National Intimate Partner and Sexual Violence Survey (NISVS): General Population Survey Raw Data, 2016/2017 (ICPSR 38960)

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 primary objectives of the National Intimate Partner and Sexual Violence Survey are to describe (1) the prevalence and characteristics of sexual violence, stalking, and intimate partner violence; (2) who is most likely to experience these forms of violence; (3) the patterns and impact of the violence experienced by specific perpetrators; and (4) the health consequences of these forms of violence.

Study Design View help for Study Design

The National Intimate Partner and Sexual Violence Survey (NISVS) is an ongoing national random-digit-dial (RDD) telephone survey of women and men in the United States that began in 2010. NISVS samples noninstitutionalized English- or Spanish-speaking persons 18 years and older and uses a dual-frame strategy that includes landlines and cell phones. It is conducted in all 50 states and the District of Columbia. The survey instrument was programmed in a computer-assisted telephone interview (CATI) system that contained the actual interview including question text, response options, interviewer instructions, and interview probes.

The current study was originally to be conducted in four consecutive 6-month periods from mid-March 2016 through mid-March 2017, with a gap for data processing and for making any necessary changes to the survey, and then two more consecutive 6-month periods from mid-March 2018 to mid-March 2019. However, due to delays in fielding the first survey period, both the first and second collections began fielding 1 week apart (collection 1 began September 9, 2016 and collection 2 began September 15, 2016), and ran concurrently through March 14, 2017. Given the increased complexity, strain on resources, and amount of data collection required to conduct these two efforts in parallel, the collection that began September 9, 2016 was extended by 3 months, ending on May 28, 2017.

The survey was administered from mid-April through mid-October of 2018 after multiple data collection changes designed to increase the response rate. Despite these efforts, the 2018 response rate showed little improvement over that for data collected from 2016 through 2017 and data collection was suspended. Further, the cooperation rate in the 2018 data declined compared to the prior periods. Most of those contacted in 2016 and 2017 participated in the survey, but less than one-third did so in 2018. Given the low cooperation rate, data from 2018 were excluded and this collection reflects data collected from 2016 and 2017 combined (referred to as NISVS 2016/2017).

Sample View help for Sample

The National Intimate Partner and Sexual Violence Survey (NISVS) employs a dual-frame, stratified random digit dial (RDD) sampling design, with continuous data collection. To meet the challenges of rising non-coverage rates in U.S. landline-based telephone surveys, NISVS implemented a dual-frame design where both landline and cell phone frames were sampled simultaneously.

List-Assisted Landline Frame. The landline sampling frame was comprised of hundred-banks of telephone numbers where each bank had at least one known listed residential number. A hundred-bank is the 100 telephone numbers that are generated by fixing the first eight digits of a telephone number and changing the last two digits (e.g., (800) 555-55XX). Known business numbers were excluded from the frame. In addition, non-working numbers were removed after sample selection through screening.

Cell-Phone Frame. The cell phone frame consisted of phone numbers in telephone banks identified as active and currently in use for cell phones. At the time the sample was drawn, directory listings of cell phone numbers were not available. Thus, list-assisted screening was not possible.

Stratification for State-level Estimates. In order to provide statistically stable estimates at the state level, survey samples were stratified by state (and by sampling frame within state), balancing between having statistically stable state estimates and reduced weight variation in national estimates from oversampling smaller states.

Within-Household Selection. When reaching a household in the landline sample, the interviewer asked about the number of males and females living in the household. In a one-adult household, the adult was automatically selected to participate. In households with only two adults, the person on the phone or the other adult in the household was randomly selected. When there were more than two adults in the household, the adult with the most recent birthday was selected. This within-household selection has been found to be less likely to lead to overrepresentation of females in the pool of respondents compared to using only the most recent birthday method for all households with more than one adult (Rizzo, Brick, and Park, 2004). Because cell phones are personal use devices, the person answering the cell phone was selected as the respondent, if eligible.

Nonresponse Phase. To increase participation, NISVS was administered as a two-phase survey. Phase One was the main data collection phase. Respondents in the first phase were offered an incentive of $10 to participate in the survey. A random subsample of non-respondents from the first phase was selected during Phase Two, with the goal of reducing non-response and non-response bias. The second phase included a substantially higher incentive ($40) to further encourage participation.

Time Method View help for Time Method

Universe View help for Universe

Non-institutionalized English and/or Spanish-speaking adults aged 18 or older in the 50 states and District of Columbia meeting sampling eligibility criteria.

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

The questionnaire is divided into several sections and includes information on demographic characteristics of the respondent, health conditions, victimization experiences (including psychological aggression and coercive control by an intimate partner, physical violence by an intimate partner, stalking victimization, and sexual violence). The questionnaire also includes information about perpetrators and follow-up questions.

Demographics - Respondents were asked their year of birth, education level, race and ethnicity, place of birth (if not U.S. born, number of years lived in the U.S.), their total household income, and zip code.

Health - Respondents were asked about the following health conditions: asthma, irritable bowel syndrome (IBS), diabetes, high blood pressure, HIV/AIDS, frequent headaches, chronic pain, difficulty sleeping, serious difficulty hearing, blindness or difficulty seeing, difficulty walking or climbing stairs, difficulty bathing or dressing oneself, difficulty concentrating due to a physical, mental, or emotional condition, and difficulty doing errands alone due to a physical, mental, or emotional condition.

Victimization - Respondents were asked to report the number of people who had perpetrated a series of violence behaviors (described below) against them.

• Psychological aggression and coercive control (PA), which includes behaviors such as insulting, humiliating, being made fun of, and coercive control behaviors: restricted access to money, restricted access to family/friends, keeping track and demanding to know whereabouts, threats of physical harm, threatened to hurt themselves or commit suicide because they were upset, made decision that should have been yours to make, and destroyed something important to you.
• Physical violence (PV), which includes behaviors such as slapping, pushing or shoving, being hit with a fist or something hard, being kicked, being hurt by pulling hair, being slammed against something, attempts to hurt by choking or suffocating, being beaten, being burned on purpose, having a partner use a knife against the victim, and having a partner use a gun against the victim.
• Stalking victimization (S), which includes a pattern of unwanted harassing or threatening tactics used by a perpetrator that causes fear or concern for the safety of oneself or others. Behaviors include: unwanted following and watching, unwanted approaching or showing up, using GPS technology or equipment to monitor or track location, leaving strange or potentially threatening items for you to find, sneaked into home or car and did things to scare you, used technology such as hidden camera, recorder, or computer software to spy from a distance, unwanted phone calls, unwanted text, photo, email or social media messages, and sent unwanted cards, letters, flowers, or presents.
• Sexual violence (SV), which includes verbal sexual harassment in a public place, unwanted sexual contact (e.g., being kissed in a sexual way, fondled or grabbed), sexual coercion (unwanted sexual penetration after being pressured in a non-physical way), rape (completed forced penetration; attempted forced penetration; alcohol or drug-facilitated completed penetration), being made to penetrate another person (completed; attempted - males only; alcohol or drug-facilitated).
• Exposure to IPV as a Minor, which included witnessing of psychological aggression or physical violence against one's parent by the parent's current or former intimate partner (asked only of respondents with children under age 18 living in the household).

Perpetrator Information - Perpetrator data were collected on the victim-perpetrator relationship and perpetrator sex for each set of behaviors. Data were gathered at the time questions were asked rather than at the end of the survey in a relationship module. Individual perpetrators are no longer linked to specific violence behaviors across the survey. Victim-perpetrator relationship information was collected for all perpetrators (up to 10 for lifetime and up to 5 for past 12 months), but additional detail was collected about the first chronological perpetrator within each violence type and subtype: respondent's age at time of experience with first perpetrator (any type), first perpetrator's age at the time of the first experience (any type), and respondent's age at time of experience with first intimate partner perpetrator, first intimate partner perpetrator's age at the time of the first experience (any type).

Follow-up Questions - Respondents who reported experiencing any of the stalking tactics were asked a series of follow-up questions to assess for victimization criteria: one or more perpetrators did the tactics on more than one occasion and made the respondent feel fearful, threatened, or concerned for safety or made physical threats. Respondents who met the stalking victimization criteria were asked questions regarding their age at the first experience and information about the perpetrator(s). Finally, respondents who met the stalking victimization criteria were asked how many perpetrators they experienced stalking with in the past 12 months and their relationship to the perpetrator(s).

Respondents who reported experienced contact sexual violence (i.e., sexual violence involving physical touch or contact) were asked a series of follow-up questions regarding the age at first victimization and perpetrator information; this information was collected for the following sexual violence sections: Unwanted sexual contact, sexual coercion, completed and attempted rape, and completed and attempted made to penetrate. Questions to assess the immediate impact of sexual violence as follows: for sexual coercion (pregnancy [females only] and sexually transmitted infections); for rape and made to penetrate combined (fear, concern for safety, pregnancy [females only], sexually transmitted infections, and physical injury).

A series of general follow-up questions were asked of respondents who reported victimization by an intimate partner in the preceding sections of the survey (i.e., psychological aggression and coercive control, physical violence; stalking; and sexual violence victimization). Respondents were asked about the impact of the violence they experienced by any intimate partner perpetrator(s). These questions included whether they were ever concerned for their safety; if they were ever fearful; presence of PTSD symptoms (nightmares; constantly on guard, watchful, or easily startled; felt numb or detached from others, their activities, or surroundings); various physical injuries; mental or emotional harm ; their need for and ability to get services (medical care, housing services, community services (crisis hotline, victim's advocate, law enforcement, or legal services); and number of missed days of work or school.

Response Rates View help for Response Rates

The overall weighted response rate for the 2016/2017 data collection for NISVS was 7.6 percent. The weighted cooperation rate for the 2016/2017 data collection for NISVS was 58.6 percent.

Original Release Date View help for Original Release Date

2024-09-05

Version History View help for Version History

Weight View help for Weight

Sample weights are essential for computing national estimates using these data. These weights reflect sampling features, non-response, coverage, and sampling variability. There are several main weight components that contribute to final sampling weights: selection, multiplicity, non-response, and post-stratification.

The selection weight accounts for the varying selection probabilities in the landline and in the cell phone frames, the within household probability of selection, and the subsampling of non-respondents in Phase Two of data collection.

The multiplicity weight component takes into consideration that some sample members had both landline and cell phone services, thereby having multiple chances of entering the survey. The non-response weight accounts for the variation in response rates within the selected sample.

The post-stratification weight adjusts the product of the selection, multiplicity, and non-response weights to match the population distribution on main demographic characteristics. This is accomplished using benchmark counts from census projections to correct for both coverage and non-response, which allows the landline and cell phone samples to be merged together.

Two main sets of weights were computed for analyzing NISVS data. Applying the same principles in constructing the various weight components, one set of weights were computed for all partial and complete interviews, while another set of weights were computed for the complete interviews only. An interview is defined as "complete" if the respondent completed the screening, demographic, general health questions, and all questions on all five sets of violence victimization, as applicable. An interview is defined as "partial" if the respondent completed the screening, demographic, and general health questions and at least all questions on the first set of violence victimization (stalking).

Users of NISVS data will need to provide the "design" and "stratum" specifications to whichever software (such as SUDAAN or a SAS complex survey procedure) is being used for analysis and apply the correct weights when producing estimates. This is necessary because the NISVS data were collected through a complex sampling design. Information about the Design and Weight Variables can be found in the Orientation to the NISVS Raw Data for Analysis.