B. COLLECTION OF INFORMATION EMPLOYING STATISTICAL METHODS
CPSC will submit each proposed survey instrument for clearance individually and will provide an explanation of the study design and purpose of the survey. Below we provide a general description of our data collection methods.
1. Respondent Universe and Sampling Methods
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 research relies on a convenience sample and is not intended to yield results that can be generalized to the overall population. Results of this research will not be used to make statements representative of the universe of study, to produce statistical descriptions (careful, repeatable measurements), or to generalize the information beyond the scope of the sample.
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. 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. Procedures for the Collection of Information
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.
All data collection and analysis will be performed in compliance with OMB standards and guidance, Privacy Act, and Protection of Human Subjects requirements.
3. Methods to Maximize Response Rates and Deal with Non-response
In recent Consumer Opinion Forum surveys, respondents were given between 3 and 4 weeks to respond. Generally, most responses occur within the first week or two, with very few responses coming afterward. During the most recent survey, CPSC Human Factors staff sent a reminder to nonrespondents 3 days prior to the closing date of the survey. The total number of responses increased by more than 20 percent during that final 3-day period (319 responses before reminder, 388 responses at the close of the survey). Staff anticipates issuing similar reminders during future surveys.
As previously noted, the information collected under this generic clearance is not intended to yield statistically representative estimates; thus, these methods to deal with non-response are adequate for this purpose.
4. Test of Procedures or Methods to be Undertaken
Before each information collection is implemented, we will pilot test the survey instrument. Lessons from the pilot test will be identified, and changes as necessary will be incorporated into the final survey instrument. All pilot tests will involve internal staff or no more than nine members of the public. If we require more than nine pretest respondents from the public, we will submit the pretest protocol for review and approval under this generic clearance.
5. Individuals Consulted on Statistical Aspects and Individuals Collecting and/or Analyzing Data
All surveys conducted to date have been in support of human factors work. Thus, CPSC Human Factors staff members have been the primary individuals involved in developing the questions, the sequence of questions, and similar survey design issues. Before a survey is launched, however, it must be internally cleared for public dissemination. Thus, the survey is reviewed and approved by management of the CPSC technical directorates, including Engineering Sciences, Health Sciences, Epidemiology, and Economic Analysis. Comments and suggested edits by these directorates are incorporated into the final survey design. CPSC Human Factors staff members are the recipients of the data upon completion of a survey, and generally perform basic statistical analyses on these data.
| File Type | application/msword |
| File Modified | 2011-02-22 |
| File Created | 2011-02-22 |