Appendix B – Methodology & Statistical Analysis

Survey Design

We identified ten desirable characteristics of a warning system for biological agents and grouped these into three categories [Table B.1]. Survey respondents evaluated Warning Systems on these characteristics using a 1-to-5 Likert scale of agreement. Respondents were solicited via email and the scoring was conducted in a Microsoft Excel spreadsheet attachment.

Experimental Design

For each of the two scenarios, Known Agent and Unknown Agent, an experimental design was conducted. First, the authors drafted a list of nine potential attack indicators placed into three groups [table B.2] and a list of expert evaluators with various backgrounds. Rather than have experts score each Indicator Group one at a time (as in a typical survey), we implemented a full factorial design with the Indicator Groups as factors. That is, experts were randomly assigned to assess one of eight Systems, varying from a baseline of no indicators to a full system of all three Groups: Business, Medical, and Sensor indicators. This design allows us to assess the value of Group combinations, i.e. do Indicator Groups complement or duplicate each other’s information for first responders?

Data Analysis

The authors sent surveys to 132 experts and further instructed them to distribute the instrument to other experts. This yielded 49 respondents, each of which analyzed one Warning System for each Scenario. Composite scores for Utility, Trustworthiness, and Resource Requirements were created by averaging the characteristics for each. Averaging the three composite scores created an Overall score. Table B.3 shows the raw scores and a measure of the components’ consistency/reliability, Cronbach’s Alpha. Note that the reliability for Trustworthiness is marginal (.65), suggesting that measurement of this element needs refinement in future studies.

Analysis of Variance (ANOVA) was performed on the four composite scores for each scenario. This statistical analysis can detect patterns in the eight combinations of Indicator Groups to isolate those that drive better scores. For example, examining the Utility Scores for the Unknown Agent in table B.3, we see that Systems that include Medical information score higher than those without Medical data do. These results are reflected in the ANOVA coefficients [table B.4] and the Systems that the models indicate as better than the baseline [table B.5].

Notes

Fit statistics and coefficients for the Known Agent models are generally larger than those of the Unknown Agent models. This suggests that warning for Unknown Agents is more difficult [table B.4].

Additional warning indicators can be added to the Preferred Systems in table 5, but doing so will not provide responders & policymakers with statistically-significant value-added.

This study was not designed to isolate the individual effects of the components of the three Indicator Groups (Medical, Business and Positive ID). This will require additional study-- for example, a larger sample employing a fractional factorial design with individual indicators as factors.

Table B.1 – Desirable Characteristics of Warning Systems & Category Assignments

Characteristics of Warning System and its Output Category
The sensor signals are available at early stages of agent release or human exposure. Utility
The set of information is timely and provides advanced warning. Utility
The set of information is actionable rather than informational. Utility
The system output and results are reliable and subject to validation. Trustworthiness
The system is highly sensitive to the agent (high recall). Trustworthiness
The system will have high specificity (low percentage of false negatives). Trustworthiness
The benefits of the system outweigh its cost. Resource Requirements
The system requires little or no input to provide an alarm function 24/7/365. Resource Requirements
The technology requirements of the system are easily met using current technology. Resource Requirements
The data requirements of the system are easily met through existing data sets. Resource Requirements

Table B.2 – Biological Attack Indicators & Group Assignments

Biological Attack Indicators Group
Medical Claims Reporting Business
Pharmacy OTC Sales Business
Absenteeism Reporting Business
Medical Surveillance: Doctors’ Offices Medical
Medical Surveillance: ER Medical
Veterinary and Agricultural Surveillance Medical
Nurse Help Line Calls Medical
Laboratory Reporting Positive ID
Sensor Reporting Positive ID
Prodromic Reporting Positive ID

Table B.3 – Means, Sample Sizes & Reliability Scores for Composite Scores

Indicator Groups

N

COMPOSITE SCORES

Utility

(α=.84)

Trust

(α=.65)

Resource Requirements

(α=.76)

Overall

(α=.86)

 

Baseline

(no indicators)

11

1.8

2.2

2.4

2.1

Scenario 1 Known Agent

 

Business

5

1.6

2.1

2.8

2.2

 

Medical

7

2.8

2.5

3.7

3.0

 

Positive ID

5

3.3

3.6

3.2

3.4

 

Business + Medical

5

1.9

1.6

2.9

2.1

 

Business + Pos. ID

6

3.6

3.5

4.0

3.7

 

Medical + Pos. ID

9

3.4

3.4

3.6

3.5

 

Business + Medical + Positive ID

6

4.6

4.2

3.9

4.2

Scenario 2 Unknown Agent

 

Business

6

1.8

1.9

2.7

2.1

 

Medical

6

2.6

2.4

3.0

2.7

 

Positive ID

6

2.2

2.1

2.2

2.1

 

Business + Medical

7

2.5

2.2

3.5

2.7

 

Business + Pos. ID

5

2.2

3.0

3.0

2.7

 

Medical + Pos. ID

5

2.6

2.3

3.1

2.7

 

Business + Medical + Positive ID

5

3.5

3.5

3.0

3.4

Table B.4 – Simplified ANOVA Models: Coefficients and R2

 

Utility

Trustworthiness

Resource Requirements

Overall

Scenario 1 Known Agent

 

R2

Intercept

Sensor

Med

Bus

Bus*Sensor

 .58

 1.9**

 1.0**

 0.6**

-0.5

 1.3**

R2

Intercept

Sensor

.52

2.1**

1.5**

R2

Intercept

Sensor

Med

.25

2.7*

0.7*

0.5*

R2

Intercept

Sensor

Med

Bus

Bus*Sensor

.56

 2.3**

 0.9**

 0.4*

-0.4

 0.9*

Scenario 2 Unknown Agent

 

R2

Intercept

Med

 .14

 2.0**

 0.8**

R2

Intercept

Sensor

Bus

Bus*Sensor

 .26

 2.2**

-0.0

-0.2

 1.3**

R2

Intercept

Med

 .13

2.5**

0.7**

R2

Intercept

Med

 .13

2.2**

0.6**

Statistical Significance: * <.05; **<.01
Non-significant factors were removed from models unless higher-order interactions were significant.

Table B.5 – Systems Preferred over Baseline and Predicted Scores

 

Utility

Trustworthiness

Resource Requirements

Overall

Scenario 1

Known Agent

Bus+Med+Sensor

Bus+Sensor

Med+Sensor

Bus+Med

Sensor

Med

4.4

3.8

3.6

2.1

2.9

2.6

Sensor

3.7

Med+Sensor

Sensor

Med

3.9

3.4

3.1

Bus+Med+Sensor

Bus+Sensor

Med+Sensor

Sensor

Med

4.2

3.7

3.6

3.2

2.7

Scenario 2

Unknown Agent

Med

2.7

Bus+Sensor

3.3

Med

3.2

Med

2.8

Note: If an Indicator Group is not present in the listed System, there is no statistically significant value-added.

Table of Contents Previous