US Welcome Survey Supporting Statement B

April 2008

SUPPORTING STATEMENT B

Collections of Information Employing Statistical Methods

Welcome to the United States Survey

OMB Control Number: 1601-0003

1. Describe the potential respondent universe and any sampling or other respondent selection methods to be used.

Given that the 11 Ports of Entry (POEs) are purposely selected for inclusion into this survey, direct inference to the universe of all foreign visitors cannot be made from the sample alone (i.e., results will be limited to the population of foreign travelers at the selected ports only). RTI and DHS will work together to assess the similarity (i.e., how representative) between these 11 POEs and the remaining frame of all, or portions of, other POEs. This assessment will allow DHS to evaluate how similar the survey results would be for any other POE with comparable characteristics.

In order to ensure that the collected samples are reflective of the target population at each POE, port-specific 1-in-N sampling paradigms will be designed and implemented. Aspects considered in the development of these paradigms will include, but are not limited to: expected traveler traffic per day, expected response rates, peak times of travelers, number of CBP processing areas, number of interviewer staff per port, expected time per interview, and the number of feasible work hours per data collection day.

We have assumed a 50% response rate among eligible sampled visitors (with only those refusing due to connecting flight time constraints being deemed ineligible). Given the Welcome to the United States Survey has not previously been conducted, this rate is assumed based upon response rates for similar intercept-type surveys, including ones conducted by Transportation Security Administration (TSA), the Delaware Division of Motor Vehicles (DMV), and the Johns Hopkins School of Medicine.

2. Describe the procedures for the collection of information.

2.1 Overview

RTI will collect data from 4 airport POEs, 4 Southwestern land border POEs, and 3 Northern land border POEs. RTI will utilize two field interviews (FIs) to travel to each port to conduct the interviews with people as they pass through CBP. At least one of the interviewers will be fluent in Spanish in order to conduct interviews with Spanish-speaking respondents. RTI plans to spend 3 ½ days at each airport and 1 ½ days at each land port. The needed number of completed interviews for each type of estimates described in Section 3.2.1 accounts for the varying length of stays. Additionally, by spending 3 ½ days at each airport RTI will better ensure coverage across the numerous peak times and terminals serviced by CBP.

RTI will work with DHS to obtain descriptive information about each port to better prepare for data collection. In addition, RTI and DHS will jointly conduct a site visit (potentially including a land and airport POE) to review and better understand CBP procedures and logistics. This site visit will allow RTI to better prepare logistically for possible scenarios that may occur during data collection. RTI will also incorporate such descriptive port information into its FI training program to enhance FI productivity and efficiency during data collection trips.

2.2 Ports

2.2.1 Port Assignments

RTI’s FIs will visit the 11 ports in our sample across 7 trips. Each of these trips will be planned in advance, taking into account geographic proximities of the selected POEs and associated travel costs so that the work is completed as efficiently as possible.

2.2.2 Interviewing Times

Interviewing times for each port will be preplanned using the peak travel time information provided by DHS. The FI training program will include review of these schedules, and the Data Collection Task Leader will ensure travel and production times align with said schedules.

2.3 Interviewer Training

2.3.1 Training Materials and Possible CBP Materials

The FI training will be conducted at RTI’s corporate headquarters in Research Triangle Park, NC. RTI will work with DHS to develop and conduct an effective training program, which will ensure FIs are provided with all information and materials necessary to effectively conduct the interviews. RTI will develop training materials such as recruiting instructions and aids, specific port information, copies of the questionnaire, potential frequently asked questions from respondents, among other things. In correlation with DHS, RTI will also provide the FIs with any additional CBP materials that may help in their training.

2.3.2 Training Program

The training program will provide the FIs with background information about the Professionalism Initiative of CBP and the Rice-Chertoff Initiative to help them understand the purpose of the study. We will train the FIs on the sampling procedures. Much of the training will focus on respondent cooperation and avoiding non-response. The training program will also include time devoted to review of the questionnaire, questionnaire administration, interaction with CBP staff, and security. RTI will provide DHS with a detailed draft training agenda to allow for feedback prior to the conduct of the training (to ensure all topics that should be presented are included). Additionally, FIs will be trained on understanding and working with the handheld electronic data collection devices. The training program is anticipated to consist of approximately 8 hours (including instruction, demonstration, and exercises).

2.4 Data Collection Procedures

2.4.1 Sample Member Selection

FIse will sample persons as they come through CBP by implementing the port-specific sampling paradigm (selecting a 1-in-N person). The FI will count people as they come through CBP using the defined rule. Once the FI identifies the person who is sampled, and after the visitor has concluded their business with CBP, he/she will approach the potential respondent and recruit them to complete the interview.

2.4.2 Recruiting

RTI will develop a recruitment protocol for the FIs to follow. This will include suggested verbiage to use when recruiting potential respondents, answers to questions that respondents may have, and additional information about the study that the FIs can use to ease respondent concerns. Specific attention will be given to the professionalism of the FIs during FI training. Appearance, tone of voice, and demeanor will be addressed to ensure the FIs best represent DHS and RTI. These same attributes will also assist FIs in avoiding refusals. During recruitment, the FI will attempt to complete an interview with the sampled respondent. If the person refuses, the FI will attempt to collect a reason for the refusal.

2.4.3 Interviewing

During the interview, which is assumed to require 5 minutes to administer, the FI will ask each question and then enter/record the answer the respondent gives on the handheld data collection device. The FI will address any concerns or questions that the respondent may have during the interview. Once the full interview is complete, the FI will provide the sample member with an in-kind incentive (to be identified and finalized in coordination with DHS) to show appreciation for the time they took to participate. Once the incentive is given, the FI will thank the respondent and then prepare for the next interview. Once all materials are gathered, the FI will then continue with the sample selection process for the next interview.

2.4.4 CAI Administration

RTI will utilize Computer Assisted Interviewing (CAI) as the mode of data collection. Each FI will be issued a handheld electronic data collection device (e.g., HP iPAQ) to administer the questionnaire to respondents. RTI will program and test each handheld device prior to training to ensure proper case identification, eventing, and questionnaire functionality (e.g., question ordering, skip patterns, and translations). FIs will receive training on how to properly use the device and administer the questionnaire. Additional procedures will be outlined concerning the saving of data, the maintenance of the issued hardware, and gaining assistance with any encountered technical difficulties. RTI programmers will provide technical support throughout each data collection trip in order to resolve problems in as close to real-time as possible.

Data that is collected for each interview will be saved on the handheld device’s inset memory card. FIs will also conduct nightly back-ups of the data, saving it onto an external, removable memory card. In the rare instance of hardware failure, this will ensure that no more than one day’s worth of data is at risk. The data for each trip (usually consisting of 2 POEs) will be downloaded at RTI upon the FIs’ return. This will allow RTI to commence analysis on the data that had been collected during the previous week.

2.4.5 Port Checklist

In collaboration with DHS, RTI will develop a Port Checklist to collect specific information about each port as observed during the data collection visit. Through discussion with DHS, we will decide who will complete this checklist, and what items need to be included on the checklist. This data may help us better understand respondent data and concerns.

3. Describe methods to maximize response rate and to deal with issues of non-response.

Several tools and procedures are being utilized in this survey to assist in maximizing the response rates and to compensate for non-response, both unit and item. These include:

1. Limiting the total interview time to 5 minutes (e.g., connecting flights or trip schedules may be negatively impacted by lengthy interviews);

2. Use of CAI data collection to mitigate the burden of the respondent from manually completing a paper interview;

3. Use of a small thank you gift as a token of appreciation

4. Ensuring that at least one of the FIs at each POE is able to administer the questionnaire in Spanish (which will be resident on the CAI handheld device).

5. Data collection efforts will be monitored in as close to real-time processing as is feasibly possible. This will allow RTI and DHS to identify problematic aspects (e.g., lower than expected response rates or high item-nonresponse), evaluate potential corrective actions, and implement adjustments as quickly as possible in order to minimize detriment (e.g., sample bias or lower than desired sample sizes) to the survey results.

6. RTI will work with DHS to obtain specific descriptive information about each port to better prepare for data collection. In addition, RTI and DHS will jointly conduct a site visit to one land port to review and better understand CBP procedures and logistics. This site visit will allow RTI to better prepare logistically for possible scenarios that may occur during data collection. RTI will also incorporate such descriptive port information into its interviewer training program to enhance interviewer productivity and efficiency during data collection trips.

7. Weight calibration steps will be utilized to account for respondent non-response. Statistical methods of imputation (e.g., “Hot” deck, predictive modeling, etc.) will be used to account for survey item non-response, when and where appropriate.

4. Describe any tests of procedures or methods to be undertaken.

RTI will evaluate the draft instrument using its Questionnaire Appraisal System (QAS). QAS is a structured, standardized instrument review methodology that assists a survey design expert in evaluating questions relative to the tasks they require of respondents, specifically with regard to how respondents understand and respond to survey questions. The QAS also allows the reviewer to evaluate the structure and effectiveness of the questionnaire form itself. In part, the QAS is a coding system (i.e., an item taxonomy) that describes the cognitive demands of the questionnaire and documents the question features that are likely to lead to response error. These potential errors include errors related to comprehension, task definition, information retrieval, judgment, and response generation. This forms appraisal can be used as a starting point for identifying particular instructions, questions, or response categories that may be problematic and could compromise the quality of the data in surveys. We will thoroughly test the CAI questionnaire functionality, including data storage and exchange. Quality control procedures will be in place throughout the survey process to detect any system errors, random or otherwise.

5. Provide the name and telephone number of individuals consulted on statistical aspects of the design and the name of the agency unit, contractor(s), grantee(s), or other person(s) who will actually collect and/or analyze the information for the agency.

Sample Design and Analysis: Michael A. Penne, MPH 919.541.5988

(RTI International)

Data Collection Design: Chris Ellis 919.541.6480

(RTI International)

Data Collection: Brian Evans 919.541.7366

(RTI International)

OMB 83‑I (10/95)