OMB Clearance Package
Project
Multi-City Taxpayer Assistance Behavioral Study
I. Introduction
Background/overview
The proposed research is designed to examine service usage and the relationship between service and compliance in a controlled environment.
The current proposal is to expand upon a pilot study that was conducted in Atlanta, GA in July and August of 2006. The current study will be conducted in 4 different cities across the country to represent the taxpayer population and also to tap into populations of interest such as taxpayers over the age of 65 and taxpayers who claim EITC. For the current study, the service channels that will be tested are telephone, walk-in assistance, and IRS.gov web assistance. The proposed research will utilize an experimental design in which participants will complete a mock tax return that is similar in content to their own tax situation. Participants will be randomly assigned to one of five groups 1) no assistance, 2) toll-free assistance, 3) walk-in assistance, 4) web-assistance, 5) toll-free, walk-in and web assistance. All participants will have access to publications applicable to their scenarios.
Objectives of data collection
The objectives of the current project are to quantify and measure the relationship between service and compliance in a controlled environment and to examine service usage of taxpayers. The current project will provide feedback regarding customer preference for service channels and how service affects downstream compliance. Data will also be analyzed to examine differences in populations of interest.
II. Methodology
Sample Design
The population for this study consists of taxpayers who are over the age of 18, and have completed their own Federal income tax return with the form 1040 (1040ez, 1040A, or 1040) series in the past 3 years.
Data Collection Date
Data will be collected the week during the months of February- May of 2009. Data will be collected during a one week period in each city. The tentative schedule for data collection is
Atlanta, GA February 17-20, 2009
Seattle, WA March 16-20, 2009
Chicago, IL April 20-24, 2009
Boston, MA May 11-15, 2009
Data to be collected
Several different categories of data will be collected.
Demographic information
Education level, age, income, ethnicity, employment status and internet usage
General Financial Attitudes
General tax information- gathered to provide participants with tax scenarios similar to their own tax situations
Filing status, member of 1040 family most recently filed, EIC claimed, taxable social security, and itemized deductions
Behavioral Data
Time to complete the mock tax scenario
Performance on a mock tax scenario
Average absolute error
Number of errors
Service Usage
Was service used
Type of service used
Questions raised during service interaction
How data will be used
Data will be used to better understand the needs and preferences of taxpayers and the impact of service on compliance. Data will also be used to support future budget requests to the Commissioner, Treasury, OMB and the Congress – and to respond to inquiries from oversight bodies such as GAO, TIGTA and the Oversight Board who frequently request estimates of the costs and benefits of tax law assistance programs.
How data will be analyzed
Analysis of the data will be conducted by researchers within IRS Wage and Investment Research and Analysis.
Data analysis will include basic and advanced statistical techniques including descriptive statistics, and Analysis of Variance (ANOVA). The analysis will be conducted using standard statistical software such as SAS and SPSS.
Who is conducting research?
Research will be conducted by Wage and Investment Research and Analysis and a vendor, Information Experts.
Location- Region/city and facilities
Research will be conducted in federal buildings within IRS training space in Atlanta, GA, Seattle, WA, Chicago, IL and Boston, MA.
Stipends
The stipend will be $60 per participant for a 2 hour session. Participants will also be eligible to earn a $40 dollar bonus if they accurately complete the mock tax scenario. The inclusion of the bonus will help simulate the motivations associated with accurately completing one’s own tax return. Because the motivational effects of the bonus end at the completion of the task the bonus will be awarded to any participant who demonstrated an effort to complete an accurate tax return. We expect that all of the participants will be awarded the bonus.
Recruitment efforts
The contractor will use two channels of recruitment, one for internet users and one for non-internet users. For internet users the contractor will be recruiting through targeted e-mails. For non-internet users the contractor will be recruiting through direct mail. If additional recruitment is needed for non-internet users, the contractor will recruit through newspapers ads, posted ads in public places, and through communications with community organizations such as the YMCA. Once the participants contact the vendor a screening guide will be used to recruit participants for the study.
Efforts to not duplicate research
This study does duplicate a test study what was conducted in Atlanta, GA during July and August of 2006. The purpose of the current test is to expand and to make a few minor methodological changes, such as higher quality recruiting to enhance the project.
Test structure/ design
The proposed research will utilize an experimental design in which participants will come in and complete a mock tax return that is similar in content to their own tax situation. Participants will be randomly assigned to one of five groups: 1) no assistance, 2) toll-free assistance, 3) walk-in assistance, 4) web-assistance, 5) toll-free, walk-in and web assistance. All groups will have access to publications applicable to their scenarios. Mock returns will be examined to determine if the availability and use of IRS taxpayer services results in a significant reduction in reporting errors as compared to the control group.
II. Participants Criteria
In order to participate in the study, participants must have the following characteristics:
Be over the age of 18
Native English Speakers
Have self-prepared a federal tax return in the past 3 years
1040 filer
Education level of grade 10 or above
30% non-internet users
IV. Security, Disclosure, Anonymity
The data returned to IRS W&I Research will not have any form of identifying information relating specific records to individual taxpayers. Nonetheless, Research Personnel will ensure that privacy, security, and anonymity of the aggregated results will receive utmost attention. Public and official access to the information will be tightly controlled. The computer files containing this tabulated information will remain password protected at all times. Data security approaching level C-2 will be accomplished using the Windows XP operating system.
V. Burden Hours
This research is expected to use a total 2,911 of burden hours.
Email 25,000 internet users to screen 2,500
Direct mail 15,000 non-internet users to screen 1,500
Total screening 4,000 to recruit 1,200 for 1,000 to show
Of the 4,000 screened 2,680 are expected to be screened out by the 5th question
25,000 read email @ 1 minutes each = 417 hours
15,000 read direct mail @ 30 seconds each = 125 hours
2,680 screened out by 5th question @ 2 minutes each = 89 hours
1,320 complete full screening @ 10 minutes each = 220 hours
1,200 final reminder contact @ 3 minute each = 60 hours
1,000 participants complete scenario and debrief at 2 hours each = 2,000hours
Total burden hours = 2,911hours
VI. Cost of Study
The estimated contractor cost of this study is $437,756. This figure includes the stipends for participants ($120,000). The estimated travel cost for eight IRS employees is $40,800.
VII. Attachments
Initial E-mail Contact Message- The attached e-mail message has been developed in conjunction with the vendor. It is expected that changes, if any, made to the e-mail will be minor and will not impact the overall burden estimate.
Direct mail- The attached direct mailing has been developed in conjunction with the vendor. It is expected that changes, if any, made to the e-mail will be minor and will not impact the overall burden estimates.
Screeners Guide- The attached screener has been developed in conjunction with the vendor. It is expected that changes, if any, made to the screener’s guide will be minor and will not impact the overall burden estimate.
Script for Follow-up Call- The attached script has been developed in conjunction with the vendor. It is expected that changes, if any, made to the script will be minor and will not impact the overall burden estimate.
Debrief questionnaire-The debrief questionnaire has been developed in conjunction with the vendor. It is expected that changes, if any, will be minor and will not impact the overall burden estimate. The time to complete the debrief questionnaire is estimated at 10 minutes each and is included in the 2 hour estimate for the study.
Instructions- The instructions will be read to the participants by the facilitator and a printed copy will also be included in the packet given to the participants. The instructions will be different for each group depending on random assignment to the different service options.
Scenarios – There will be 20 different mock tax scenarios in total covering 4 general topics, EITC, taxable social security, standard deduction greater than itemized deduction, and itemized deduction and 1040A standard deductions. A scenario representing each of the four most common filing statuses (single, married, head of household, and married filing separately) will be created for each of the four topics. Sample scenarios are attached.
| File Type | application/msword |
| File Title | OMB Package |
| Author | ysflb |
| Last Modified By | qhrfb |
| File Modified | 2008-12-22 |
| File Created | 2008-12-11 |