The current Harris Interactive panel consists of approximately 4 million adults in the United States with over 132,000 participants who are parents of 10- to 14-year-old children and are actively participating in research. Harris Interactive recruits members to join the panel by issuing invitations through the Internet, telephone, mail, e-mail, print, or in person. Their recruiting partners have included major Internet portals, television stations, global corporations, and their panel members. In each case, panelists agreed to receive e-mail invitations on a periodic basis from Harris Interactive. By general policy, panelists receive between two and three invitations to respond to unique surveys on general topics each month. They may also receive one to two reminder invitations for each survey. Because Harris Interactive maintains socio-demographic data on all panelists, it is possible to select only panelists with certain characteristics to be invited to participate in a survey. In the present case, only parents of children aged 10 to 14 will be selected for invitation to participate.
The panel is not representative of any population in a statistical sense because respondents were not recruited by probability sampling. However, Harris Interactive uses targeted sampling and weighting to build samples that reflect the known distributions by socio-demographic characteristics of the general population—e.g., in this case, the race/ethnicity of parents in the United States.
Because the primary objective of the 4parents.gov Web site survey is to collect customer satisfaction data (from parents asked to visit the Web site), the use of the Harris Interactive online panel is an appropriate modality for this study. Furthermore, because parents of children aged 10 to 14 are already identified in the Harris Interactive panel, drawing from this sample for the survey will yield useful results in a very cost-effective manner.
The 4parents.gov Web site survey will be conducted as a one-time online survey among parents of 10- to 14-year-olds in the United States selected from the Harris Interactive online panel. Data collection will be conducted in fall 2009, during the rollout of a television ad that focuses specifically on the 4parents.gov Web site. The sample will total 800 parents and will consist of 100 mothers and 100 fathers each of four groups of children (boys aged 10 to 12 and 13 to 14, and girls aged 10 to 12 and 13 to 14), for a total of 400 mothers and 400 fathers.
The survey will be approximately 30 minutes in length, including 15 minutes for responding to a questionnaire, and 15 minutes for visiting assigned sections of the 4parents.gov Web site. The survey will assess all sections of the Web site selected by OPA for assessment among the sample of parents, who are directed to the Web site via the online survey. After some initial questions, parents will be asked to rate their degree of interest in various Web site topics and then will be provided a hyperlink to the Web site with the request that they find two topics that have been assigned to them. Sections will be assigned to respondents partly on the basis of their expressed interest in the topics and partly on the need to assign all sections to adequate numbers of respondents. Parents will be asked to view the two sections for about 5 minutes each and then return to the online survey to give feedback about the content of the sections, navigating the Web site, and their planned use of the Web site. The survey will also collect data on the respondents’ attitudes, beliefs, and behaviors related to parent-child communications about sex; these data will help to explain the parents’ reactions to the Web site. Information about study participants’ children will also be collected, including their stage of pubertal development and interest in romantic relationships; these data will also help explain parents’ opinions about the Web site. Socio-demographic characteristics (e.g., age, gender, marital status, race, ethnicity, education, and income) will not need to be asked in the survey, but will be included in the dataset constructed by Harris Interactive using socio-demographic data that are routinely maintained on the panel. Harris Interactive will provide weights in the dataset that can be used to weight the survey sample to be as similar as possible to the national distributions by race and ethnicity.
We will conduct tests of statistical significance of differences among groups of parents in their assessment of their assigned Web site topics (taking into account their degree of interest) and their reactions to the 4parents.gov Web site. We will also utilize multivariate logistic regression models to examine the independent relationships of several characteristics—socio‑demographic characteristics, parents’ attitudes toward teen sex, their previous experience talking with their child about sexual intercourse, and children’s stage of pubertal developmental and interest in sexual relationships—with parents’ interest in various Web site topics and their evaluations of these sections, as well as more general assessments of the Web site. The emphasis of the study is on customer satisfaction of a panel of parents who are asked to visit the 4parents.gov Web site.
In partnership with Harris Interactive, a sample will be selected from members eligible to participate in the survey: English-speaking members of their panel living in the United States who are parents or parent surrogates (e.g., stepmother, grandfather, foster parent) of children aged 10 to 14. The sample will total 800 persons, consisting of 100 mothers and 100 fathers of each of four groups of children (boys aged 10 to 12 and 13 to 14, and girls aged 10 to 12 and 13 to 14), for a total of 400 mothers and 400 fathers. A sample of eligible panel members will be assigned to participate in the survey. They will be invited to do so via a notice in their password-protected e-mail account that the survey is available for completion. Nonrespondents will receive two e-mail reminders from Harris Interactive requesting their participation in the survey. If respondents respond to some questions but do not complete the survey, they will be sent e-mails that remind them that they have only a few questions left and should return to complete the survey in order to receive their incentive. Copies of the e-mail notifications are in Attachment 7. The survey will be self-administered and accessible any time of day for a designated period. The survey will be password-protected so that participants can complete the survey only once. Informed consent will be sought from the parents for participation in the Web survey. A copy of the consent form is included in Attachment 5. Parents will consent by selecting the appropriate link on the Web screen. Panel members may leave the panel at any time and may choose not to take any survey assigned to them. They may also choose to participate in the survey but not to answer any particular question.
An incentive of 100 points (roughly equivalent to $1 cash) and a chance at winning a bi-monthly Harris Interactive sweepstakes will be offered to participants who complete the survey. The incentive is intended to recognize the time burden placed on them, encourage their cooperation, and convey appreciation for contributing to this important study. A detailed description of Harris Interactive’s panel recruitment methodology is in Attachment 8.
As data collection progresses, data collection production reports from Harris Interactive will be used to track survey completion rates by the eight classes of respondents (defined by gender of parent and age and gender of child), so that adjustments can be made in their efforts to convert break-offs and new respondents can possibly be recruited. The frequencies with which each section of the Web site has been visited and assessed by respondents will be tracked. Initially, we will assign parents to visit two sections that they have rated as being of relatively high interest. However, to ensure that all sections are assessed by sufficient numbers of parents, we will, if needed, assign parents to sections that they have not rated as being of highest interest but that have been infrequently assigned. Parents’ level of interest in their assigned topics will be taken into account in the analysis.
The following procedures will be used to maximize cooperation and to achieve the desired high response rates:
About 10% of the total sample will be invited to do the survey on the first day, and the remainder will be invited 1 or 2 days later.
All respondents who do not do the survey will automatically receive reminder invitations 48 hours after the initial invitation e-mail has been sent.
Respondents who do not return to the survey after beginning it will be sent special reminders to return to the survey to complete it.
Harris Poll Online incentives in the amount of 100 points (equivalent to $1 cash) and a chance at the Harris Interactive bi-monthly sweepstakes will be offered to participants who complete the survey.
To ensure that the final sample consists of the 100 mothers and fathers with a child in the targeted gender and age group, a rolling tally of completed surveys by these characteristics will be kept. New invitations will be sent out to parents with a child in the categories needed to reach the targeted balanced sample. This method will ensure that a total sample of 800 parents will be achieved, with the targeted distribution by parent’s gender and child’s gender and age.
Harris Interactive will provide an e-mail address to all sampled individuals and invite them to e-mail any questions or concerns about any aspect of the study.
Harris Interactive data collection staff will work with RTI project staff to address concerns that may arise.
Pilot test data collection was conducted from [will insert DATE] to [will insert DATE]. Harris Interactive contacted nine panel members to participate in the pilot testing of the 4parents.gov Web site survey. The pilot questionnaire asked panel members about any difficulties they had with the survey itself (questions that lacked clarity, were too sensitive, or too difficult to answer; the overall length; problems accessing the Web site; etc.). The results of this pilot test showed: [to be filled in].
The agency official responsible for receiving and approving contract deliverables is:
Allison Roper
240-453-2806
Office of Population Affairs/DHHS
1101 Wootton Parkway, Suite 700
Rockville, MD 20852
The persons who designed the data collection are:
Helen Koo, DrPH
919-541-6351
RTI International
3040 Cornwallis Rd
Research Triangle Park, NC 27709
Kevin C. Davis, MA
919-541-5801
RTI International
3040 Cornwallis Rd
Research Triangle Park, NC 27709
The person who will collect the data is:
Dawn Amodeo
323-522-6599
Harris Interactive
250 W. 1st Street
Claremont, CA 91711
The person who will analyze the data is:
Helen Koo, DrPH
919-541-6351
RTI International
3040 Cornwallis Rd
Research Triangle Park, NC 27709
Abreu, D. A., & Winters, F. (1999). Using monetary incentives to reduce attrition in the survey of income and program participation. Proceedings of the Survey Research Methods Section of the American Statistical Association.
Shettle, C., & Mooney, G. (1999). Monetary incentives in U.S. government surveys. Journal of Official Statistics, 15, 231-250.
| File Type | application/msword |
| Author | DHHS |
| Last Modified By | DHHS |
| File Modified | 2009-07-10 |
| File Created | 2009-07-10 |