services

Pat has been working long hours and feels bad about the lawn, given

that the neighbors' lawns are always trimmed and tidy.

Pat would much rather sit on the couch and watch TV than spend time

cutting the grass.

Pat is considering hiring a lawn service and looks at the business card

below that was left on the mailbox last week:

Interaction: p=.037

Virtue: p=.217

Vice: p=.015

Controlling for conscientiousness

Preliminary Results

STUDY 2:

NC Sustainable Seafood

(invoking guilt)

Laying on the guilt

Cognitive dissonance anyone?

“Unrelated Study”

• A relative of Pat’s is getting married in the Outer Banks, a chain of islands off

the North Carolina coast. It’s going to be a quick trip so Pat is attending solo. Pat

arrives at the hotel and finds that check-in is at 2:00. It’s currently noon and Pat

has not had lunch. In the hotel lobby, Pat glances at a newspaper and sees an ad

for a seafood restaurant. It reads:

•

Come try the $11.95 in-and-out lunch special at the Ocean Harvest Restaurant.

At Ocean Harvest, we serve only locally caught sustainable seafood paired

with fresh organic produce.

If you're not feeling like fish, we also have a number of vegetarian and vegan

options.

Preliminary Results

STUDY 3:

Hijacked Benefit Auction

(virtue vs. vice products)

Preliminary Results

Takeaway

• Consumer reception of green marketing depends on

where they are coming from.

• Green is good and taking the good with the bad is

hard to do

– Service operators must be aware

of what drives people to consume

their service

– Not everyone can go green

In conclusion…

• Variety of ways for green messaging to affect

consumer behavior

– Explicit program

– Servicescape

– Customer characteristics

• Green is/isn’t good

• Reseach / Channel your customers

Questions / Comments?

Scales

• Injuctive Pride (Original)

Please indicate the extent to which you might feel this way while

staying at this hotel:

Unethical / Ethical

Immoral / Moral

Selfish / Altruistic

In the wrong / In the right

Ashamed / Proud

Irresponsible / Responsible

Wicked / Virtuous

More Group Comparisons

• Contrast codes were used to test a pattern whereby, in

addition to a main effect of participation, a reward reduced

satisfaction for participants, but increased it for non-

participants (2,1,-2,-1), (t(3,189) = 2.917, p = .009)

• H1 Pairwise comparisons specified all <.02

• When compensation was offered, there was no significant

difference between the control group and those who

declined to participate in a green program (p = .345).

Not a Mood Effect

• Mood

• Neutral Object Satisfaction Questionaire

• Neutral Image Evaluations

More Group Comparisons

• Of participants in the green program, only those who were

compensated with a virtue incentive were significantly

more satisfied than participants in the control condition.

– (Mcontrol = 7.381, Mparticipated_virtue = 7.873, p = .039)

• Among those who did not participate in the green program,

only those for whom compensation was not offered were

less satisfied than the respondents in the control condition.

– (Mcontrol = 7.381, Mdidnotarticipate_nocompensation = 6.561, p = .002)