Disaster Research Center

Jenniffer Santos-Hernández Disaster Research Center

University of Delaware

Developing Informed Radar Technology: The social dimensions of risk communication

This work was supported by the Engineering Research Centers (ERC) Program of the National Science Foundation under NSF Cooperative Agreement No. EEC-0313747. Any opinions, findings and conclusions, or recommendations expressed in this material are

those of the authors and do not necessarily reflect those of the National Science Foundation.

The CASA Project

Inter-disciplinary, Multi-institution research effort

ERC Director: Dr. David McLaughlin, UMASS, Amherst

Director of Industry, Government, and End User Partnerships: Brenda Phillips

Senior Social Science Faculty: Havidán Rodríguez and Walter Díaz

Other faculty associated to the DRC-CASA project: William Donner and Joseph Trainor

DRC-CASA Graduate students: Jenniffer M. Santos-Hernández

DRC-CASA Undergraduate students: Claudia Flores, Paige Mikstas, Yesenia Rodríguez, Spencer Schargorodski, Kathleen Shea, Stephen Shinn, Jasmine Wynn

Social Scientists in CASA

How improved forecasting can reduce the exposure and vulnerability of individuals and property to every-day and extreme weather events?

What factors inform weather related decisions at different levels?

How are warnings communicated to the general population?

Under what conditions are these warnings interpreted correctly?

Through the use of field research, focus groups, in-depth interviews, and surveys, we are examining how the end-user community, particularly emergency managers and the general public, access, interpret, utilize, and respond to weather forecasts

Use of both qualitative and quantitative approaches

Research Efforts Survey of emergency managers’ access and use of weather

information In-depth interviews with emergency managers, weather

forecasters, and other emergency management related personnel to understand the processes by which emergency managers acquire, manage, and use weather information (Oklahoma and Puerto Rico)

Quick-response research after Hurricane Katrina Quick-response research after tornado warnings Phone Survey on response to tornado warnings Social Vulnerability Index for Puerto Rico Online GIS integrated platform – Disaster Decision Support

Tool Evaluation of the implementation of FEMAs CERT program

in Puerto Rico

Objectives – Public Response Phone Survey

Explore and describe public response and the household decision making process following a severe weather warning or a hazard event

Using Computer Assisted Telephone Interviewing (CATI), explore the public’s response to severe weather warning/events in communities in Oklahoma, Kansas, Minnesota, Illinois, Mississippi, Tennessee and Alabama in 2008 and 2009.

Develop quantitative and predictive models, which are based on initial extensive qualitative research with emergency managers and the general public following severe weather events

DRC-CATIDeployment

Grounded Approach: Qualitative – Quantitative Quick Response – Phone Surveys. Two-Step Sampling Process

County Selection Criteria

Demographic and Socio-Economic Characteristics

RaceIncome

EducationAge

Event Characteristics

NWS ProductMagnitudeDamage

Media Coverage

Event Characteristics

NWS ProductMagnitudeDamage

Media Coverage

Inside of Test Bed Scenarios

- Severe Weather- Tornados

- Event with Watch/Warning

- Event NO Watch/Warning- False Alarm

Outside of Test Bed Scenarios

- Tornadoes- Event with Warning- Event NO Warning

Methodology

Questionnaire

127 questions in total yielding about 429 variables:

• Damage to home, business, or other property • Shelter availability and preferences• Social Vulnerability• Social Networks• Insurance coverage • Effectiveness of Siren Systems• Behavioral outcomes of lead time • Social and environmental cues• Protection of possessions and pets• Reception of warnings and watches• Understanding of warnings and watches• Questions on false alarms • Geographic warning specificity• Past experience with other disasters

Methodology

GENESYS Sampling Systems: Genesys provided samples based on DRC sampling requests in the impacted areas

Over 600 interviews completed in counties in Oklahoma, Minnesota, Kansas, Illinois, Mississippi, Tennessee and Alabama.

Average duration of interviews: 35 minutes Calls were made 1-3 weeks after event

Demographic Characteristics

Male Female Refused0.00%

10.00%

20.00%

30.00%

40.00%

50.00%

60.00%

70.00%

80.00%

30.50%

69.00%

0.50%

Gender

Demographic Characteristics

81.00%

10.50%

1.00%

3.00%2.00%

2.50%

Racial Composition

WhiteBlackPacificAmerican IndianOtherRefused

Demographic Characteristics

18-24

25-34

35-44

45-54

55-64

65 and u

p0%

5%

10%

15%

20%

25%

30%

35%

40%

3%7%

10%

21%25%

35%

Age

Demographic Characteristics

Less

than $24,9

99

$25,000 to

$49,999

$50,000 to

$74,999

$75,000 to

$99,999

$100,000 to

$149,999

More th

an $150,000

0%

5%

10%

15%

20%

25%

30%

19%

25%23%

12%14%

7%

Annual Income

Were you aware that a tornado or severe storm had been observed in the surrounding area before it got to your town?

Yes84%

No 12%

Don’t know3%

Did you receive a warning or notification of a tornado or severe storm in your region?

No Yes Don’t Know0.00%

10.00%

20.00%

30.00%

40.00%

50.00%

60.00%

70.00%

80.00%

90.00%

100.00%

9.30%

85.85%

4.85%

NoYesDon’t Know

From whom did you receive this information?

53.8%

42.3%

Parents

Siblings

Friends

Neighbors

Fire Department

Emergency Manager

Hospital

Mass Media

Other: Siren

11% 4%

5%

40%

25%

6%

2%

2% 5%

After receiving the warning or notification, what did you do?

Sought more informationLooked outside a windowContinued what they were doingTook shelterProtective ActionsProtected private goodsProtected petCalled othersOther

When you first found out a tornado or severe storm was present inside or near your town or city, about

how many minutes did it take before it hit your neighborhood? (Average = 27.9 minutes)

0_5 6_10 11_15 16_20 21_25 26_30 36+0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

9.6%

22.9%

10.8%

14.5%

1.2%

20.5% 20.5%

Did the tornado sirens in your community go off?

Yes No There are no sirens

Don’t know0.00%

10.00%

20.00%

30.00%

40.00%

50.00%

60.00%

70.00%

80.00%72.80%

13.60%9.40%

4.20%

Did you look outside to verify whether the tornado or severe storm was coming?

Yes No0.00%

10.00%

20.00%

30.00%

40.00%

50.00%

60.00%

70.00%

80.00%

66.90%

33.10%

Did you receive information from the Internet during the last 30 minutes before the tornado or severe storm arrived?

Yes No Don’t know0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

14%

85%

1%

Series1

Why did you not receive information from the internet?

8%8.9%

2.7%

57.1%

9.8%

3.6%8.1% 1.8%

Power OutageDo not have access to the Internet Already seeking shelterComputer offNo access to computerEnough Information

Did you receive information from television during the last 30 minutes before the tornado or severe storm arrived?

No Yes Don’t Know0.00%

10.00%

20.00%

30.00%

40.00%

50.00%

60.00%

70.00%

80.00%

90.00%

16.10%

84.00%

0.90%

Did you take any actions to protect yourself, your family, or your property from the hazard event?

Yes No Don’t Know0.00%

10.00%

20.00%

30.00%

40.00%

50.00%

60.00%

70.00%

58.60%

41.15%

0.25%

What information led you to seek shelter?

Personally saw tornado approaching

Saw tornado or storms approaching on TV

Somebody called me

NWS specific information about what action to take

Local TV gave specific information about what action to take

Radio gave specific information about what action to take

Sirens

Tornado warning

Other

Don’t Know

Refused

0.0% 10.0% 20.0% 30.0% 40.0% 50.0%

6.4%

12.4%

2.0%

3.2%

40.8%

11.6%

34.0%

6.0%

19.2%

0.8%

0.4%

NOAA Radio Ownership

66%

31%

2%1%

Do you own a NOAA weather radio?

No

Yes

Don’t know

Refused

70%

8%

3%4%

13% 2%

1 (Never)

2

3

4

5 (V. Frequently)

Don’t know

How often would you say you listen to a NOAA radio for information about

tornadoes or severe storms?

Tornado Watch & Warningand False Alarms

Respondents appear to have difficulty in understanding the differences between watches and warnings and what is a false alarm

Participants seem to understand that watches and warnings represent some type of danger, but they are unable to clearly differentiate between these two concepts

Watch Definition: Examples “I think the watch is the more dangerous one” “Same as a warning” “When the TV flashes yellow” “They put it up on the TV and tell you what

time it will be in your area and when to take shelter”

“They feel like there’s one [tornado] in our vicinity”

“A tornado is on the ground near your house” “Tornado was been sighted in my area” “Watch for the tornado coming to you”

Watch Definition

Agrees with NWS definition

47%

Disagrees: confuses watch with warning

13%

Dis-agrees:

Scientific terms13%

Disagrees with NWS definition

27%

Could you please describe what you think a tornado watch is?

Warning Definition

Agrees with NWS definitions

43%

Disagrees: confuses watch with warning

9%

Disagrees: Scientific terms

8%

Disagrees with NWS definitions

40%

Could you please describe what you think a tornado warning is?

False Alarm Definition

Agrees with NWS definition

13%

Common Use19%

Disagrees: Scientific

terms3%

Disagrees with NWS definition

65%

Could you please describe what a false alarm is?

In your opinion, how trustworthy are the weather forecasts provided in your region?

1 2 3 4 5 Don’t Know Refused0

50

100

150

200

250

Not Trustworthy Very Trustworthy

Next StepsContinue CATI Survey; expand sample

size and geographic areas

Develop predictive models on protective action:Binary logistic model to predict protective action

following severe weather warning or a hazard event

Estimate the probability that the dependent variable will assume a certain value (e.g., take protective action or not) based on a number of independent variables

Canon (1994) asserts that technology is not socially neutral and that we must have an understanding of the context in which it is implemented.

Technology matters, but what really matters is the application of the substantive knowledge that we generate regarding how individuals respond (or not) to severe weather events and how can we improve their response in order to minimize the devastating impacts associated with these events.

Technology and the social dimensions of risk communication