Prof. Anthony Patt, Potsdam Institute and Boston University Pablo Suarez, Boston University

Prof. Anthony Patt, Potsdam Institute and Boston UniversityPablo Suarez, Boston UniversityChiedza Gwata, University of Zimbabwe

Seasonal Forecasts,Scientific Information,& Subsistence Farming

Research funded by the NOAA Office of Global Programs

Presentation at the September 2003 SARCOFLusaka, Zambia

1. Farmers didn‘t seem to understand the forecast.

2. Farmers didn‘t seem to trust the forecast.

3. Farmers didn‘t seem to use the forecast very much.

Observations

1. Could farmers understand a probabilistic forecast?

2. Could the forecasts be made more credible?

3. If farmers understand and believe a forecast, will they use it to make different decisions?

Questions

Could farmers understand a probabilistic forecast?

Experiment:a. Play a simple betting game 5 timesb. Play a different game that mimics realityc. Play the simple betting game again

Could farmers understand a probabilistic forecast?

Experiment:a. Play a simple betting game 5 timesb. Play a different game that mimics realityc. Play the simple betting game again

Could farmers understand a probabilistic forecast?

Psychological experiment:a. Play a simple betting game 5 timesb. Play a different game that mimics realityc. Play the simple betting game again

Could farmers understand a probabilistic forecast?

Experiment:a. Play a simple betting game 5 timesb. Play a different game that mimics realityc. Play the simple betting game again

By the way, women did much better than men.

Could farmers understand a probabilistic forecast?

Teaching method during workshop

Could farmers understand a probabilistic forecast?

Above Normal—25%

Above Normal—25%

Below Normal—45%

Below Normal—25%

El Niño Stays

El Niño Goes Away50%

50%

25% Above Normal 40% About Normal 35% Below Normal

About Normal—50%

About Normal—30%

Teaching method during workshop

Could the forecasts be made more credible?

Experiment: three door game

1 2 3

Could the forecasts be made more credible?

1 2 3

Experiment: three door game

Could the forecasts be made more credible?

1 lose 3

Experiment: three door game

Could the forecasts be made more credible?

1lose

3

?

Experiment: three door game

Could the forecasts be made more credible?

Psychological experiment: three door game

lose

P(win) = 0.33

lose WIN!

P(win) = 0.67

Could the forecasts be made more credible?

What if:a. Somebody gives advice to switch?b. You have observed switching succeeding?c. The advisor has an incentive to help you win (wins same prize)?

lose

P(win) = 0.67

lose WIN!

P(win) = 0.67

Experiment: three door game

Could the forecasts be made more credible?

Experiment: three door game

Could the forecasts be made more credible?

By the way, women did much better than men.

Experiment: three door game

If farmers understand and believe a forecast, will they use it to make different decisions?

Explain and work with information.

If farmers understand and believe a forecast, will they use it to make different decisions?

Listen to experiences on the ground.

If farmers understand and believe a forecast, will they use it to make different decisions?

Build networks with and involve local authorities.

If farmers understand and believe a forecast, will they use it to make different decisions?

Discuss major constraints, including climate.

If farmers understand and believe a forecast, will they use it to make different decisions?

Assist community to organize a sustainable process,+ examine whether worthwhile.

1. Could farmers understand a probabilistic forecast?

2. Could the forecasts be made more credible?

3. If farmers understand and believe a forecast, will they use it to make different decisions?

Questions

1. Could farmers understand a probabilistic forecast?

2. Could the forecasts be made more credible?

3. If farmers understand and believe a forecast, will they use it to make different decisions?

Questions

Yes, yes, and yes!

1. Could farmers understand a probabilistic forecast?

2. Could the forecasts be made more credible?

3. If farmers understand and believe a forecast, will they use it to make different decisions?

Questions

Yes, yes, and yes!(But it takes work)