Determinants of littering: an experimental analysis.
Khawaja, Fatima Salim Ki ; Shah, Anwar
Littering, the improper disposal of small quantities of waste, is
one of the main causes of environmental degradation. To protect the
environment from this degradation, we need to factor out the
determinants of littering behaviour. In this study, we conduct a
controlled laboratory experiment to examine whether people would avoid
littering if the social cost of this behaviour was internalised. Based
on the microeconomic theory relating to externality, we test whether
penalising littering decreases its level compared to when it is not..
The results indicate that when the cost of production of littering is
internalised, the level of littering drops.
JEL Classification: C92, D62
Keywords: Litter, Externality, Private Cost, Internalisation,
Experiment, Baseline Treatment, Cost Treatment
1. INTRODUCTION
The careless and improper disposal of small quantities of waste is
defined as littering [Geller, el al. (1982); Stokols and Altman (1987);
Keenan (1996)]. Keep America Beautiful (2009) shows that improper litter
disposal poses a threat not only to human, animal, and plant life, but
also to man's economic prosperity. Raffoul, et al. (2006) show that
along with polluting surface and ground water, littering is one of the
main causes of transmission of diseases like dengue and leptospirosis.
Additionally, littering makes the environment unpleasant, causes vehicle
accidents and puts children at risk of cuts and infections in public
places like parks and grounds.
Studies show that various factors lead to littering. For example,
people litter when they are too lazy to dispose their waste properly
[Ojedokun and Balogun (2011)], when they argue that everyone else is
doing it [Campbell (2007)] and when there is lack of waste bins nearby
[Williams (1997); Keep America Beautiful (2009); Ojedokun and Balogun
(2011)]. The findings of other researches also suggest that litter
begets litter [Dur and Vollaard (2012)], and that littering persists
when the private cost of disposing litter correctly is very high
[Torgler, et al. (2008)].
Most of the available literature uses field surveys for data
collection which makes it difficult if not impossible to isolate the
impact of a certain variable of interest without controlling for the
other factors. In this study we isolate the impact of the
internalisation of the social cost of littering on the level of
production of littering by using a controlled laboratory experiment. In
a controlled laboratory experiment it is possible for the experimenters
to examine the impact of a single factor on littering while controlling
for the remaining factors that may influence littering behaviour.
Our experiment has two treatments: in the Baseline treatment, the
participants do not bear the private cost for littering whereas in the
second treatment (which we term as the 'Cost treatment'), each
participant bears the private cost for littering. Based on the
microeconomic theory regarding externality, we hypothesise that an
individual will litter more when littering has no private cost as
compared to when there is a private cost associated with littering. This
private cost may take various forms such as disapproval by others
(social cost), remorse and guilt (conscious cost), jail or doing
community service (legal cost), or fines etc. (monetary cost).This study
examines the impact of monetary cost only. The rationale is that
monetary cost is easier to quantify in comparison to the rest of the
private costs. Our study finds that the internalisation of the social
cost of littering decreases the level of production of littering. On the
other hand, littering is more prevalent when individuals do not bear any
private cost for the creation of littering.
The rest of the paper is organised as follows: the next section is
devoted to literature review; Section 3 explains the theoretical
framework and the hypothesis; Section 4 discusses results, and Section 5
draws the conclusion.
2. LITERATURE REVIEW
A volume of studies is available on the issue of littering. Some
studies address the question 'why people litter?', some
examine the factors affecting the amount and frequency of littering,
while some discuss ways to improve littering behaviour for a cleaner
environment. A complete review of such literature is beyond the scope of
the present study; hence we provide a brief review of studies relevant
to our research question in the following paragraphs.
Various researchers have explored the role of laziness in littering
behaviour. For example, a study conducted in Nigeria shows that
littering is an automatic and routine behaviour as it is easy and
comfortable [Ojedokun and Balogun (2011)]. Campbell (2007) concludes
that many people litter simply because they do not bother to walk up to
a waste basket. Likewise, Williams (1997) states that one of the major
reasons of littering is indolence.
Dur and Vollaard (2012) test whether litter begets litter and find
that littering increases significantly when regular cleaning is not
done. Some other studies show that the presence of litter is an
incentive for further littering. [Krauss, et al. (1978); Reiter and
Samuel (1980); Heberlein (1971); Robinson (1976)]. Other contributing
factors to littering are age. Research shows that younger
people--especially between the ages of 17 and 19--tend to litter more
than older people [Beck (2007)]. Moreover, when the personal cost of
disposing waste materials correctly is high in terms of time and energy
spent, individuals litter more [Torgler, et al. (2008)].
Ojedokun and Balogun (2011) further add that low altruism, low
self-efficacy, low locus of control, and low self-concept lead to a
positive attitude towards littering whereas higher education and
awareness lead to a negative attitude towards littering. Arafat, et al.
(2007) and Al-Khatib, et al. (2009) report that socio-cultural factors
such as gender, marital status, monthly income, religious convictions,
education level, age, and type of residence all have an impact on
littering behaviour. Littering is also more frequent in situations when
the person is in a hurry, the item is biodegradable, there is a sense
that someone else will pick it up, and when the item is not recyclable
[Keep America Beautiful (2009)]. Several studies further state that
littering practice is often the result of non-availability of a trash
basket nearby [Finnie (1973) and Keep America Beautiful (2009)].
The above analysis shows that extensive research is available on
littering. However, to our knowledge, there is no study except of
Torgler, et al. (2008) which evaluates the impact of internalising the
social cost of littering. The focus of Torgler, et al. (2008) to an
extent, is close to our study; yet, their study is based on European
Values Survey, (1999-2000) and is much broader in nature. On the other
hand, we focus on a single variable and examine whether internalising
the social cost of littering decreases the levels of littering. While
the subject matter of our question has been discussed extensively in
microeconomic theory, little empirical evidence from a controlled
laboratory experiment exists.
3. THEORETICAL FRAMEWORK AND HYPOTHESIS
The study is based on the concept of negative environmental
externalities. Environmental externalities refer to the economic concept
of "uncompensated environmental effects of production and
consumption that affect consumer utility and enterprise cost outside the
market mechanism"[Glossary of Environment Statistics (1997)]. When
negative externalities are associated withan economic activity,
governments usually intervene in the form of bans, taxes, fines, etc.
The purpose of these measures is to act as deterrents and to internalise
the negative impact of economic activities on a society. Following the
basic assumption that all economic agents are rational and wish to
maximise their welfare, it makes intrinsic sense that if there is a
monetary cost associated with an action such as inappropriate removal of
waste, the economic agents will become wary of what they throw on the
ground and endeavour to curtail it in order to capitalise on their
earnings.
Based on the idea of negative externality discussed above, the
present study examines whether the negative externality of littering can
be internalised by associating a cost to the act by imposing a fine on
the litterers. The hypothesis of the study is thus: littering is likely
to be lower if the private cost of its production is high.
3.1. Experimental Procedure
The experiment consists of two treatments, namely the Baseline
treatment and the Cost treatment. The participants of the experiment are
undergraduate students belonging to the School of Economics,
Quaid-i-Azam University, Islamabad.
The participants of the experiments were selected at random by
using the ballot system to decide which semester's students would
be asked to sit for which treatment. We wrote down the number of
semesters on paper chits and put them in a box. (1) This was followed by
a blind selection of two chits from the box. Consequently, students of
third semester were asked to participate in the Baseline treatment
whereas the students of first semester were asked to participate in the
Cost treatment (all students were informed beforehand that participation
was voluntary). The experiment was conducted in October 2012. The detail
of each treatment follows:
3.1.1. Baseline Treatment
In the Baseline treatment, there were 40 participants. Each
participant was given some coloured sheets with squares drawn on them, I
pair of scissors and 2 envelopes. One of the envelopes was labelled
"Use Me for Squares" and the other was labelled "Use Me
for Waste Material". The instructions required participants to cut
as many squares drawn on the coloured sheets as possible within three
minutes and put those squares into the envelope labeled "Use Me for
Squares". (2) There was no monetary cost of littering and the
monetary compensation to the participants followed the criterion given
below:
Monetary Reward = (PKR 10* Number of completed squares)--(PKR 0
*Number of pieces of litter)
Participants were informed that 10 percent (4 out of 40) from among
them will be randomly chosen using the ballot system (all participants
were seated on numbered chairs) and paid PKR 10 for each completed
square at the end of the experiment. (3) The experiment lasted half an
hour. The randomly chosen participants were paid cash at the end. The
earnings of each of these 4 participants were between PKR 40 to 50. The
participants were not paid any show-up fee.
3.1.2. Cost Treatment
The second treatment was similar to the Baseline treatment: the
only difference was that the instructions clearly mentioned that the
monetary compensation to the randomly selected individuals would be
deducted if extra pieces of paper were found around them (i.e. if they
produced litter). As such, a private cost for littering was introduced
in this treatment. The rest of the procedure was the same as in the
Baseline treatment but with a different group of students. The
compensation to the randomly chosen participants was given based on the
following criterion:
Monetary Reward = (PKR 10* Number of completed squares)--(PKR 5*
Number of pieces of litter)
The randomly chosen participants were paid cash at the end of the
experiment. The earnings of each of these 4 participants were between
PKR 20 to 30.
4. RESULTS AND DISCUSSION
For statistical analysis, we report one-tailed Wilcoxon rank-sum
test as the data of our study does not come from a population with a
normal distribution. The reason to report one tailed Wilcoxon rank-sum
test is that we have ex-ante hypothesis about the impact of cost on
littering.
Table 1 shows the descriptive statistics across treatments. The
table compares the mean, mode, and standard deviation of the number of
completed squares, the number of litter pieces inside the waste
envelope, and the number of litter pieces outside the waste envelope
across treatments. The mean number of squares cut by each individual in
the Baseline and Cost treatments is 4.4 and 4.1, respectively. However,
the mode remains the same at 0 in both treatments. On the other hand,
the mean number of litter pieces inside the waste envelopes in the
Baseline and Cost treatments is 1.0 and 2.18, respectively, while the
mean number of litter pieces outside the waste envelope in the Baseline
and Cost treatments is 7.25 and 4.75, respectively. In addition, there
is a sharp drop in the mode from 9 in the Baseline treatment to 0 (zero)
in the Cost treatment. It appears that internalisation of cost does
decrease the level of litter outside the waste envelope; however, some
participants show careless behaviour towards littering even after
incorporating the cost. This is evident from the standard deviation of
the level of litter outside the waste envelope in the Cost treatment as
compared to the Baseline treatment.
Although Table 1 shows a summary of the results across treatments,
the detailed behaviour of each participant is unobservable. Therefore,
we show the frequency distributions of each of the three decisions made
by participants in both treatments.
4.1. Distribution of Completed Squares Across Treatments
Figure 1 presents the distribution of squares across treatments.
Taking the number of squares as independent observations, the Wilcoxon
rank-sum test shows that the distribution of the number of squares in
the Baseline treatment is not significantly different than the
distribution of number of squares in Cost treatment (p = 0.16).This
result shows that internalising the social cost of littering does not
significantly affect the efficiency of participants in terms of the
production of squares.
[FIGURE 1 OMITTED]
4.2. Distribution of Litter Pieces Inside the Waste Envelope Across
Treatments
Figure 2 presents the distribution of litter pieces inside the
envelopes across the treatments. We can see that the number of
participants who did not use the waste envelope at all is higher in the
Baseline treatment as compared to the Cost treatment. We also observe
that more participants use the waste envelope for litter disposal in the
Cost treatment than in the Baseline treatment. The mean number of litter
pieces inside waste envelope in the Baseline and Cost treatments is 1.0
and 2.18, respectively.
Taking individual pieces of litter in the waste envelope as
independent observations, the Wilcoxon rank-sum test shows that the
distribution of proper disposal of litter in the waste envelop in the
Cost treatment is significantly higher than the proper disposal of
litter in waste envelope in the Baseline treatment (/?<0.05). This
finding is in line with our hypothesis.
[FIGURE 2 OMITTED]
This result implies that littering without care is a general
behaviour in the Baseline treatment. Conversely, this behaviour is not
too common in the Cost treatment as some of the participants are more
conscious of their littering behaviour. This shows that internalising
private cost does indeed control littering behaviour.
4.3. Distribution of Litter Pieces Outside the Waste Envelope
Across Treatments
Figure 3 presents the distribution of litter pieces outside the
waste envelope across treatments. We can see that the number of
participants who do not produce litter outside the waste envelope is
high in the Cost treatment. On the contrary, there is a fair number of
participants who littered outside the waste envelope in the Baseline
treatment. Taking the number of waste pieces outside the waste envelope
as independent observations, the Wilcoxon rank-sum test shows that the
distribution of number of waste pieces outside the waste envelope in the
Baseline treatment is higher than the distribution of number of waste
pieces outside the waste envelope in the Cost treatment (p < 0.01).
Therefore, this finding is also in line with our hypothesis.
[FIGURE 3 OMITTED]
4.4. Total Waste Disposal Across Treatments
Table 2 shows the total number of completed squares, total number
of litter pieces inside and outside the waste envelope across
treatments. We can see that the total number of litter pieces outside
the waste envelope is higher in the Baseline treatment (290 pieces) as
compared to the Cost treatment (190 pieces). Wilcoxon Rank-Sum test
shows that the sum of the number of litter pieces outside the waste
envelope in the Baseline treatment is lower than the number of litter
pieces outside the waste envelope in the Cost treatment (p<0.05). It
is also worth mentioning that the number of squares is not affected much
due to internalisation of the social cost. This shows that the
internalisation of social cost make people litter-conscious without
compromising their efficiency in the form of the production of squares.
5. CONCLUSION
This study tests whether internalisation of the social cost of
littering decreases the level of littering. We investigate the answer of
our research question by conducting a controlled laboratory experiment.
In the Baseline treatment, the participants are asked to cut as many
squares as possible within a mentioned time in return for a monetary
reward with no cost for littering. In the Cost treatment, participants
bear monetary cost for littering.
We find that most of the participants in the Baseline treatment did
not bother to open the envelope labeled "Use Me for Waste
Material". They littered the area around them. On the other hand,
participants in the Cost treatment mostly put the litter pieces in the
waste envelope without compromising on the number of squares. The
findings support our hypothesis indicating that littering is more
frequent and common when the private cost to littering is not
internalised as opposed to when there is a penalty for littering.
The results of the study have implications especially in those
areas where litter is a significant problem, e.g. academic institutions
like the Quaid-i-Azam University, Islamabad. It is suggested that in
order to discourage littering behaviour and achieve environmentally
favourable outcomes, one step in the right direction would be adequate
provision of waste baskets complemented by imposition and implementation
of fines on littering. If monetary fines are not possible, then
developing a norm among faculty and students to criticise individuals
who are littering is also likely to work. In other words, the study also
suggests that where formal institutions are weak, activities having
negative externality could be curtailed through informal institutions
such as social norms and peer effect. Nevertheless, the limitations of
the study are that littering behaviour could depend on gender, risk
preferences and demographic features, which we did not control for.
There is also scope for further study on the subject matter by
addressing it from different perspectives and by refining the
experimental model further. A few suggestions towards that goal would be
to see how results differ if the size of the waste envelope is
increased/decreased, the average age of the control group is altered,
the time for the participants for cutting the squares is increased, or
the introduction of a heavier fine on the litterers.
APPENDIX A
INSTRUCTIONS FOR BASELINE TREATMENT
You are now taking part in an economic experiment. Please do not
talk or communicate in any way with other participants during the
experiment. Please also remember to turn off your cellphones. Should you
have any questions, please raise your hand and one of us will come to
assist you. At the end of the experiment one of you will be randomly
selected and paid in cash in private for your participation.
Although there are many people participating in today's
experiment, everyone is working independently. This means that if you
are the one who is randomly selected, your earnings will be based
entirely on your decisions--what the others do has no effect on your
earning. It is therefore important that you take your time to understand
the instructions.
In the experiment you are required to cut as many rectangles drawn
on the coloured sheets provided to you as you can with in the span of 3
minutes. You will use the scissors provided to you to do this. The time
will be measured using the clock in front of you. You will be asked to
stop cutting as soon as the three minutes time is over.
At the end of 3 minutes one of you will be selected at random. For
this purpose you will write your seat number on a piece of paper and put
it in a box. One of the experimenters will randomly pick one number in
front of you from that box. If you are the one who is randomly selected
in the draw, you will be paid Rs. (10) per completed rectangle.
Below are two examples to help you understand the experiment. These
examples are not meant as a guide for behaviour in the experiment.
Example 1
Assume that you cut 15 rectangles within 3 minutes and at the end
of 3 minutes you are the one who is randomly selected in the draw. Your
earnings at the end of the experiment will be 15*10= Rs.150 which will
be paid to you in cash in private at the end of the experiment.
Example 2
Assume that you cut 15 rectangles within 3 minutes and at the end
of 3 minutes you are NOT the one who is randomly selected in the draw.
Your earnings at the end of the experiment will be NIL.
If you have any questions, please raise your hand to attract the
attention of one of the experimenters. Otherwise, please proceed to
answer the questions below. Once you have finished answering the
questions, please raise your hand and one of the experimenters will come
to check your answer.
Control Questions
(1) If you have completed 6 rectangles by the end of three minutes
and your seat number is selected at random, how much will be your
earning?
Rs 0 Rs 60 Rs 100
(2) If you have completed 6 rectangles by the end of three minutes
and your seat number is not selected at random, how much will be your
earning?
Rs 0 Rs 20 Rs 60
INSTRUCTIONS FOR COST TREATMENT
You are now taking part in an economic experiment. Please do not
talk or communicate in any way with other participants during the
experiment. Please also remember to turn off your cellphones. Should you
have any questions, please raise your hand and one of us will come to
assist you. At the end of the experiment one of you will be randomly
selected and paid in cash in private for your participation.
Although there are many people participating in today's
experiment, everyone is working independently. This means that if you
are the one who is randomly selected, your earnings will be based
entirely on your decisions--what the others do has no effect on your
earning. It is therefore important that you take your time to understand
the instructions.
In the experiment you are required to cut as many rectangles drawn
on the coloured sheets provided to you as you can with in the span of 3
minutes. You will use the scissors provided to you to do this. The time
will be measured using the clock in front of you. You will be asked to
stop cutting as soon as the three minutes come to an end.
At the end of 3 minutes one of you will be selected at random. For
this purpose you will write your seat number on a piece of paper and put
it in a box. One of the experimenters will randomly pick one number in
front of you from that box. If you are the one who is randomly selected
in the draw, you will be paid Rs.10 per completed rectangle.
However, Rs 5 per piece of extra paper found around you will be
deducted from your total earning.
Below are a few examples to help you understand the experiment.
These examples are not meant as a guide for behaviour in the experiment.
Example 1
Assume that you cut 15 rectangles within 3 minutes and at the end
of 3 minutes you are the one who is randomly selected in the draw and
you have no pieces of extra paper on your desk or down on the floor.
Your earnings at the end of the experiment will be 15*10= Rs 150 which
will be paid to you in cash in private at the end of the experiment.
Example 2
Assume that you cut 15 rectangles within 3 minutes and at the end
of 3 minutes you are the one who is randomly selected in the draw, but
you have 2 pieces of extra paper on your desk or down on the floor. Your
earnings at the end of the experiment will be 15*10= Rs 150 - (2*5) =
Rs.140 which will be paid to you in cash in private at the end of the
experiment.
Example 3
Assume that you cut 10 rectangles within 3 minutes and at the end
of 3 minutes you are the one who is randomly selected in the draw, but
you have 20 pieces of extra paper on your desk or down on the floor.
Your earnings at the end of the experiment will be 10*10= Rs.100 -
(20*5) = 0 and you will not be paid any money.
NOTE: There is no negative earnings in the experiment. For example
if you have 30 piece of extra papers on your desk or down on the floor
in Example 3 and you are the one who is randomly selected in the draw,
your earnings at the end of the experiment will also be 0.
Example 4
Assume that you cut 20 rectangles within 3 minutes and at the end
of 3 minutes you are NOT the one who is randomly selected in the draw.
Your earnings at the end of the experiment will be NIL.
If you have any questions, please raise your hand to attract the
attention of one of the experimenters. Otherwise, please proceed to
answer the questions below. Once you have finished answering the
questions, please raise your hand and one of the experimenters will come
to check your answer.
Control Questions
(1) If you have completed 6 rectangles by the end of three minutes
and no extra piece of paper is found on the desk or down on the floor in
front of you; moreover, if your seat number is selected at random, how
much will be your earning?
Rs 0 Rs 60 Rs 100
(2) If you have completed 6 rectangles by the end of three minutes
and your seat number is selected at random, but I piece of extra paper
is found on the floor in front of you, how much will be your earning?
Rs 0 Rs 55 Rs 60
(3) If you have completed 6 rectangles by the end of three minutes
and your seat number is not selected at random, how much will be your
earning?
Rs 0 Rs 20 Rs 60
REFERENCES
Al-Khatib, I. A., H. A. Arafat, R. Daoud, and H. Shwahneh (2009)
Enhanced Solid Waste Management by Understanding the Effects of Gender,
Income, Marital Status, and Religious Convictions on Attitudes and
Practices Related to Street Littering in Nablus-Palestinian Territory.
Waste Management 29:1,449-455.
Arafat, H., et al. (2007) Influence of Socio-economic Factors on
Street Litter Generation in the Middle East: Effects of Education Level,
Age, and Type of Residence. Waste Management and Research 25:4, 363-370.
Beck, R. W. (2007) Literature Review-Litter. A Review of Litter
Studies, Attitude Surveys and other Litter-Related Literature: Final
Report. Keep America Beautiful.
Campbell, D. F. (2007) People Who Litter. Wigan, UK. ENCAMS.
Dur, R. and B. Vollaard (2012) The Power of a Bad Example--A Field
Experiment in Household Garbage Disposal (Tinbergen Institute Discussion
Paper No. 12-061/1).
Finnie, W. C. (1973) Field Experiments in Litter Control.
Environment and Behaviour.
Geller, E. S., R. A. Winett, and P. B. Everett (1982) Preserving
the Environment: New Strategies for Behaviour Change. Elmsford, NY:
Pergamon.
Glossary of Environment Statistics (1997) In Studies in Methods.
New York: United Nations.
Heberlein, T. A. (1971) Moral Norms, Threatened Sanctions, and
Littering Behaviour. University of Wisconsin--Madison.
Kagel, J. H., C. Kim, and D. Moser (1996) Fairness in Ultimatum
Games with Asymmetric Information and Asymmetric Payoffs. Games and
Economic Behaviour 13:1, 100-110.
Keenan, R. J., C. E. Prescott, J. P. Kimmins, J. Pastor, and B.
Dewey (1996) Litter Decomposition in Western Red Cedar and Western
Hemlock Forests on Northern Vancouver Island, British Columbia. Canadian
Journal of Botany 74:10, 1626-1634.
Krauss, R. M., et al. (1978) Field and Laboratory Studies of
Littering. Journal of Experimental Social Psychology, 109-122.
Littering Behaviour in America: Results of a National Study (2009)
Washington: Keep America Beautiful.
Ojedokun, O. and S. K. Balogun (2011) Psycho-sociocultural Analysis
of Attitude Towards Littering in a Nigerian Urban City. Ethiopian
Journal of Environmental Studies and Management 4:1.
Raffoul, L., R. Mahon, and R. Goodridge (2006) Roadside Litter in
Barbados: Sources and Solutions.
Reiter, S. M. and W. Samuel (1980) Littering as a Function of Prior
Litter and The Presence or Absence of Prohibitive Signs. Journal of
Applied Social Psychology, 45-55.
Robinson, S. N. (1976) Littering Behaviour in Public Places.
Environment and Behaviour 8:3, 363-384.
Stokols, D. and 1. Altman (Eds.) (1987) Handbook of Environmental
Psychology (Vols.1-2). New York: John Wiley.
Torgler, B., B. S. Frey, and C. Wilson (2009) Environmental and
Pro-social Norms: Evidence on Littering. The BE Journal of Economic
Analysis and Policy 9:1.
Torgler, Benno, Marla A. Garcia-Valinas, and Alison Macintyre
(2008) Justifiability of Littering: An Empirical Investigation.
Williams, E., R. Curnow, and P. Streker (1997) Understanding
Littering Behaviour in Australia. Beverage Industry Environment Council.
Fatima Salim Khawaja <Fatima.s@gmail.com> and Anwar Shah
<anwar@qau.edu.pk> are Assistant
Professors at the Department of Economics, Quaid-i-Azam University,
Islamabad.
(1) It is pertinent to mention that at the time of the study, three
batches were enrolled in the School of Economics (QAU). The first batch
students were in their 5th semester while the second and first batch
students were in their third and first semester, respectively. Our
protocol is between treatments, reason being that we have to collect and
count the number of litter pieces produced after students have left the
experimental room. Hence, we could not have the within treatment
protocol. Additionally, we invited students of different semesters for
the two treatments because inviting students of the same semester for
both treatments could contaminate the data as students could discuss the
experiment with each other during the interval time period. Running both
treatments at the same time could solve the issue of running both
treatments on the same semester students, however, the constraint of
space and number of experimenters did not make that possible.
(2) See Appendix A for the full instructions used in the Baseline
treatment and Cost treatment.
(3) Questions inevitably arise concerning the non-seriousness of
participants regarding payoff's in this game. However, the
procedures employed (paying off in 1 out of 10 subjects) are comparable
to other experimental research in this area. For example, in Kagel, et
al. (1996) subjects participate in 10 bargaining periods against
different opponents in an ultimatum game and they are paid conditional
on their bargaining outcomes for one period which is selected randomly
at the end of the session.
Table 1
Descriptive Statistics Across Treatments
No. of Completed Litter Pieces Inside
Squares Waste Envelope
Baseline Cost Baseline Cost
Treatment Treatment Treatment Treatment
Mean 4.4 4.1 1 2.18
Mode 5 4 0 0
Std. Dev. 1.30 1.88 2.60 3.28
Litter Pieces Outside
Waste Envelope
Baseline Cost
treatment Treatment
Mean 7.25 4.75
Mode 9 0
Std. Dev. 4.38 4.45
Table 2
Total Number of Squares, Litter Pieces Inside and Outside
Waste Envelope Across Treatments
Sum of Litter
Pieces Sum of Litter
Sum of Inside Waste Pieces Outside
Squares Envelope Waste Envelope
Baseline Treatment 176 40 290
Cost Treatment 164 87 190
