The National Survey of Teen Relationships and Intimate Violence (STRiV), [United States], 2013-2020 (ICPSR 36499)

Time Period(s) View help for Time Period(s)

Date of Collection View help for Date of Collection

Data Collection Notes View help for Data Collection Notes

Study Purpose View help for Study Purpose

The National Survey of Teen Relationships and Intimate Violence (STRiV) examined the changing nature of adolescent dating relationships, particularly those marked by adolescent relationship abuse (ARA). More specifically, this study was designed to produce nationally representative estimates of the prevalence of multiple forms of ARA among youth (ages 10-18), to document the characteristics of abusive relationships during adolescence, to assess ARA risk factors, and to situate these estimates within the environment of adolescents' key social relationships and communications.

The goals of this study included the following:

1. Provide a national portrait of the prevalence of varying categories of ARA victimization and perpetration, including levels of physical and emotional injury, and describe how exposure to these forms of ARA varies by gender, socio-economic status and other key demographic characteristics.
2. Identify specific conditional attitudes, dating relationship characteristics, and peer network dynamics associated with ARA risk, and to determine whether these pathways are uniquely gendered.
3. Investigate specific consequences of ARA to dating relationships, and whether the adolescent discloses information about ARA incidents to peers and others and their responses.

Study Design View help for Study Design

Researchers recruited a nationally representative sample of 5,105 households with at least one resident youth between the ages of 10 and 18 to complete an adult parent or caregiver baseline survey and a separate child survey online from October 2013 to January 2014. Of the 5,105 recruited households, 556 of the sample actively refused to participate in the study and 1,816 did not respond to the invitation. While the GfK/KnowledgePanel used to collect respondents consisted of a database of resident children, a further 279 households were screened ineligible (for example, the expected youth did not reside in the household), equivalent to about 7% of the responding sample. The 2,354 completed youth surveys represent a participation rate of 50% among eligible households. A somewhat larger group of parents agreed to do the baseline survey (n=2,645); 2,354 of the children of these parents completed a baseline survey.

If the recipient parent or adult caregiver (PCG) consented to participate, a web survey algorithm randomly chose an eligible child in that household to participate in the study. Next, the participating child and PCG received an invitation by e-mail to complete the surveys, which were presented sequentially with child assent required prior to child participation. For the youth survey, the PCG consent rate (for self and child participation) was 82.6%, and the child assent rate was 98.3%. Communications with the PCG stressed the importance of allowing the selected child to complete the survey privately. While the 40-minute youth survey was available in English or Spanish, few youth respondents selected the Spanish option (n=128, 5%). Each youth and parent was provided a small cash incentive ($10 each) for completing the baseline survey.

At each data collection wave, GfK reviewed the STRiV cohort status to prepare invitations for those who remained active Knowledge Panelists and those who had withdrawn from the KnowledgePanel (i.e., accepting no new studies) but remained active in the STRiV study; a small group of respondents who had withdrawn permanently from the KnowledgePanel and all ongoing studies were lost to follow-up. Households were offered a $20 incentive for completing the baseline surveys through the GfK points system.

Sample View help for Sample

Respondents to the National Survey on Teen Relationships and Intimate Violence (STRiV) were recruited from the GfK/KnowledgePanel, a national household address-based probability sample (50,000+ members ages 18 and older) of the United States. A full description of the panel is available from GfK.

From an address-based sampling (ABS) frame covering approximately 97% of U.S. households, randomly sampled households were invited to join KnowledgePanel through postal invitations (English and Spanish) and by telephone follow-up. Using dual sample frames, panel members were further recruited via listed and unlisted telephone numbers, telephone and non-telephone households, and cell phone only households, as well as households with and without internet access. To assure national representativeness, researchers applied the KnowledgePanel statistical weights (see "Weights" below).

Time Method View help for Time Method

Universe View help for Universe

US households with at least one resident youth age 10-18.

Unit(s) of Observation View help for Unit(s) of Observation

Data Type(s) View help for Data Type(s)

Mode of Data Collection View help for Mode of Data Collection

Description of Variables View help for Description of Variables

STRiV includes the following variables:

Response Rates View help for Response Rates

50% of eligible households participated in the baseline survey. 62.5% of households completing the baseline survey also participated in the Wave 2 follow-up. The Wave 3 response rate of youth baseline respondents was 70.5% and the Wave 4 response rate of youth baseline respondents was 69.5%. The Wave 5 response rate of youth baseline respondents was 61.3% and the Wave 6 response rate of youth baseline respondents was 67.6%.

Original Release Date View help for Original Release Date

2017-06-30

Version History View help for Version History

Weight View help for Weight

Provided in the datasets are panel base weights for youth and parent/caregiver that take into account a range of sampling and non-sampling error (e.g., non-response to panel recruitment and panel attrition), and was employed in a probability proportional to size (PPS) selection method for drawing sub-samples from KnowledgePanel. Using U.S. Census demographic and geographic distributions, a sample-specific post-stratification process was conducted (applying an iterative raking procedure) to adjust for survey nonresponse and elements related to the study-specific sample design (oversampling households with youth), resulting in a weighted sample distribution at baseline (wave 1) that approximates the 2010 U.S Census estimates, and adjusts for nonresponse at each subsequent wave.