Attachment 3-A0001 Shoe Study Reference Brief.pdf

Executive Overview

December 2018

Behavioral Science Research Support for the Next Generation Shoe Screening

Technology: Overview of Three Studies

Background

As part of the Apex Screening at Speed (SaS) Program, the Department of Homeland

Security Science and Technology Directorate (DHS S&T) is researching and developing next-generation technologies to enhance the passenger screening process (DHS S&T, 2016). DHS

S&T is currently working with Pacific Northwest National Laboratory (PNNL) to develop a millimeter wave shoe scanning system that would allow passengers to wear their shoes through the security checkpoint. In 2017, PNNL developed and successfully tested a laboratory prototype of a shoe scanner system (Jones et al., 2017). In the second phase of development, which began in FY18, PNNL is developing a refined engineering test unit for the

SaS program at DHS S&T.

Early in FY18, the Behavioral, Economic, and Social Science Engine (BESS-E) partnered with the Apex SaS team to provide social science support for this effort. Broadly speaking, social science support from the BESS-E team provides analysis of the social, economic, and behavioral implications of an Apex program’s research, implementation, and diffusion of new technologies, programs, and policies. By matching social science capabilities to Apex program needs, BESS-E is able to provide basic and applied research insights that address the human impact of the Apex programs (DHS S&T, 2018).

The purpose of RTI International’s work as part of the BESS-E team for the current project was to provide behavioral science research support by gathering data on the types of footwear worn by air travelers, to identify stakeholder and passenger perceptions of checkpoint security (particularly as it relates to footwear screening), and to gather feedback on the use

This material is based upon work supported by the Department of Homeland Security Science and Technology Directorate. Any opinions, findings, conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the Department of Homeland Security.

Cleared for Public Release

Executive Overview 2 and integration of a shoe scanning technology in the passenger screening process. The

BESS-E/RTI team’s overall approach involved three separate but related studies:

• A survey of the air traveling public’s footwear preferences, views on safety and security (particularly as they relate to shoe screening), and other relevant perceptions about air travel (see Hedge & King, 2018a).

• An observation study conducted at airports in each of the seven Transportation

Security Administration (TSA)-designated regions to gather data about footwear characteristics of travelers (see King & Hedge, 2018).

• An interview study with TSA officers at a sample of airports concerning implementation issues about shoe scanning, use of shoe scanners, and related operational issues (see Hedge & King, 2018b).

This Executive Overview provides a high-level summary of the development, administration, and analysis activities across these three studies; and summary findings pertinent to development and implementation of a shoe scanning technology.

Study One: A Survey of the Air Traveling Public

Approach

The Survey of the Air Traveling Public was developed and administered to a representative sample of air travelers. As such, it provides a snapshot of perceptions, perspectives, and preferences of respondents over the last 12 months, with particular emphasis on:

• travel habits/patterns (e.g., frequency, destination, purpose),

• membership in travel/security programs (e.g., Precheck, Global Entry),

• footwear characteristics (e.g., size, style),

• perspectives on air travel (e.g., preferences, concerns),

• perspectives on air travel safety and security (e.g., safety versus expediency), and

• demographic characteristics (e.g., height, weight, age, geographic location of residence).

Results

Data were collected from 1,118 air travelers, who traveled for business or leisure, and mostly on domestic flights. The majority of these travelers (69.9%) stated that they had been satisfied or very satisfied with their travel experiences over the last year. Over 44% of

Executive Overview 3 respondents noted that they belonged to no frequent flyer program, and over 75% said they belonged to no expedited screening program. Over 90% of the sample indicated that their shoe size was between 6 and 12. They also noted that their choice of footwear was largely driven by comfort and convenience (63.7%) and ease of removal when passing through security (19.3%). In warm weather, sneakers (42.2%) and sandals/flip-flops (27.8%) were the footwear of choice; in cold weather, sneakers (52.1%) and boots (19.2%) were preferred.

Survey respondents appeared to adopt a balanced approach to security and expediency, believing screening should occur more efficiently than is currently the case, yet expressing a willingness to wait as long as it takes to ensure that individuals with harmful intentions will be detained. Relatively few (21.7%) saw current screening procedures at U.S.

airports as bordering on infringement of their rights as U.S. citizens. In fact, the majority of respondents (53.3%) believed that the removal and scanning of shoes was important for flight safety, and over 75% suggested that eliminating the need for shoe removal at the checkpoint would save as least some time moving through security. Finally, over 64% of those surveyed said they would be willing to limit choices of footwear if it meant saving time going through security; a large but somewhat reduced percentage of respondents (45.9%) professed a willingness to wear TSA-approved footwear to improve speed of movement through the checkpoint.

Study Two: An Observation Study of Footwear Worn by the Air Traveling

Public

Approach

The intent of the Observation Study was to gather data about the types of footwear worn by the air traveling public. To capture a realistic sample of the shoes being worn by air travelers, RTI and DHS S&T researchers traveled to airports in all seven regions of the United

States. A data observation checklist was developed to gather observation data in an efficient, effective, consistent, and reliable manner across sites and data collectors, and allowed data collectors to record observations about shoe style, sole thickness, sole material, heel height and thickness, and shoe size. At each airport, observation data were collected on different days of the week (Friday, Saturday, Sunday, or Monday), and times of the day (morning, afternoon, and evening). Seasonal variations in footwear were also incorporated into the study by visiting Southern airports in September/October and Northern airports in

October/November. Observations were made on 5,089 passengers’ footwear between

September and November 2018.

Executive Overview 4

Results

Most travelers wore size 6 to 12 footwear; most males wore size 10–12 footwear, while most females wore sizes 6–9. Rubber-soled and rubber-heeled sneakers were the most commonly observed footwear. Other observed sole and heel materials included leather, cork, foam, wood, plastic, rope, metal, air pockets, LED lights, and polyurethane springs. The majority of footwear upper materials (the part covering the foot) consisted of leather or suede.

Canvas, cloth, plastic, rubber, mesh, metal, fur/faux fur, rope, snakeskin, LED lights, and costume jewelry were also seen.

Most soles, defined as the vertical layer(s) of material under the part of the foot extending from the tip of the toes to the point where the arch of the foot meets the heel of the foot, were thin—less than ½” thick—although over a quarter had a medium, ½” to 1½” thick sole. Most heels, defined as the vertical layer(s) of material under the heel of the foot where the footwear increases in thickness from the sole, were ½” to 1½” thick; a quarter were thin

(less than ½” thick) and slightly less than a quarter were thick (over 1½”). Heels, boots, and sandals/flip-flops were the styles most likely to have thick heels. Rubber was the most common heel material, regardless of heel thickness. Leather, cork, and wood materials were more likely to be seen in thick heels. As sole thickness increases so too does heel thickness.

Study Three: Perspectives on Shoe Scanner Technology Implementation—

Interviews with Transportation Security Officers

Approach

The intent of the Interview Study was to gather insights from TSA officers about integrating footwear scanning technology into TSA’s passenger screening standard operating procedures. To capture perspectives across airports in perceptions and feedback on the use and integration of a footwear scanner, RTI and DHS S&T researchers conducted interviews at four airports with a total of 23 officers between September and November 2018. The interviews asked officers to reflect on challenges encountered when new technologies are introduced into the security screening process and to recommend success strategies for implementing shoe scanning technology.

Results

Thematic analysis of interview data revealed four broad technology adoption considerations: the traveling public, the TSA officers, the technology, and the operational environment. Interviewees noted that getting buy-in from the traveling public and officers using the technology is important. Interviewees stated that sufficient officer training, staffing, and

Executive Overview 5 operational testing is a must before the technology is put into operational use. Suggestions for shoe scanning technology placement included: after the travel document check station, integrated into the Advanced Imaging Technology (AIT) system, or after the AIT while passengers wait for bags/personal items to come through x-ray. Interviewees pointed out that common materials such as mud, ice, and lawn chemicals could create shoe scanning problems which need to be considered during design.

Overall most interviewees believed that a shoe scanning technology could make the security process run smoother and more efficiently and reduce passenger security anxieties.

However, if the technology merely slows the checkpoint down by adding another step in the screening process, few officers believe it is worth the time, effort, and expense.

Implications for a Shoe Scanning Technology

Based on the findings from a survey of the traveling public, an observation of air travelers’ footwear choices during travel, and interviews with TSA officers, the following are observations relevant to use of a shoe scanning technology in the airport screening process.

Design

• Almost all travelers wore footwear sized 6 to 12.

• Sneakers were the most popular shoe style; sandals/flipflops were also common in warm weather, as were boots in cold weather.

• Most footwear had leather/suede upper material; canvas/cloth, plastic/rubber, and mesh were also popular.

• Most footwear worn by the air traveling public had a rubber sole less than a ½” thick with a rubber heel ½” to 1½” thick; heels, boots, and sandals/flip-flops were most likely to have a heel over 1½” thick.

• Rubber was the most common heel material, regardless of heel thickness; leather, cork, and wood materials were more likely to be seen in thick heels.

• Most footwear closed via laces or slipped on the foot.

• Prosthetics, and shoe inserts, paraphernalia, and anomalies can cause shoes to alarm.

• Materials such as mud, ice, lawn chemicals, and similar shoe debris can create scanning problems.

• A majority of travelers said they would be willing to limit choices of footwear if it meant saving time going through security.

Executive Overview 6

Adoption/Implementation

• Communication is critical to gain buy-in from both the traveling public and TSA officers, and to educate them on the value of the technology and changes to required security procedures; needs to occur prior to planned implementation.

• Travelers’ trust in the technology from both a personal safety and security perspective is critical, as is officer trust in the accuracy and reliability of the technology.

• TSA staff need to receive thorough training on the technology before implementation; and additional staffing requirements should be considered.

• Operational testing of the new technology prior to adoption and implementation is essential; in the field, the technology can perform differently than in a laboratory; the technology needs be easy to use, durable, and user-friendly.

• Implementation plans need to be strategic based on each airport because of differences in operational environments.

• Technology placement is important to ensure that the shoe scanner does not slow down passenger screening; placement suggestions included after the Travel Document Checker (TDC), integrated into the AIT, or after the AIT.

• If the new technology does not improve safety and security, and make the process more efficient and effective then it is not worth the time, effort, and expense.

Executive Overview 7

References

DHS S&T (2016). Apex Screening at Speed program. Retrieved from:

https://www.dhs.gov/sites/default/files/publications/Apex%20Screening%20at%20Speed %20508%20Approved%20Final_0.pdf

DHS S&T (2018). Behavioral, Economic, and Social Science engine. Retrieved from:

https://www.dhs.gov/sites/default/files/publications/3443_BESS- E_Fact_Sheet_20180330.pdf

Hedge, J. W., & King, S. M. (2018a). A survey of the air traveling public. First Responders Group, Science and Technology Directorate, U.S. Department of Homeland Security.

Hedge, J. W., & King, S. M. (2018b). Perspectives on shoe scanner technology implementation: Interviews with Transportation Security Officers. First Responders Group, Science and Technology Directorate, U.S. Department of Homeland Security.

Jones, A., Hartman, T., McMakin, D., Sheen, D., Hall, T., Teseschi, J., … Munoz, E. (2017).

Millimeter-wave shoe scanner system – Phase 1 final report (PNNL-26733). Richard, WA: PNNL.

King, S. M., & Hedge, J. W. (2018). An observation study of footwear worn by the air traveling public. First Responders Group, Science and Technology Directorate, U.S. Department of Homeland Security.

https://www.dhs.gov/sites/default/files/publications/Apex%20Screening%20at%20Speed%20508%20Approved%20Final_0.pdf https://www.dhs.gov/sites/default/files/publications/Apex%20Screening%20at%20Speed%20508%20Approved%20Final_0.pdf https://www.dhs.gov/sites/default/files/publications/3443_BESS-E_Fact_Sheet_20180330.pdf

Executive Summary

November 2018

An Observation Study of Footwear Worn by the Air Traveling Public

The purpose of the current project was to provide a snapshot of air traveler footwear choices and characteristics. The Observation Study of Footwear Worn by the Air Traveling

Public (Observation Study) was conducted at airports in each of the seven Transportation

Security Administration (TSA) regions in a manner that simulated seasonal differences. At each of the airports, observation data were collected on different days of the week (Friday, Saturday, Sunday, or Monday), and times of the day (morning, afternoon, and evening).

Observations were made on 5,089 passengers’ footwear between September and November

2018.

Most travelers wore size 6 to 12 footwear; the majority of males wore size 10–12 footwear, while the majority of females wore sizes 6–9. Rubber-soled and rubber-heeled sneakers were the most commonly observed footwear. Other observed sole and heel materials included leather, cork, foam, wood, plastic, rope, metal, air pockets, LED lights, and polyurethane springs. The majority of footwear upper materials (the part covering the foot) consisted of leather or suede. Canvas, cloth, plastic, rubber, mesh, metal, fur/faux fur, rope, snakeskin, LED lights, and costume jewelry were also seen.

The vast majority of soles were thin—less than ½” thick—although over a quarter had a medium, ½” to 1½” thick sole. Most heels were ½” to 1½” thick vertically; a quarter were thin

(less than ½” thick) and slightly less than a quarter were thick (over 1½”). Heels, boots, and sandals/flip-flops were the styles most likely to have thick heels.

This material is based upon work supported by the Department of Homeland Security Science and Technology Directorate. Any opinions, findings, conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the Department of Homeland Security.

Research Brief

Background

As part of the Apex Screening at Speed (SaS) Program, the Department of Homeland

Security Science and Technology Directorate (DHS S&T) is researching and developing next-generation technologies to enhance the passenger screening process (DHS S&T, 2016). DHS

S&T is currently working with Pacific Northwest National Laboratory (PNNL) to develop a millimeter wave shoe scanning system that would allow passengers to wear their shoes through the security checkpoint. In 2017, PNNL developed and successfully tested a laboratory prototype of a shoe scanner system (Jones et al., 2017). In the second phase of development, which began in FY18, PNNL is developing a refined engineering test unit for the

SaS program at DHS S&T.

Early in FY18, the Behavioral, Economic, and Social Science Engine (BESS-E) partnered with the Apex SaS team to provide social science support for this effort. Broadly speaking, social science support from the BESS-E team provides analysis of the social, economic, and behavioral implications of an Apex program’s research, implementation, and diffusion of new technologies, programs, and policies. By matching social science capabilities to Apex program needs, BESS-E is able to provide basic and applied research insights that address the human impact of the Apex programs (DHS S&T, 2018).

The purpose of RTI International’s work as part of the BESS-E team for the current project is to provide behavioral science research support by gathering data on the types of footwear worn by air travelers, to identify stakeholder and passenger perceptions, and to gather feedback on the use and integration of a millimeter-wave footwear scanner in the passenger screening process. The BESS-E/RTI team’s overall approach involves three separate but related components:

A survey of the air traveling public’s footwear preferences, views on safety and security (particularly as they relate to shoe screening), and other relevant perceptions about air travel (see Hedge & King, 2018)

Interviews with Transportation Security Officers at a sample of airports concerning implementation issues about shoe scanning, use of shoe scanners, and related operational issues

An Observation Study conducted at airports in each of the seven TSA-designated regions to gather data about footwear characteristics of travelers

Research Brief 2

The current document describes the third component of the project––the Observation

Study––and provides a description of the activities involved in the development, administration, and analysis of data obtained from unobtrusive observation of the air traveling public at airports across the United States.

Objectives of the Study

The intent of the Observation Study was to gather data about the types of footwear worn by the air traveling public. To capture a realistic sample of the shoes being worn by air travelers, it was important for RTI and DHS S&T researchers to travel to airports in all seven regions of the United States. To gather observation data in an efficient, effective, consistent, and reliable manner, we developed a data observation checklist that allowed data collectors to record observations about shoe style, sole thickness, sole material, heel height and thickness, and shoe size. These data could then be used to inform decisions about further development and testing of the prototype millimeter wave shoe scanner being developed by PNNL.

The remainder of this Research Brief describes the methodology applied, including development of the data collection checklist, the data collection procedures, and data analysis plan. Also included are the results of those analyses, grouped into three primary sections (demographics, footwear choices and characteristics of the air traveling public), and the characteristics of footwear soles and heels as they relate to factors important to the development of the shoe scanning technology; and finally, a summary of findings.

Methodology

Development of the Observation Checklist Instrument

Based on input from the DHS S&T BESS-E and SaS teams, and PNNL, and a review of available sources, RTI identified footwear and air traveler characteristics around which to develop observation items. Observed items included the following:

Traveler demographics o Age o Gender o Purpose of travel

Footwear upper (the part that covers the foot) characteristics o Style o Material o Closure style

Footwear sole and heel characteristics o Sole thickness o Sole material

Research Brief 3 o Heel thickness o Heel material

The project team drafted a coding scheme to target each of these observation areas.

These draft items then went through a series of reviews and revisions, with emphasis on improving response options, redundancy, and instrument length.

Once we developed a final version of items, we programmed the instrument as a web-based data collection checklist. The web-based instrument was pretested by members of the

RTI team at a large, public gathering place in central North Carolina. A review of pretest data and data collection procedures led to collapsing response options (e.g., combining slip-on, loafer, and boat shoe styles into one category), clarifying response options, and adding data collection standardization aids (e.g., standalone data collection visual aid (see Appendix A)).

Prior to data collection at airports, six data collectors participated in observer calibration training. This training familiarized data collectors with the observation checklist and the differences in footwear characteristics. The training also served to standardize data collection by orienting data collectors to the same observation strategies and by training data collectors to observe and record footwear characteristics consistently and similarly.

Procedures

Observation data were collected at seven category X or category I1 airports within the continental United States between September and November 2018. The seven airports were each located within a unique TSA region resulting in all seven TSA regions being represented.

Logistics for each data collection were arranged with TSA headquarters staff and TSA staff local to each airport. Data collection at each airport took place over either a Friday and

Saturday or a Sunday and Monday. This was done to increase the likelihood of capturing a mixture of leisure travelers (Saturday and Sunday) and business travelers (Friday and

Monday). Data collections were also scheduled in a manner that simulated seasonal differences. We collected data at airports in warmer, southern climates in September and at airports in cooler, northern climates in October and November. We collected data at domestic and international terminals at six of the seven airports. At the remaining airport, we collected data at only one terminal; however, the airline serving this terminal flies domestic and international routes.

We collected data on computer tablets via the web-based checklist across three 1-hour time periods each day at each airport for a total of six 1-hour data collections per airport.

1 TSA classifies U.S. airports into one of five categories (X, I, II, III, IV) based on various factors such as the number of annual takeoffs and the number of passengers screened through security. In general, Category X airports have the highest passenger volume and Category IV airports have the lowest volume.

Research Brief 4

Observations occurred at TSA security checkpoints as departing passengers formed queues to go through security or at points where arriving passengers exited the TSA-secured gate area.

We collected data on TSA Pre-Check and standard security passengers. Table 1 displays the total number of observations recorded in each TSA region and the category of the data collection airport within each region.

Table 1

TSA Region Category Number of

Observations

1 I 442

2 I 492

3 X 973

4 X 812

5 X 1,256

6 X 439

7 X 675

Total Observations 5,089

Data Analysis Plan

Because the primary purpose for gathering these survey data was to support the ongoing efforts by DHS S&T and PNNL to develop a shoe scanning technology, our data analysis focused on descriptive statistics to provide readily usable information for this work.

Specifically, our descriptive analyses focused on demographics, footwear choices and characteristics, and sole and heel characteristics. We also conducted cross-tabulations to examine the relationships between variables. Cross-tabulations allow examination of relationships within the data that might not be readily apparent when analyzing total survey responses. These cross-tabulations help answer questions such as: What upper materials are associated with various footwear styles? What sole materials comprise the thickest soles?

Which shoe styles have the thickest heels?

Results

The results are grouped and discussed in three primary subsections—demographics, footwear choices and characteristics, and sole and heel characteristics.

Demographics

The air traveling public observed for this study were approximately half female (52.7%) and half male (47.3%). Nearly all (97.4%) were between the ages of 13 and 74, with 1.5% appearing to be below the age of 13 and 1.2% appearing to be over the age of 74. Most passengers appeared to be traveling for leisure (75.6%) rather than for business (23.3%).

Footwear in men’s size 6 to size 12 was most frequently observed (86.4%). For standardization and consistency, footwear sizes were recorded in men’s footwear sizes

Research Brief 5 regardless of passenger gender. Figure 1 shows the distribution of footwear sizes observed by gender.

Figure 1. Footwear Size Frequency by Gender

Footwear Choices and Characteristics

Sneakers were the most common style of footwear worn by air travelers. Boots, slip-on/loafers/boat shoes, sandals/flip-flops, and men’s dress shoes were the next most commonly worn styles, with each being worn at roughly the same frequency. Heels were the least often worn style. Figure 2 shows the frequency of each style. Regardless of specific footwear style, the vast majority of footwear worn were of a casual design (82.8%); the remaining 17.2% were dress shoes.

Lace-up and slip-on closure styles were the most common, accounting for 86.3% of all footwear (44.6% and 41.7%, respectively). Buckle (6.6%) and zipper (5.5%) closures were the next most common followed by Velcro (1.1%). The remaining 0.2% of closure styles fell into one of the following categories: button, plastic clip, snap, or strap.

Most footwear worn by the air traveling public had upper material consisting of leather/suede (43.5%). Canvas/cloth (24.2%) and plastic/rubber (18.8%) were also common materials, as was mesh (11.8%). Metal made up a half-percent of observed materials. Other materials observed were fur/faux fur, rope, snakeskin, LED lights, and costume jewelry. As shown in Figure 3, leather/suede was most commonly found on boots and men’s dress shoes, while canvas/cloth, mesh, and plastic/rubber were most common on sneakers.

0.8%

17.9%

77.0%

4.3% 0.0%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%

80.0%

3 or smaller

3 to 5 6 to 9 10 to 12 13 to 15

Female Passenger Footwear Sizes

0.9% 0.7%

19.2%

72.8%

6.4%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%

80.0%

3 or smaller

3 to 5 6 to 9 10 to 12 13 to 15

Male Passenger Footwear Sizes

Research Brief 6

Figure 2. Footwear Style Frequency

Sole and Heel Characteristics

For this study, we defined the sole as the vertical layer(s) of material under the part of the foot extending from the tip of the toes to the point where the arch of the foot meets the heel of the foot. Observations were recorded for the thickest point of the sole.

Of the three sole thickness ranges—less than ½” (thin), ½” to 1½” (medium), and over 1½” (thick)—most footwear worn by the air traveling public has a thin, less than ½” thick sole (72.2%).

Another 26.8% have a medium, ½” to 1½” thick sole, and 1% have a sole over 1½” thick. Figure 4 displays the frequency of sole thickness ranges within each footwear style.

Rubber comprised 90% of sole materials observed. Leather, cork, and foam were observed on 5.6%, 2.3%, and 1.6% of footwear, respectively. Other materials observed were plastic, rope, wood, air pockets, LED lights, and polyurethane springs. Rubber was the most common material within each thickness range. Figure 5 displays the frequency of sole materials within each sole thickness range. Materials in the “other” category are those listed above—plastic, rope, etc.

Men's Dress Shoe 13%

Heel 3%

Boot 19%

Sandal / Flip-Flop 13%

Slip-on / Loafer / Boat Shoe 17%

Sneaker 35%

Research Brief 7

Figure 3. Upper Material Frequency Within Footwear Styles

Figure 4. Sole Thickness Frequency Within Footwear Styles

0.00% 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% 70.00% 80.00% 90.00% 100.00%

Men's Dress Shoe

Heel

Boot

Sandal / Flip-Flop

Slip-on / Loafer / Boat Shoe

Sneaker

Metal Other Plastic / Rubber Mesh Canvas / Cloth Leather / Suede

.6

.3

.8

.0

.4

.8

.6

.7

% 1

.6

.8

.7

.4

.8

.0

.6

.2

.8

.8

H E E L M E N ' S D R E S S S H O E

S L I P - O N / L O A F E R / B O A T

S H O E

S A N D A L / F L I P - F L O P

B O O T S N E A K E R

Thin - Less than 1/2" Medium - 1/2" to 1 1/2" Thick - Over 1 1/2"

Research Brief 8

Figure 5. Sole Material Frequency Within Sole Thickness Ranges

We defined the heel as the vertical layer(s) of material under the heel of the foot where the footwear increases in vertical thickness from the sole. Footwear without an increase in vertical thickness under the heel of the foot was coded as not having a heel.

Seventy-seven percent of footwear observed had a distinct heel structure. Across all categories of footwear with a heel, over half (58%) had a medium, ½” to 1½” thick heel. About a quarter had thin, less than ½” thick heel; the remainder (17.9%) had over 1½” thick heels. There was more variance in heel vertical thickness within and across footwear styles than there was in sole vertical thickness (Figure 6).

As with footwear sole material, rubber was also the most prevalent heel material (83.7%). Leather, wood, cork, and foam were also common heel materials (6.2%, 4.2%, 2.3%, and 2.1%, respectively). Plastic, metal, cloth, rope, air pockets, LED lights, and polyurethane springs were also observed. Rubber again was the most common material within each heel thickness range (Figure 7).

As would be expected, footwear with a thin sole could have a heel of varying thickness, whereas footwear with increasingly thick soles tended to have a thicker heel. Figure 8 displays the frequency of heel thicknesses within each sole thickness range.

.1

.6

.7

.4

.3

.9

.3

.9

% 1

.2

.1

.5

.0

.1

.9

% 1

.3

T H I N - L E S S T H A N 1 / 2 " M E D I U M - 1 / 2 " T O 1 1 / 2 " T H I C K - O V E R 1 1 / 2 "

Rubber Leather Cork Foam Other

Research Brief 9

Figure 6. Heel Thickness Frequency Within Footwear Styles

Figure 7. Heel Material Frequency within Heel Thickness Ranges

.0

.9

.5

.6

.0

.7

.9

.0

.0

.7

.7

.5

.1

.1

.5

.6

.2

.8

H E E L M E N ' S D R E S S S H O E

B O O T S A N D A L / F L I P - F L O P

S L I P - O N / L O A F E R / B O A T

S H O E

S N E A K E R

Thin - Less than 1/2" Medium - 1/2" to 1 1/2" Thick - Over 1 1/2"

.5

.1

.6

.2

.7

% 1

.2

.7

.5

% 8 .4

.8

.2

% 1

.1

.2

.4

.4

.3

.3

.6

.8

.4

.0

T H I N - L E S S T H A N 1 / 2 " M E D I U M - 1 / 2 " T O 1 1 / 2 " T H I C K - O V E R 1 1 / 2 "

Rubber Leather Cork Wood Foam Plastic Other

Research Brief 10

Figure 8. Heel Thickness Frequency Within Sole Thickness Ranges

Summary of Findings

The Observation Study was conducted at airports in each of the seven TSA regions in a manner that simulated seasonal differences. As such, it provides a snapshot of the types and characteristics of footwear worn by the air traveling public. The study was designed to capture information relevant to the DHS SaS and BESS-E teams in support of the development of a shoe scanning technology. Observations were made on 5,089 passengers who appeared to travel mostly for leisure.

Almost all travelers wore footwear sized 6 to 12.

Sneakers were the most popular shoe style.

Most footwear closed via laces or slipped on the foot and had leather/suede upper material.

Canvas/cloth, plastic/rubber, and mesh were also popular upper materials.

Most footwear worn by the air traveling public had a rubber sole less than a ½” thick with a rubber heel ½” to 1½” thick.

Heels, boots, and sandals/flip-flops were most likely to have a heel over 1½” thick.

Rubber was the most common heel material, regardless of heel thickness.

Leather, cork, and wood materials were more likely to be seen in thick heels.

As sole vertical thickness increases so too does heel vertical thickness.

.6

.3

.0

.4

% 7

.0

.0

%1

.0

% 2

.7

.0

T H I N - L E S S T H A N 1 / 2 " S O L E M E D I U M - 1 / 2 " T O 1 1 / 2 " S O L E T H I C K - O V E R 1 1 / 2 " S O L E

Thin - Less than 1/2" Heel Medium - 1/2" to 1 1/2" Heel Thick - Over 1 1/2" Heel

Research Brief 11

DHS S&T (2016). Apex Screening at Speed program. Retrieved from:

https://www.dhs.gov/sites/default/files/publications/Apex%20Screening%20at%20Speed %20508%20Approved%20Final_0.pdf

DHS S&T (2018). Behavioral, Economic, and Social Science engine. Retrieved from:

https://www.dhs.gov/sites/default/files/publications/3443_BESS- E_Fact_Sheet_20180330.pdf

Hedge, J. W., & King, S. M. (2018). A survey of the air traveling public. First Responders Group, Science and Technology Directorate, U.S. Department of Homeland Security.

Jones, A., Hartman, T., McMakin, D., Sheen, D., Hall, T., Teseschi, J., … Munoz, E. (2017).

Millimeter-wave shoe scanner system – Phase 1 final report (PNNL-26733). Richard, WA: PNNL.

https://www.dhs.gov/sites/default/files/publications/Apex%20Screening%20at%20Speed%20508%20Approved%20Final_0.pdf https://www.dhs.gov/sites/default/files/publications/Apex%20Screening%20at%20Speed%20508%20Approved%20Final_0.pdf

Research Brief 12

Appendix A

Data Collection Visual Aid

Men’s Dress Shoe Heel

Boot

Sandal Slip-on / Loafer / Boat Shoe

Sneaker

Sole Thickness Heel Thickness

Thin Medium Thick Thin Medium Thick

Up to ½” ½” to 1½” Over 1½” Up to ½” ½” to 1½” Over 1½”

September 2018

A Survey of the Air Traveling Public

The purpose of the current project was to provide a snapshot of air traveler perceptions, perspectives, and preferences about footwear choices, and views on safety and security

(particularly as it relates to shoe screening). The Survey of the Air Traveling Public was developed and administered to a representative sample of individuals who had flown – for leisure or business – in the last year. Data were collected from 1,118 air travelers. Choice of footwear was largely driven by either comfort and convenience or ease of removal when passing through security. In warm weather, sneakers and sandals/flip-flops were the footwear of choice; in cold weather, sneakers and boots were preferred.

Survey respondents appeared to adopt a balanced approach to security and expediency, believing screening should occur more efficiently than is currently the case, yet expressing a willingness to wait as long as it takes to ensure individuals with harmful intentions will be detained. Relatively few air travelers see current screening procedures at U.S. airports as bordering on infringement of their rights as U.S. citizens. The majority of respondents believed that the removal and scanning of shoes was important for flight safety, and three-quarters of air travelers suggested that eliminating the need for shoe removal at the checkpoint would save at least some time moving through security.

The survey also probed more deeply whether travelers would be willing to consider limiting their choice of footwear. When asked a generic question about willingness to limit footwear choices in order to save time going through security, nearly two-thirds of respondents said they would be willing. However, this number fell to less than half of all survey respondents when they were asked if they would be willing to wear a TSA-approved footwear style to save time going through security.

This material is based upon work supported by the Department of Homeland Security Science and Technology Directorate. Any opinions, findings, conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the Department of Homeland Security.

As part of the Apex Screening at Speed (SaS) Program, the Department of Homeland

Security Science and Technology Directorate (DHS S&T) is researching and developing next-generation technologies to enhance the passenger screening process (DHS S&T, 2016). DHS

S&T is currently working with Pacific Northwest National Laboratory (PNNL) to develop a millimeter wave shoe scanning system that would allow passengers to wear their shoes through the security checkpoint. In 2017, PNNL developed and successfully tested a laboratory prototype of a shoe scanner system (Jones et al., 2017). In the next phase of development, which began in FY18, PNNL will develop and deliver a refined engineering test unit to DHS S&T.

Early in FY18, the Behavioral, Economic, and Social Science Engine (BESS-E) agreed to provide support to the SaS program in this effort. The BESS-E team analyzes the social, economic, and behavioral implications of an Apex program’s research, implementation, and diffusion of new technologies, programs, and policies. By matching social science capabilities to Apex program needs, BESS-E is able to provide basic and applied research support that addresses the human impact of the Apex programs (DHS S&T, 2018).

The purpose of RTI International’s work as part of the BESS-E team is to provide behavioral science research support by gathering data on the types of footwear worn by air travelers, to identify stakeholder and passenger perceptions, and to gather feedback on the use and integration of a millimeter-wave footwear scanner in the passenger screening process.

The BESS-E/RTI team’s overall approach involves three separate but related components:

An Observation Study conducted at airports in each of the seven Transportation

Security Administration (TSA)-designated regions to gather data about footwear characteristics of travelers

A survey of the air traveling public’s footwear preferences, views on safety and security (particularly as it relates to shoe screening), and other relevant perceptions about air travel

Interviews with TSA representatives—both at a sample of airports throughout the

United States and at TSA headquarters—concerning shoe scanning, use of shoe scanners, and related issues

The current document provides a description of the activities involved in the development, administration, and analysis of data from the Survey of the Air Traveling Public.

Objectives of the Study

The intent of the survey was to gather information about air travelers’ footwear preferences, perspectives on travel, and issues related to security and safety. These data can be used to inform decisions about shoe scanner design and whether air travelers would be amenable to limiting the types of footwear worn to the airport if it expedites their screening time.

The remainder of this Research Brief includes a description of the methodology applied, (including development of the survey instrument, the data collection procedures, and data analysis plan) the results of those analyses, grouped into four primary sections (demographics, general perceptions and perspectives about air travel, footwear choices and preferences when flying, and safety and security issues) and a summary of findings.

Methodology

Development of the Survey Instrument

Based on input from the DHS S&T BESS-E and SaS teams and PNNL, and a review of available literature and previous studies of air travel, RTI identified content areas around which to develop survey items. Broad content areas included:

travel habits/patterns (e.g., frequency, destination, purpose) membership in travel/security programs (e.g., Precheck, Global Entry) footwear characteristics (e.g., size, style) perspectives on air travel (e.g., preferences, concerns) perspectives on air travel safety and security (e.g., safety versus expediency) demographic characteristics (e.g., height, weight, age, geographic location of residence).

The project team wrote draft items to target each of these content areas. These draft items then went through a series of reviews and revisions, with emphasis on improving questions, response options, overall readability, redundancy, and survey length.

Once a final version of survey items was developed, the full instrument was pretested with members of The GfK Group’s KnowledgePanel (described in greater detail below). A review of pretest data led to the randomization of answer options on one item. All other items performed as expected.

Participants

The survey was conducted with households from The GfK Group’s KnowledgePanel.

KnowledgePanel is a probability-based panel designed to be representative of the U.S.

population. To qualify for this survey, participants had to be age 18 or over and have flown in the last 12 months as determined by the screener question in the survey. Table 1 displays the number of participants sampled, the number who completed the screener portion of the survey, and the number of eligible participants who completed the full survey.

Table 1

Participants Sampled Number who Completed the

Screener Number of Eligible Participants who Completed the Survey

4,564 2,629 1,118

To select a sample from GfK’s KnowledgePanel, active members of the panel were weighted to demographic benchmarks (gender, age, race/ethnicity, education, Census region, household income, home ownership status, and metropolitan area) obtained from the latest

March supplement of the U.S. Census Bureau’s Current Population Survey. Using the derived weights, a study sample representative of the U.S. population was selected. After data collection, the sample was examined for differential responding across demographic groups.

The previously derived weights were then adjusted to ensure that the final dataset remained representative of the U.S. population.

Procedures

Survey data collection started on August 15, 2018. Once assigned to a survey, KnowledgePanel members received a notification e-mail letting them know there was a new survey available for them to take. This e-mail notification contained a custom link that sent members to the survey questionnaire without requiring any further login or password verifications. To secure high rates of response, e-mails were used to encourage nonresponding panel members to complete the survey. E-mail reminders were sent on day three of the field period. GfK provided a modest incentive to encourage participation and foster member loyalty. Data collection ended on August 24, 2018.

Data Analysis Plan

Because the primary purpose for gathering these survey data was to support the ongoing efforts by DHS S&T and PNNL to develop a shoe scanning technology, our data analysis focused on descriptive statistics in order to provide readily usable information for this work. Specifically, our descriptive analyses focused on demographics, general perceptions and perspectives on air travel, footwear choices, and safety and security issues. Means, standard deviations, and frequencies were computed for each item. We also conducted numerous cross-tabulations to examine the relationships between variables. Cross tabulations allow examination of relationships within the data that might not be readily apparent when analyzing total survey responses. These cross-tabulations help answer questions such as: Are the subset of respondents who said screening of passengers and luggage should be accomplished much more quickly and efficiently than it is currently also the ones who are more willing to limit their choice of footwear if it means saving time going through security? Or are those who believe the removal and scanning of footwear is important for flight security also willing to wait as long as it takes to ensure that individuals with harmful intent will be detained?

Results

The results are grouped and discussed in four primary subsections—demographics, general perceptions and perspectives about air traffic, footwear choices and preferences when flying, and safety and security issues.

Demographics

The height of nearly every respondent (99%) was in the 5- to 6-foot range with the average respondent measuring 5 feet 5 inches tall (SD for feet = 0.4; SD for inches = 3.4). The vast majority (92.4%) of respondents weighed between 101 and 250 pounds. Figure 1 shows height and weight distributions of respondents. Like the U.S. population, 50.7% of our respondents were female and 49.3% were male. Respondent average age was 47.14 years

(SD = 16.7).

Half of the respondents had a bachelor’s degree or higher (49.9%), while 28.1% had at least some college. The remainder had a high school education (17%) or less than high school

(5%). Most respondents identified as White, non-Hispanic (65.8%) or Hispanic (15.3%). Black, non-Hispanic and Other, non-Hispanic were roughly equally selected (7.8% and 9.8%, respectively), with the remainder identifying as two or more races, non-Hispanic (1.3%). All regions of the United States were well represented, although the South (31.8%) and the West

(28.5%) had slightly more representation than the Northeast (20.1%) and the Midwest (19.6%).

Figure 1. Respondent Height and Weight Frequencies

Figure 2 displays the frequency of respondent footwear size by gender. Female respondents mostly wear size 6 to 9. Male respondents mostly wear size 10 to 12. Both genders had members who wear size 16 or larger, while only female respondents wear size 5 or smaller.

Figure 2. Respondent Shoe Size Frequency by Gender

Perceptions and Perspectives on Air Travel

Survey respondents flew for leisure more than for business. Only 32% had taken a business-only flight in the past 12 months and only 38.4% had taken a combined business/leisure air travel trip. On the other hand, 94.2% had taken at least one leisure flight in the previous year. Almost a third of respondents (30.4%) noted that when traveling for leisure, 0.7%

35.5%

46.6%

16.6%

0.5% 0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

35.0%

40.0%

45.0%

50.0%

4'11" or shorter

5'0" to 5'5"

5'6" to 5'11"

6'0" to 6'5"

6'6" or taller

Respondent Height

0.5%

29.5%

44.1%

19.0%

5.6%

0.8% 0.5% 0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

35.0%

40.0%

45.0%

50.0%

100 or less

101 -

151 -

201 -

251 -

301 -

351 or more

Respondent Weight (lbs)

0.1% 4.1%

74.9%

18.3%

1.5% 1.0% 0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%

80.0%

2 or smaller

3 to 5 6 to 9 10 to 12 13 to 15 16 or larger

Female Respondent Shoe Sizes

0.0% 0.0%

23.0%

65.9%

10.0%

1.1% 0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%

80.0%

2 or smaller

3 to 5 6 to 9 10 to 12 13 to 15 16 or larger

Male Respondent Shoe Sizes they travel alone; an additional 45.2% travel with a significant other; and just under a quarter

(24.4) travel with a significant other and children.

Regardless of the purpose, most flights were domestic. Europe was the next most frequent destination regardless of flight purpose. Most respondents both checked and carried on bags (59.6%). Just under a third of respondents (29%) carried on baggage only, with the remaining 11.4% checking bags only. Most survey respondents expected to take about the same number of flights in the next 12 months as they had taken in the previous 12 months

(59.5%). Roughly an equal number thought they would take fewer flights (21.9%) or more flights (18.6%). When asked how satisfied they have been over the last 12 months with their air travel experience, 69.9% of respondents said they have been satisfied or very satisfied, while just 9.1% voiced any dissatisfaction with their travel experience.

Respondents were also asked about the importance of various factors when making air travel decisions. Figure 3 displays the frequency at which each factor was chosen as the number 1 most important factor when making air travel decisions. Convenience of the airport and airline schedules were by far the most frequently selected response options. Airport security was not given much consideration as one of the top three most important factors.

Figure 3. Most Important Factors When Making Air Travel Decisions

Respondents were queried concerning the importance of a variety of preflight experiences at the airport and asked to rank them. As shown in Figure 4, most respondents ranked speed of getting through security as most important, followed by efficiency of the boarding process.

43.1%

38.2%

11.4%

5.0%

2.0%

0.1%

0.1%

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

Convenience of the airport

Airline schedules

Total time spent in travel

Safety of air travel

Total travel price

Airport security

Quality of inflight experience

Figure 4. Most Important Preflight Experience

Respondents were also asked about their affiliation with frequent flyer and expedited security programs. In all 44.1% said they did not belong to a frequent flyer program; another

39.7% said they belonged to a program, but their status gave them no extra privileges. Only

16.2% noted status with extra privileges. An even larger percentage of respondents (75.6%) noted that they belonged to no expedited security screening program. Of the remaining quarter of respondents who belong to some screening program, over 60% of this group belong to the

TSA Precheck program. In addition, of those individuals who said they currently do not belong to a screening program, only 12.4% suggested they had plans to join in the next 12 months.

Sixty percent of individuals with status in a frequent flyer program belonged to an expedited security screening program. On the other hand, only 27% of individuals with frequent flyer status but no extra privileges and only 9.8% of individuals not enrolled in a frequent flyer program belonged to an expedited security screening program.

Footwear Choices

Two primary factors were offered as driving footwear choice: comfort…