Sin t

1
Communication and Cooperation Combining
Qualitative Approaches and ExperimentalEconomics
to Study Common Pool Resources. Maria Claudia
Lopez Juan Camilo Cardenas
2
Outline of the Presentation

• Research questions
• The experiment
• Communication
• Methodology
• Qualitative information
• Quantitative information
• Conclusion and further questions

3
Research Questions

• From the literature on experimental economics,
  where communication is allowed it is clear that
  communication works to improve cooperation. But
  why?
• By analyzing the discussions, is it possible to
  forecast how the group will perform during the
  experiment?
• When conducting experiments in the field, are we
  adding different/new explanations of why
  communication works?

4
The Experiment

• Common Pool Resource Experiments.
• 8 players per group
• Effort extraction from 0 to 8
• 10 rounds baseline- 10 rounds with communication
• 5 minutes of communication in between rounds
  (recorded and taped)
• 3 different regions of Colombia
• 17 groups

5
Payoff table
6
(No Transcript)
7
Communication Words alone, without a sword
(Ostrom and Walker 1989) What are the
predictions?

• Cheap talk
• Non binding,
• Non enforceable,
• Does not change the structure of the game.

8
Plausible reasons for communication to enhance
cooperation
9
Methodology

• Combining qualitative approaches with
  statistical analysis to investigate the rich
  information generated through the experiments.

10
The Methodological Challenge

• We know that communication works,
• We have some hypotheses to explain why it works,
• But, we do not have a unique tool to examine the
  communication process.

11

• I borrowed from focus groups the way they
  analyze the qualitative information.
• Non parametric statistical analysis were used

12
Methodology (next)

• We transcribed and analyzed each of the group
  communication exchanges.
• We coded their communications in three different
  categories
• High cooperation
• Medium cooperation
• Low cooperation
• We compared our prediction with the actual levels
  of cooperation during the experimental sections.

13
Qualitative informationWhy Focus groups?

• In a focus group the researcher is expected to
  create an atmosphere such than people could
  express their opinions (Krueger,1994)
• 6 different requirements
• People Groups of 4 to 12. Small enough that
  everybody has a chance to speak, but big enough
  to allow diversity.
• Similar task within different groups.
• People with some common caracteristcs
• Information is being gathered for later use.
• All the information recorded and taped can be
  used.
• The discussion is moderated by an external agent.

14
When analyzing and coding the information I was
looking for

• What were they talking about?
• What type of decision were they making?
• Do they arrive to an agreement?
• What kinds of arguments people were they using?
• Are they maintaining the agreement?
• How are they maintaining the agreement?
• Were they changing the agreement over time?
• What happened when they noticed that people were
  following/ not following the agreement?

15
(No Transcript)
16

• Very low numbers in the months when we go to the
  forrest, the less one goes to the forrest, the
  less it is destroyed. (Números bajitos en los
  meses en que vamos al bosque, Entre menos se va
  al bosque menos se destruye.)
• Listen my friends, pay attention to me for a
  second. Many times we are writing a number and,
  unfortunately, we are making them too high and
  the scores are coming out too low. (Oiga
  compañeritos me atienden un segundito. Muchas
  veces estamos colocando un número y
  desafortunadamente los estamos colocando muy
  altos y los puntajes se nos vienen muy bajos.)

17
Frequency of game decisions in the final three
rounds before communication in each selected
category.
18
Frequency of game decisions in the first round
after communication in each selected category.
19
(No Transcript)
20
When does communication work in the field?

• All the arguments from the literature were found
  in our analysis.
• But we also found, that a leader who understands
  the game and is able to explain the game to the
  other participants is important.
• And, context matters. Being able to relate what
  happened in the experiment to what happens in
  their forest enables them to understand the game
  much better.

21
Conclusions

• The predictions made through qualitative analysis
  were accurate. Differences per category were
  statistically different.
• Communication works, but not always. There were
  big differences among groups.
• The hypotheses found in the literature to explain
  why communication works were also found in the
  field.
• Leadership and context are two new factors that
  help understand why communication works in the
  field.

22
More research to be done in this direction

• Communication works but why is it so fragile?
• What happens with communication within groups?
  What consequences does intra-group communication
  bring to inter-group communication?