OMB0135_Justification_2010

INFORMATION COLLECTION REQUEST (ICR):

OMB 83-I SUPPORTING STATEMENT AND PRIVACY IMPACT ASSESSMENT

A. Justification

1. Information to be collected and circumstances that make the collection of information necessary

The U.S. Consumer Product Safety Commission (CPSC) is an independent federal regulatory agency that was created in 1972 by Congress in the Consumer Product Safety Act. In that law, Congress directed the Commission to “protect the public against unreasonable risks of injuries and deaths associated with consumer products.”

Most of the work of the staff of the CPSC Division of Human Factors involves applying human factors data and principles to the design and evaluation of consumer products. Some of these data are available through human factors texts, professional journals, and the like. However, relevant research and data on some human characteristics and attributes, such as certain population stereotypes and consumer perceptions related to specific product use, often are not readily available to the staff. Obtaining relevant data typically would require the staff to conduct research; however, the timeframes available in which to complete an assessment are usually too short to plan and complete formal research studies. Additionally, the staff often lacks the necessary funds to carry out these studies. Thus the staff often must rely solely on its own expert opinions or judgments. The ability to pose questions or scenarios to consumers to obtain their opinions and perceptions could provide the staff with preliminary information from which more accurate expert judgments can be made. This information could also be used to identify areas worthy of additional research by the staff. To obtain this information, the staff has been involved in two related information-collection activities: a Consumer Opinion Forum and the Respondent Registration process.

The Consumer Opinion Forum is an internet-based survey or poll that is available for voluntary participation by interested consumers 18 years of age and older through the CPSC website. The Consumer Opinion Forum periodically posts a set of questions, scenarios, or similar information (collectively referred to as “questions” from this point on) to solicit opinions and perceptions from respondents. Before being posted in the Consumer Opinion Forum, each set of questions is prepared by the CPSC Human Factors staff, cleared within the CPSC for public dissemination, and submitted to OMB with a quick-turn-around approval request.

Once a new set of questions is posted in the Consumer Opinion Forum, potential respondents receive e-mail invitations to respond. Individual participants are not sent invitations more frequently than once every four weeks. New questions may be prepared and posted in the Consumer Opinion Forum more frequently than this, but invitations to respond are sent only to those participants who have not received invitations within the prior four weeks.

From the time a set of questions is posted on the Consumer Opinion Forum, participants are given a minimum of two weeks to respond; the CPSC Human Factors staff has completed two surveys to date, and participants were given at least four weeks to respond in each survey. To provide a response, participants go to the CPSC website, log into the Consumer Opinion Forum, and respond to the questions posted. At the end of the posting period, no more responses to the questions are permitted and the Human Factors staff can retrieve the responses.

The Respondent Registration process is the method by which consumers can choose to participate in the Consumer Opinion Forum. Only those consumers who visit the CPSC website and voluntarily register for the Consumer Opinion Forum are permitted to participate in the Forum. To register, volunteers are required to fill out an online registration form that asks for the following information:

• The respondent’s e-mail address

• A personal password for access to the Consumer Opinion Forum

• The respondent’s sex

• The respondent’s date of birth

• The zip code of the respondent’s primary residence

• The respondent’s number of children

• The sex and date of birth for each child

Registration commenced before any questions were posted in the Consumer Opinion Forum, but potential respondents are permitted to register at any time and for as long as the Forum is available for participation. The intended population group of respondents includes all “adult” users of consumer products over the age of 18 except for current employees of CPSC. However, given the nature of the Forum, only those consumers with internet access and an e-mail address are able to participate. As of July 30, 2009, 2300 people have registered to participate in the Consumer Opinion Forum. The staff does not have a predetermined upper limit or goal for the total number of registrants/respondents, and plans to allow as many respondents as are interested to register.

2. Use and sharing of collected information

The information collected by the Consumer Opinion Forum is used by the CPSC Human Factors staff to make more informed and accurate expert judgments on consumer behavior. Precisely how the information is used depends on the information needed at the time. To illustrate how this information might be used, brief descriptions of the two surveys already completed by the staff follow.

The first survey focused on obtaining consumer experiences with recalled products that are under the jurisdiction of CPSC. Questions asked included how respondents typically learn about product recalls, whether and how respondents have registered consumer products, whether and how respondents have participated in product recalls, and problems respondents might have encountered when trying to participate in a recall. The second survey focused on obtaining consumer experiences with electrical outlets that contain ground fault circuit interrupters, or GFCIs. Questions asked included where GFCI receptacles are located in and around the home, whether respondents know how to test them, whether respondents actually do test them and how often, and whether their GFCI receptacles have ever failed to work. Other possible uses for information obtained through the Consumer Opinion Forum include obtaining typical product behavioral or use patterns, identifying comprehension problems with various warning language, and determining possible interpretations of warning pictorials.

The responses collected through the Consumer Opinion Forum, and the demographic information associated with those responses and collected through the Respondent Registration process, may be made available to the public through the CPSC website. The e-mail addresses and personal passwords of the respondents, however, will not be shared outside the CPSC. All information collected during the Respondent Registration is stored in a registration database. When a potential respondent logs into the Consumer Opinion Forum using his or her e-mail address and password, this information is compared against information in this database (i.e., the e-mail address and password) to confirm that he or she is registered to participate. This process is automated. As participants provide responses to the Consumer Opinion Forum, the responses and the associated demographic information—retrieved from the registration database—are loaded into a separate, response database. Once all responses are collected, the information in the response database is accessible to the CPSC Human Factors staff and may be made available to the public (e.g., through the CPSC website). The respondents’ e-mail addresses and passwords, however, will not be made public and will only be accessible to those CPSC staff who are granted access to the registration database.

3. Use of information technology (IT) in information collection

All information collection is done electronically through the CPSC’s Internet Information Server. The staff restricts participation in the Consumer Opinion Forum to those people who register online through the CPSC website. The information collected during the Respondent Registration process is held in a registration database. Once registered, respondents must log into the Consumer Opinion Forum through the CPSC website, using the respondent’s e-mail address and a password, to provide the staff with responses. All responses are recorded online and transferred automatically into a response database.

This method of information collection was selected because it allows the staff to securely collect information about the characteristics of the respondents without having to repeatedly collect this information each time a respondent provides a response. During the registration process, consumers are asked to provide the staff with certain personal demographic information, such as their sex and their date of birth (see #1, Information to be collected and circumstances that make the collection of information necessary, for a complete list of information to be collected). Once a respondent has registered and later logs onto the website to provide the staff with a response, the response will be associated with the respondent’s profile, allowing the staff to know the characteristics of the individual associated with that response. Although this information could be requested from the respondent each time he or she responds to a question posed in the Consumer Opinion Forum, the staff believes this would be unnecessarily redundant and prone to error. Additionally, the staff believes that providing the requested information through the CPSC website is more secure and less prone to transcription error than having the respondent provide the information by phone, e-mail, or an alternate method.

4. Efforts to identify duplication

The intent of the Consumer Opinion Forum is to obtain information that is not readily available elsewhere. Some information on consumer opinions and perceptions related to product use is available online. For example, the websites Epinions (www.epinions.com) and Amazon.com permit people to “review” or comment on various products, including consumer products of interest to the staff. These comments sometimes provide insight into consumer perceptions about what a product is believed to be intended for, how the product was used or misused by the individual, etc. The staff sometimes searches these websites and can identify usage patterns or similar information from them. For the most part, however, this is not the website’s focus or intent, and the information that is available from these websites is limited. For example, respondents to those websites often limit their responses to whether the product was “good” or not, or whether the product met their expectations. Other websites may have similar online forums, bulletin boards, or message boards that allow people to provide feedback or comments about product use; for example, the Woodworker’s Association website (www.woodworking.org). The staff intends to rely on these other websites to the extent possible. However, the staff often encounters unique situations or products that are not dealt with on these websites. In these cases, the ability to pose questions and scenarios to consumers is valuable.

The Human Factors staff meets occasionally to identify new sets of questions to consider posting in the Consumer Opinion Forum. The originator of each set of questions proposed is asked to discuss efforts expended to locate existing research or information on the topic. The Human Factors staff will not consider a question unless it concludes that effort has been put into locating research that could be used to answer the question and that such research is not available, unless there is a reason to believe that the data that are available are not accurate.

5. Impact on small businesses

The information is not collected from small businesses or other small entities. We expect individual consumers to participate in the Forum.

6. Consequence to Federal program or policy activities if collection is not conducted or is conducted less frequently

Currently, the Human Factors staff must sometimes rely solely on expert judgment about consumer behavior, perceptions, and similar information related to consumer products and product use. Not conducting the information collection activity, therefore, would not reduce the quality of assessments currently completed by the staff. However, conducting the information collection activity provides the staff with evidence that informs the staff’s assessments and can provide insight into consumer perceptions and usage patterns that may not be anticipated by the staff. Hence, the staff believes that conducting this information collection activity improves the quality of the staff’s assessments. Conducting the collection activity less frequently would provide the staff with fewer opportunities to obtain the necessary information.

7. Special circumstances requiring respondents to report information more often than quarterly or to prepare responses in fewer than 30 days

Some of the human factors assessments for which the results of the Consumer Opinion Forum may be particularly useful include Product Safety Assessments (PSAs). PSA requests are received from the CPSC Office of Compliance, and may ask for such things as an assessment of how consumers are likely to interpret an instruction or warning, or how likely it is that a consumer will use a product in a particular way. The staff may be able to use the Consumer Opinion Forum to identify information of this sort. However, the staff cannot predict when it might receive a request for a PSA. Frequently, such requests are received more often than quarterly, and the staff’s assessments must be completed in much less time than one quarter (i.e., three months). Although the surveys already completed by the staff have not been posted in the Forum more often than quarterly, restricting the ability to pose questions more than once every quarter could potentially limit the usefulness of the information collection activity.

Additionally, PSA requests often have due dates that fall within 30 days from the date in which the request is submitted. Given that questions that are to be posed to respondents through the Consumer Opinion Forum must be internally cleared for public dissemination before being posed to respondents, the amount of time available to post and await responses in support of PSA-related work could easily be less than 30 days. Some information sought through the Consumer Opinion Forum will not be needed in fewer than 30 days. However, providing interested respondents with 30 days or more for responses to all questions posed in the Consumer Opinion Forum would potentially limit the Forum’s usefulness, especially for PSA-related work.

8. Agency’s Federal Register Notice and related information

An FR Notice was published on November 13, 2009. Comments were not received.

9. Decision to provide payment or gift

The Commission will not provide any payment or gift to participants in the Consumer Opinion Forum.

10. Assurance of confidentiality

All responses to the Consumer Opinion Forum are entirely voluntary. As described earlier, those who register to participate in the Forum occasionally receive e-mail notifications that a new set of questions has been posted and that they are invited to respond. To provide responses, the potential respondent must go to the CPSC website and log into the Consumer Opinion Forum. If the potential respondent chooses not to respond to a particular set of questions, he or she can simply ignore the notification/invitation. Additionally, the potential respondent may log into and later exit the Forum without responding to the posted questions. Similarly, during the Respondent Registration process, an individual may choose not to provide the information requested. However, in so doing, the individual will not be permitted to participate in the Forum.

During the Respondent Registration process, respondents are informed that their participation is voluntary and that information exchanged during registration will be encrypted for privacy during transit using a VeriSign SSL (Secure Sockets Layer) Certificate. By clicking on a VeriSign logo, respondents can see the CPSC’s specific server ID information and verify that cpsc.gov is a legitimate VeriSign secure site. Respondents also are informed that their e-mail address and personal password will not be distributed outside the CPSC.

Each respondent will be identified by his or her e-mail address. However, these e-mail addresses will not be included in any reports or summaries of the data collected through the Consumer Opinion Forum. The CPSC Office of General Counsel has reviewed the proposed information collection activities and has found that the Privacy Act does not apply because the data are not maintained in a “system of records” from which information is retrieved by the name of the individual or by some identifying number, symbol, or other particular assigned to the individual. The CPSC staff will not access or retrieve records (i.e., responses to the Consumer Opinion Forum) using e-mail addresses; instead, the staff will use demographic criteria to access and retrieve records.

11. Questions of a sensitive nature

The information to be collected is described in #1, Information to be collected and circumstances that make the collection of information necessary. The staff will not pose questions of a sensitive nature, such as sexual behavior or attitudes, religious beliefs, or other matters commonly considered private.

12. Estimate of hour burden to respondents

As of July 30, 2009, the number of people who have registered to participate in the Consumer Opinion Forum is 2300. The staff has not established an upper limit or goal for the total number of respondents, and registration continues to be open to anyone accessing our website and demonstrating an interest in participating. Based on the rate at which people are registering, however, the staff doubts that the total number of registrants will increase substantially over the next few years and most likely will not exceed the staff’s original estimate of 5000 respondents.

The staff estimates that the total time for each respondent to register to participate in the Consumer Opinion Forum will likely be not more than 10 minutes, depending on the number of children in the respondent’s household. As described in #1, Information to be collected and circumstances that make the collection of information necessary, registration information includes e-mail address, password, designation of sex, date of birth, zip code, number of children, and the sex and date of birth for each child.

The two main purposes for having respondents register are to provide CPSC Human Factors staff with basic demographic information on the respondents while reducing the overall burden on respondents. Registering respondents will permit the Human Factors staff to invite or notify respondents, using the e-mail addresses they provide, when a new set of questions has been posted in the Consumer Opinion Forum. The likely alternative would be to simply post questions on the Forum and to ask interested respondents to check the website regularly to see when a new question has been posed. The Human Factors staff believes that having respondents register would reduce the total burden hours placed on them.

Additionally, requiring respondents to register would provide the staff with related demographic data without having to ask for this information every time the respondent provides a response. Gathering this information will also permit the staff to avoid sending notifications to participants who are unlikely to provide the staff with useful information because of their demographics. For example, the staff may be interested in understanding how parents, in particular, are likely to respond to a certain scenario; therefore, the staff may not be interested in responses from consumers without children. Similarly, the staff may be interested in posing certain questions only to females, or to consumers 65 or older. By having respondents register, the Human Factors staff can limit its invitation to only those respondents of interest, thereby limiting the burden on respondents whose demographics are not consistent with the intended target population.

Once respondents have logged into the Consumer Opinion Forum, they will also be given the opportunity to update their registration information as needed. For example, respondents who give birth to another child may choose to update their registration information so that it includes the new child, and that child’s sex and date of birth. Updating this information would take only a fraction of the total estimated initial registration time.

When a new set of questions is posted on the Forum, potential respondents will receive e-mail invitations to respond. Individual respondents will not be sent invitations more frequently than once every four weeks. New questions may be prepared and posted in the Consumer Opinion Forum more frequently than this, but invitations to respond will only be sent to respondents who have not received invitations within the last four weeks.

Logging into the Consumer Opinion Forum most likely takes less than one minute since it only requires respondents to enter their e-mail address and personal password. The amount of time required for a respondent to provide responses to the questions varies considerably depending on the specific number, type, and complexity of questions being posted. This variability does not allow the staff to make a single estimate of the total time to respond. The Human Factors staff, however, has decided on a self-imposed upper limit of 15 minutes total. For each survey, the staff estimates the time required to complete it, and only posts those surveys that are estimated to take no more than 15 minutes to complete. The staff has launched and closed two surveys and has obtained the starting and ending times for each completed survey, but determining the actual time required to complete a survey is difficult because respondents can begin a survey and return later to complete it. The resulting data show that the average completion times for survey #1 and survey #2 were 9.5 minutes and 5.0 minutes, respectively, but the median completion times are more likely to reflect the true “average” time required by a respondent to complete the surveys; these times for survey #1 and survey #2 were 4.3 minutes and 3.4 minutes, respectively. The 95^th percentile completion times for these surveys—that is, the time in which 95 percent of respondents completed the surveys—were 16.3 minutes and 11.5 minutes, respectively.

13. Estimate of total annual cost burden to respondents

During the approximately two years¹ that the Consumer Opinion Forum has been operating, 2300 people have registered to participate. Based on the staff estimate that it takes no more than 10 minutes to register, the registration burden is estimated to be about 383 burden hours, or about 192 burden hours per year.

During the same timeframe, the staff has completed two surveys and is in the process of clearing a third. This works out to roughly one survey per year. The number of respondents to survey #1 was 385, which was equivalent to a response rate of about 44 percent (based on the number of registered participants at that time), and the median time required to complete the survey was 4.3 minutes. The total burden for survey #1, therefore, was about 28 hours. The number of respondents to survey #2 was 343, which was equivalent to a response rate of about 23 percent (based on the number of registered participants at that time), and the median time required to complete the survey was 3.4 minutes. The total burden for survey #2, therefore, was about 19 hours. Based on these data, the staff estimates that the burden to respondents for the next survey is unlikely to exceed 84 hours (44 percent response rate for 2300 potential respondents at about 5 minutes per survey). Surveys may become more frequent, however, and the staff foresees the possibility of up to four surveys per year. This increased frequency would result in a total annual burden of up to about 337 hours and could approach 733 hours if the total number of people who register to participate reaches 5000 (44 percent response rate for 5000 potential respondents at about 5 minutes per survey for four surveys).

Based on the above numbers, the total estimated burden for new registrations and surveys, combined, is unlikely to exceed 925 hours annually (no more than 733 hours for four surveys per year, plus no more than about 192 hours for new registrations). Thus, assuming an hourly rate of $29.39,² the total burden cost is unlikely to exceed $27,000 per year.

14. Estimate of annualized costs to the federal government

Based on the staff’s experience over the last couple years, the total staff time for preparing questions for the Consumer Opinion Forum, maintaining the Forum, and analyzing the responses from the Forum is about one staff month per year, or about one staff month per survey. If one assumes that the surveys could occur up to four times per year, the total staff time could be four staff months annually. The resulting annualized cost to the federal government is estimated to be $55,360 (4 staff months total × $13,840 estimated cost per staff month.)

15. Program changes or adjustments

Based on our initial experience, we have adjusted the hour burden to respondents to more accurately reflect the actual time required by respondents to complete surveys. Additionally, the number of surveys completed annually is considerably less than initially estimated. Although the staff anticipates surveys occurring with greater frequency than has been the case recently, the total number of surveys annually is still likely to be much less than originally estimated. The staff has also updated the hourly rate employed to calculate the estimated burden costs.

16. Plans for tabulation and publication

Not applicable

17. Rationale for not displaying the expiration date for OMB approval

Not applicable.

18. Exception to the certification statement

Not applicable.

2. Collections of Information Employing Statistical Methods

Not applicable.

1 The registration portion of the Consumer Opinion Forum was launched in June 2007.

2 U.S. Department of Labor. (2009). Employer Costs for Employee Compensation – March 2009 [On-line]. Available: http://www.bls.gov/news.release/pdf/ecec.pdf.