Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Feb 1.
Published in final edited form as: Violence Against Women. 2011 Feb;17(2):251–266. doi: 10.1177/1077801210397756

APPLYING EXPERIENCE SAMPLING METHODS TO PARTNER VIOLENCE RESEARCH: SAFETY AND FEASIBILITY IN A 90-DAY STUDY OF COMMUNITY WOMEN

Tami P Sullivan a, Enna Khondkaryan b, Nancy P Dos Santos a, Erica N Peters a
PMCID: PMC3079539  NIHMSID: NIHMS262438  PMID: 21307033

Abstract

An experience sampling method (ESM) rarely has been applied in studies of intimate partner violence (IPV) despite the benefits to be gained. Because ESM approaches and women who experience IPV present unique challenges for data collection an empirical question exists: is it safe and feasible to apply ESM to community women who currently are experiencing IPV? A 90-day, design-driven feasibility study examined daily telephone data collection, daily paper diaries, and monthly retrospective semi-structured interview methods among a community sample of 123 women currently experiencing IPV to study within-person relationships between IPV and substance use. Findings suggest that ESM is a promising method for collecting data among this population and can elucidate daily dynamics of victimization as well as associated behaviors and experiences. Lessons learned from the application of ESM to this population are also discussed.

Keywords: Partner violence, Experience Sampling Method, Research methods, Feasibility, Safety

An experience sampling method (ESM) rarely has been applied in studies of violence against women, or intimate partner violence (IPV) specifically, despite the method’s increasing application and successful implementation in other fields of research over the past three decades (Hektner, Schmidt, & Csikszentmihalyi, 2007). ESM is a data collection method whereby respondents self-report experiences and behaviors in their natural environment and in near real time, which improves the reliability and validity of data by reducing recall bias (Csikszentmihalyi & Larson, 1987; Hektner et al., 2007; Stone, Shiffman, Alienza, & Nebeling, 2007). The absence of ESM in IPV research is disappointing because, despite the challenges of applying ESM to investigations of women who experience IPV, there potentially are extraordinary benefits. Benefits include ESM’s ability to provide data to better elucidate the phenomenology of IPV; examine proximal relationships between IPV, behaviors, and experiences; explicate temporal relationships; and identify contingencies between IPV and related phenomena which can be targeted in intervention.

Concerns about and challenges of conducting research with women who have experienced violence and trauma are well documented. Principal concerns are centered on the physical safety and mental well-being of participants (Black, Kresnow, Simon, Arias, & Shelley, 2006; Campbell, Adams, Wasco, Ahrens, & Sefl, 2010; Carlson et al., 2003; Clements & Holtzworth-Munroe, 2009; Garcia-Moreno, 2002; Griffin, Resick, Waldrop, & Mechanic, 2003; Langford, 2000; Newman & Kaloupek, 2004; Newman, Walker, & Gefland, 1999; Walker, Newman, Koss, & Bernstein, 1997). Given the sensitive nature of research questions, the most fundamental challenges are centered on the willingness of women to participate in research (Rosenbaum & Langhinrichsen-Rohling, 2006). Regarding these concerns and challenges, findings of the above cited studies show that the majority of women who participate in research on victimization and trauma tolerate the interview(s) well and do not report symptoms of distress upon completion. Furthermore, the majority typically express benefiting from participating in the study. These findings suggest that women who have been victimized are safe to participate in research on victimization and trauma, are open to discussing their experiences in research, generally tolerate the experience well, and generally perceive the experience positively. However, regarding the application of ESM studies, concerns, challenges, and women’s willingness to participate are unknown.

Application of ESM in the Current Study of IPV and Substance Use

Specifically in regard to research on IPV and substance use, no studies have used ESM to examine within-person, proximal relationships between women’s daily experiences of IPV and events of substance use. To detect if contingent relationships exist between IPV victimization and substance use [e.g., as suggested by the self-medication hypotheses (Khantzian, 1997, 2003)], within-person associations must be analyzed rather than between-person associations. Within-person associations avoid the problem of the ecological fallacy (Robinson, 1950), whereby data, because they have been aggregated for a group, misrepresent experiences of individuals in that group. ESM approaches, which gather data intensively to study within-person associations, present unique challenges for data collection. No precedent exists for the application of ESM approaches to gather detailed, sensitive information from this special population of women. Thus, an empirical question exists: is it safe and feasible to apply ESM to community women who currently are experiencing IPV?

The aim of the project is to examine the effectiveness of two ESM approaches that yielded high response rates and reliable data among other populations (Rosenbaum, Rabenhorst, Reddy, Fleming, & Howells, 2003; Searles, Perrine, Mundt, & Helzer, 1995; Todd, Armeli, Tennen, Carney, & Affleck, 2003) and one retrospective approach to study within-person relationships between IPV events and drug and alcohol use on a day-to-day basis. A 90-day, design-driven feasibility study examined daily telephone data collection (i.e., ESM), daily paper diaries (i.e., ESM), and monthly retrospective semi-structured interview methods among a community sample of women currently experiencing IPV. The expert recommendations of Stone, Shiffman, Csikszentmihalyi, Hektner, and others (Hektner et al., 2007; Stone & Shiffman, 2002; Stone et al., 2007) regarding experience sampling and other methods of intensive longitudinal data collection guided the design of the study. In the absence of specific guidelines for community women who are experiencing IPV, decisions about study design were made by the investigative team based on the principal investigator’s clinical expertise working with women who experience IPV, research expertise conducting cross-sectional studies with this population, and other study investigators’ research expertise conducting ESM studies among the general population.

METHODS

Participants

One hundred twenty-three women were randomly assigned to one of three conditions for a duration of 90 days – (1) daily telephone data collection (n = 47), (2) daily paper diaries (n = 36), and (3) monthly, retrospective in-person interviews (n = 40). All women participated in baseline and follow-up assessments. Those in the telephone data collection and paper diary conditions recorded event-level IPV and/or substance use data on a daily basis. Participants in the monthly retrospective condition recorded event-level data for each of the previous 30 days during three separate monthly, retrospective in-person interviews.

Participants were recruited from an urban community in the Northeast with a population of 123,626 (U.S. Census Bureau, 2001). Recruitment materials were posted in local businesses such as grocery stores, laundromats and shops, selected state offices such as the Departments of Adult Education and Employment, and in waiting rooms, bathrooms, and exam rooms of urban-area primary care clinics and emergency departments. Eligibility was determined via a phone screen. Inclusion criteria were: (a) female gender, (b) age 18 or older, (c) the use of any amount of drugs or alcohol in the previous 30 days, (d) current involvement in a heterosexual intimate relationship of at least six months duration with current contact at least twice a week, (e) presence of physical victimization of the woman in the past 30 days, by her current male partner as measured by selected screening questions from the Conflict Tactics Scale-2, CTS-2 (Straus, Hamby, Boney-McCoy, & Sugarman, 1996), and (f) residency in the greater-urban area. Exclusion criteria were (a) inpatient psychiatric hospitalization within the last year, and (b) current residence in a shelter/group home (determined a priori because structured living environments affect women’s ability to have contact with their partners, the experience of IPV, and the use of substances). At the time of development of this manuscript, 123 women had completed the study and data collection was ongoing; the following information is about and subsequent analyses are performed with data from these 123 women.

Procedures

All participants completed: (a) a baseline interview consisting of a 3-hour, semi-structured, computer-assisted interview (and for participants in the telephone data collection and paper diary conditions, a condition-specific, 45-minute training session regarding how to participate in the study for 90 days); (b) 90 days of daily data collection, the method of which was determined by randomization to one of the following three conditions: 1. daily telephone data collection which involved calling into the phone system each day for 90 days; 2. daily paper diary condition which involved filling out paper surveys each day and mailing them in for 90 days; or 3. monthly retrospective interviews which involved retrospectively reporting on 30 days during each of three in-person interviews, (c) a follow-up interview on or after day 91 of the study where women participated in a two-hour, structured, computer-assisted interview (for participants in the monthly retrospective interview condition, their report of the final 30 days of the study took place during this interview). After completion of the interview, all participants were debriefed. Throughout the study they were provided with a list of community resources for employment, food, and benefits assistance, mental health and substance use treatment, and domestic violence services. Participants were asked if they wanted assistance accessing resources and assistance was provided when requested.

Participants were remunerated in cash for their participation and were eligible for payments immediately after each in-person interview and, for daily participation, every 30 days (although, if a participant requested payment before the end of the 30 day period payment was made). All participants were remunerated $45 for each baseline and follow up interview. Participants in the telephone data collection and paper diary conditions were remunerated $2 per phone call or paper diary. Those in the monthly retrospective interview condition were remunerated $45 for each interview. Total possible remuneration for telephone data collection and paper diary condition participants was $270 and monthly retrospective interview condition participants was $180.

Daily data collection assessed the occurrence and timing of physical, sexual, and psychological IPV victimization and perpetration, alcohol use quantity and frequency, drug use frequency (by specific substance), and other key issues that impact the occurrence of IPV or substance use (e.g., face-to-face contact with partner, incarceration or hospitalization of participant, etc.). The protocol was standardized across conditions.

Telephone data collection

A toll-free number (i.e. 1–800-number) and four telephone lines were dedicated for the study. Participants were provided with a laminated business card with the study telephone number and instructions for using the telephone data collection system. On this card was a study assigned identification number which allowed participants to access this specific study’s questions via the telephone data collection system; the identification number also allowed the study team to monitor participant’s data. To maintain confidentiality and minimize risk to participant safety, (a) the laminated card noted the “Yale Breast Cancer Study” as a header and (b) a © copyright symbol preceded identification numbers to deter others from comprehending the content and purpose of the laminated card. The survey was the PI’s recorded voice providing instructions and asking close-ended questions. Participants were required to use the telephone’s keypad to enter their responses. During their baseline interviews, participants were instructed on how to call into the telephone system and respond to close-ended questions regarding any events of IPV or substance use and the start and stop times of these events (e.g. pressing the 1 button for “yes,” the 2 button for “no;” or entering start and stop times followed by the number sign button). The potential existed that participants may not have reported IPV or substance use on a given day to more quickly complete the call. As a deterrent, participants were required to answer approximately the same number of questions each time, regardless of IPV or substance use activity (Searles, Helzer, & Walter, 2000). Calls were not monitored by a live person. However, women were able to connect directly to a domestic violence hotline while completing a study call by pressing numbers on the phone’s key pad. Daily telephone data were uploaded and monitored every two to three days. Three days of missed calls resulted in a follow-up phone call from the research assistant to the participant to assess her safety and barriers to participation.

Paper diaries

Participants in this condition were given daily paper diary sheets (i.e., brief surveys) and each was pre-printed with study questions and pre-filled with a participant’s identification number and appropriate date. During the baseline interview, women were instructed to record daily, any events of IPV or substance use and the start and stop times of events. Participants were provided with stamped, addressed envelopes and were instructed to mail diaries to the research team daily. To minimize risk to participant safety, envelopes were pre-printed with a return address specifying “Breast Cancer Study.” Each day, a research assistant collected mail, recorded the stamped postmark, and logged-in the diaries received. These compliance-enhancing procedures were implemented to limit the amount of back-filling that might otherwise occur. Three days of missed diaries resulted in a follow-up phone call from the research assistant to the participant to assess her safety and barriers to participation. Data from the diary sheets/surveys were not manually entered into a database but rather, were scanned (i.e, teleforms).

Monthly, retrospective, in-person interviews

Participants in this condition completed four in-person interviews, during three of which they self-reported IPV and substance use events for the previous 30 days (daily IPV and substance use information for the prior 30 days was not collected at the baseline interview so that the 90-day data collection windows were identical across the three conditions). The interview format was derived from the timeline calendar methods devised by Sobell and Sobell (1995). Participants were prompted with a daily calendar of the previous 30 days. The only information preprinted on the calendars was standard United States holidays. During each monthly retrospective interview, the interviewer asked the participant to note days of personal significance and these were marked on the calendar. Next, days on which events of IPV occurred were elicited and participants were asked to report the types of IPV that occurred and the start and stop times of those events. The same procedures used to collect information about IPV were followed to collect information about substance use.

Measures

Feasibility

For each condition, feasibility was assessed by participants’ (a) perceived safety in completing assessments, (b) reported ease of participation, (c) reported honesty, (d) compliance with the instructions (i.e., recorded information by telephone daily, recorded and mailed information daily, and attended scheduled monthly interviews), (e) percent attrition by methodology (f), preference for methodology assigned, or alternate methodology (i.e., daily telephonic recording, written daily diary, and in-person monthly interview), and (g) degree and pattern of missing data. Assessments of feasibility, with the exceptions of compliance with instructions, percent attrition, and degree and pattern of missingness, were measured during the follow-up interview using a self-report questionnaire developed by the first author. Assessment of reported honesty, reported ease, and preference for assigned versus alternate interviewing methodology was modeled after the measures used by Corkrey and Parkinson (2002). Although the questions were condition-specific there were a number of questions asked of all participants. Respondents rated their level of honesty, ease of complying with instructions for their assigned condition, and safety in participating on a 5-point Likert scale. Data on safety, ease of participation, reported honesty, and preference for assigned vs. alternate method of participation are the focus of this manuscript.

Intimate partner violence

Women’s physical, psychological, and sexual IPV victimization, as well as negotiation and injury, were measured by the 78-item CTS-2 (Straus et al., 2003). To gain more specific information about sexual IPV victimization, these constructs also were measured by the Sexual Experiences Survey, SES (Koss & Oros, 1982). For the purposes of these analyses, the CTS-2 physical IPV and injury and the SES sexual IPV victimization scores were used. At baseline, a referent time period of three months was used to assess the woman’s IPV victimization by her current partner via the woman’s self-report.

Data Analysis

Descriptive statistics are provided for participants’ self-reported safety, ease of participation, reported honesty, and preference for assigned vs. alternate method of participation. For certain items, response options that were categories or a Likert scale were collapsed for ease of presentation and comprehension. The following is an example of collapsing response categories. A safety question is asked regarding whether or not conflict in their relationship (a) increased, (b) stayed the same, or (c) decreased as a result of study participation. “Stayed the same” and “decreased conflict” were collapsed so that “increased conflict” became the reference group. An example of collapsing Likert scale response options is in relation to asking telephone data collection and paper diary participants if they felt that they had to rush through completing the daily surveys. Original response options ranged from 1 “not at all” to 5 “very much.” Responses were collapsed so that 1 or 2 came to describe “rarely or never rushed through daily survey.” As indicated by the type of data, ANOVAs or chi square tests of independence were conducted to determine if groups (i.e., methods/conditions) differed across multiple domains. Group differences could only be analyzed for data which was comparable across groups (i.e., where question wording was comparable/equivalent).

Feasibility of Implementing ESM Methods

The initial six months of the project were devoted to consulting with colleagues and informatics specialists experienced with conducting telephone data collection before developing the daily protocol for any of the conditions. The telephone data collection method had to drive the development of the daily protocol for all conditions because (a) the protocol was standardized across conditions, (b) the telephone data collection system limits the number of and way in which questions can be asked, and (c) there is a limit to the amount of time participants are willing to spend on the phone to complete assessments before participation is experienced as burdensome and leads to non-compliance and attrition. Thus far, 24 months have been devoted to recruitment (enrollment and randomization is ongoing).

The university’s institutional review board did not immediately approve the study’s protocol due to concern that paper diaries pre-printed with questions on IPV and substance use might pose risk to women’s safety if the diaries were read by their partners. Therefore, a mini safety study was conducted by piloting the paper diary condition with 10 women. From the pilot study we learned that the majority of women were not at all concerned about their partners finding the surveys, and those who were concerned were confident that they could conceal the diaries. Based on these findings, institutional review board approval was granted.

RESULTS

Sample Characteristics

The average age of participants was 47.19 years (SD = 7.88). The majority of women were unemployed (72%), with a mean level of education of 11.90 years (SD = 1.25), a mean annual household income of $14,043 (SD = $13,296), and two children. The mean number of years in the relationship was 6.73 (SD = 6.24) with more than half of the sample married or cohabitating (58%). Eighty percent were African American, 8% were White, 7% were Latina, 2% were American Indians/Alaskan Natives, 2% were identified as multiracial, 1% identified as “other.”

By nature of inclusion criteria, all women experienced at least one act of physical victimization in the past 30 days. During the three months prior to study entry, 69% experienced severe physical victimization; 58% sustained minor injuries as a result of IPV and 32% sustained severe injuries. Forty-two percent experienced no sexual victimization, 15% experienced sexual coercion or attempted rape, and 40% experienced rape by their partners. Regarding substance use, alcohol, marijuana, and cocaine use were the most prevalent with 86%, 48%, and 33% of women, respectively, reporting use in the three months prior to study entry. Eighty-six percent of the sample smoked cigarettes in the three months prior to study entry with 73% of smokers doing so daily.

Attrition was low across the three conditions. Of the 49 telephone data collection participants who completed baseline interviews, 47 completed the study. Of the 40 paper diary participants who completed baseline interviews, 36 completed the study. Of the 41 monthly retrospective condition participants who completed baseline interviews, 40 completed the study. The overall attrition rate was 5.3%. Chi square analysis indicates non-significant differences between groups in regard to attrition, χ2 (2, n = 131) = 2.59, p < .10.

Safety of Participants

Information about participants’ safety, ease of participation, reported honesty, and preference for method of participation is noted in Table 1. Data show that it was safe for women to participate. Of note, nearly 100% of participants reported that their physical safety was never put at risk as a result of participating – regardless of the method of participation. One participant reported that her safety was “somewhat” put at risk but stated that there was no violence as a result and that her “partner became suspicious that I was talking to someone else on the phone.” Furthermore, depending on condition, between 66% and 73% of participants reported that their partners knew they were participating in the study and of those, 62% to 77% knew the focus of the study was on conflict in their relationships. Finally, of those partners who knew the study was about conflict, approximately 78 to 91% had neutral or positive reactions (e.g., he did not react, was not angry, or thought it would help the relationship).

Table 1.

Safety, Ease, and Feasibility Information Obtained During Follow up Interview to 90 Day Participation

Telephone
data
collection		 Paper
diaries		 Monthly
retrospective
interviews		 Between
groups
differencesa
(n = 47)
% (n)		 (n = 36)
% (n)		 (n = 40)
% (n)
Participants Perceived Safety
Participant’s physical safety was never put at risk as a result of study participation 98% (46) 100% (36) 100% (40) ns
Partner knew that participant was participating in the study and of those, he knew the study was about conflict in the relationship 66% (31) 72% (26) 73% (29) ns
77% (24) 73% (19) 62% (18) ns
Participation in study led to increased conflict in relationship 0% (0) 2.8% (1) 2.5% (1) ns
Specifically, partner knew that participant was calling in/ coming to study appointmentsb 60% (28) -- 83% (33) **
    Partner had a neutral or positive reaction when he learned of her study participation (e.g., he did not react, was not angry, or thought it would help the relationship) 75% (21) -- 91% (30) ns
Participant not all concerned that partner would notice her calling in to study daily 75% (35) -- --
Not at all hard to find a place to keep the daily surveys -- 89% (32) --
Participant did not have to hide blank daily survey from partner -- 83% (30) --
Partner saw blank daily survey -- 47% (17) --
    Partner had a neutral or positive reaction to seeing the blank surveys(e.g., he did not react, was not angry, or thought it would help the relationship) -- 88% (15) --
Participant did not have to hide daily surveys after completed with information about IPV and substance use -- 78% (28) --
Partner saw daily survey with completed information -- 39% (14) --
    Partner had a neutral or positive reaction to seeing daily survey that was complete with information on IPV and substance use (e.g., he did not react, was not angry, or thought it would help the relationship) -- 100% (14) --
Not at all important to mail daily surveys without anyone (including partner) knowing -- 64% (23) --
Not at all concerned that partner would learn about study appointments -- -- 90% (36)
Participants Reported Ease of Participation/Honesty, Preference for Assigned vs. Alternate Method of Participation
On a scale of 1–5, it was very easy (5) to participate 70% (33) 81% (29) 55% (22) ns
It was not at all difficult to share experiences openly and honestly 62% (29) 72% (26) 83% (33) ns
The 90 day duration of the study was "just right" 81% (38) 78% (28) -- ns
Training prepared participant very well to participate daily 79% (37) 92% (33) -- ns
Rarely or never rushed through the daily survey or did not report things that happened to complete the survey more quickly 75% (35) 86% (31) -- ns
It was not at all hard to remember to call in/complete the paper diaries 26% (12) 64% (23) -- ***
Instructions and questions on the phone survey were easy to understand 92% (43) -- --
It took the right amount of time to answer the questions over the phone 77% (36) -- --
Not hard to find a phone from where to call 87% (41) -- --
Questions on daily survey were easy to understand -- 92% (33) --
It did not take long to complete the daily survey -- 95% (34)
It was easy to find transportation to the interviews -- -- 78% (31)
It was not hard to find childcare for the interviews -- -- 92% (36)
I did not have to miss any work to participate in the interviews -- -- 100% (40)
Preference for assigned (vs. alternate) method of participation 57%(27) 67%(24) 58%(23) ns
Open in a new tab

Note. All percentages reported are valid percents.

ns – non-significant differences between groups

**

p < .01;

***

p < .001

a

Significance information is noted in this column only if between groups differences were tested (i.e., no information is noted for items that were asked/apply to only one group/method).

b

This general question was not asked of paper diary participants. Rather, paper diary participants were asked about partners’ knowledge of their participation with greater specificity. See questions below (e.g., “partner saw blank diary survey”, “partner saw daily survey with completed information”).

Feasibility of ESM Methods for Data Collection

Data in Table 1 also show that women found it easy to participate. Sixty-five to seventy-eight percent of participants reported that it was very easy to participate, and the large majority (78% to 81%) reported that the 90 day duration of the study was “just right”. Of the participants in the telephone data collection and paper diary conditions, 79% and 89%, respectively, said that the condition-specific training provided during the first interview prepared them well to participate on a daily basis. Regarding completion of the telephone survey or paper diaries, the majority of women (75% and 86%, respectively) reported that they rarely or never had to rush through the survey to complete it more quickly. Regarding participation in the monthly retrospective interviews, the majority of women reported that it was easy to find transportation to the interviews (78%; the research facility is accessible by public transportation), it did not take too long to complete the interviews (70%), it was not hard to find childcare (92%), and they did not have to miss work to participate (100%). The following summarized participants’ preference for assigned vs. alternate method of participation. Of paper diary participants, 67% preferred their assigned method, 28% would prefer the telephone data collection method, and 5% would prefer the monthly retrospective method. Of the telephone data collection participants, 57% preferred their assigned method, 34% would prefer the daily paper diary method and 8% would prefer the monthly retrospective method. Of the monthly retrospective interview participants, 57% preferred their assigned method, 25% would prefer the daily paper diary method, and 17% would prefer the telephone data collection method.

As indicated by the type of data, ANOVAs or chi square tests of independence were conducted to determine if groups (i.e., methods/conditions) differed on the characteristics reported in Table 1. Group differences could only be analyzed for items which were equivalent across groups; the final column of Table 1 notes significance/non-significance only for those items compared. There were no significant between-method differences in the characteristics in Table 1, with the exception of one safety item and one ease of participation item. A significantly greater percentage of participants in the monthly retrospective interview condition (83%) vs. the telephone data collection condition (60%) reported that their partners knew of their participation in the study (χ2 (df = 1, n = 87) = 5.42, p < .01), and a significantly greater percentage of participants in the paper diary condition (64%) vs. the telephone data collection condition (26%) reported that it was not at all difficult to remember to participate (i.e., complete the paper diaries or call in, respectively (χ2 (df = 1, n = 83) = 12.29, p <.001)].

DISCUSSION

Findings regarding the safety and feasibility of conducting ESM studies with women who currently are experiencing IPV are promising. Most importantly, data show that ESM can be applied safely. Nearly 100% of women reported that their physical safety was never at risk as a result of participating in the study. Of course a number of procedures were implemented to ensure participant safety; however, this finding is surprising given the large number of participants whose partners were aware they were participating in a study about the conflict in their relationships. Further, that many of the women’s partners reacted neutrally or positively when learning about the focus of the study on conflict is surprising – especially given the prevalence of severe victimization, injury, and sexual coercion/rape in this sample. Participants’ partners who exhibited positive reactions to their involvement in the study seemed to feel that the study could help to alleviate distress in their relationship. The latter findings suggest the need to take a closer look at the dynamics of IPV between women and men to gain a better understanding of the dynamics of the relationship in general.

Regarding ease of participation, results clearly show that women who currently experience IPV are able and willing to participate in an intensive ESM study about their victimization (and substance use) and further, report doing so easily. Promising findings in this domain include: (a) ease of recruiting targeted sample and retention throughout the study, (b) the high frequency with which women endorsed the 90-day duration of participation as “just right” (rather than “too long” or “too short”), and (c) the success of training to prepare women to participate daily via telephone survey or paper diary. Further, regardless of condition, women reported that it was easy to participate including sharing their experiences openly and honestly, understanding survey questions, finding transportation to interviews, and finding childcare.

With only two exceptions, there were no differences between the three conditions regarding safety and feasibility. Though there is no empirical evidence to support hypotheses regarding differences, these findings were unexpected. Depending on the feasibility domain under investigation, it would have been reasonable to assume that safety, for example, would have differed by condition. The first difference between groups is regarding women’s partners’ knowledge of their participation in the study. Compared to women in the monthly retrospective interview condition, fewer women in the telephone data collection condition reported that their partners knew that they were participating in the study. Women in the telephone data collection condition may have put forth greater effort to conceal participation from their partners than those in the monthly retrospective condition (i.e., to make calls, women may have waited until they were alone, went to a separate room, used a cell phone or phone of a friend or neighbor); therefore, partners may have had less awareness of participation. The second difference between groups is regarding women’s ability to remember to participate on a daily basis. It was more difficult for women in the telephone data collection condition than in the paper diary condition to remember to complete the survey daily. Perhaps having the paper diary forms at home as a visual reminder was helpful to remind participants to complete the survey.

Lessons learned about conducting ESM with women who currently are experiencing IPV

The following are lessons learned about conducting ESM with women who currently are experiencing IPV based on the authors’ experiences applying the ESM approaches highlighted here. These lessons are shared to inform the development of future research of this nature so that other investigators’ challenges can be avoided or made less difficult to overcome. First, there were challenges to obtaining IRB approval and these challenges delayed commencement of recruitment. Second, it was important to consult with colleagues and informatics specialists experienced with conducting telephone data collection before developing the daily protocol for any condition. Third, because participant-specific forms are required for participation in the paper diary and monthly retrospective conditions a substantial amount of time is needed from research assistants (equivalent to one full-time research assistant through out the study) to prepare and process those forms and the resulting data. This challenge is mentioned so that researchers budget accordingly lest complete these forms themselves. Fourth, anticipate life events/changes that may occur during the study among participants and plan the survey questions to capture this information as it may affect relationships under study or attrition; for example, incarceration or admission to in-patient treatment during the study affects the occurrence of IPV and substance use. Fifth, to reinforce participation, it is important to allow remuneration as frequently as possible for participants in the daily data collection conditions; one suggestion for future studies is to offer a graduated remuneration scale which might help to reduce attrition. Sixth, future studies should further investigate participants’ negative responses to, dissatisfaction with, and barriers to participation.

Limitations

Study limitations are worthy of note. Different operationalizations of ESM exist in addition to the one used in this study. For example, one operationalizations/approach expects participants to report behaviors and experiences as they are occurring. For obvious safety and related reasons, ESM operationalized in this manner could not be applied to the phenomenon of IPV victimization. However, future studies of IPV among community women could capture data/experiences more frequently than was done in this study (e.g., multiple times a day – however, not as IPV is occurring) to even further reduce difficulties with recall. Although participants experienced a range of severity and frequency of IPV and related injury, it is important to note that the most severely victimized women likely are not represented in this sample given the independence and freedom that was required of women to participate. Finally, it is possible that the amount of remuneration for which participants were eligible (up to $270 for women in the telephone data collection or paper diary conditions and up to $180 for women in the monthly retrospective interview condition) influenced results whereby partners were more accepting of women’s participation because of the financial gain. However, none of the women reported comments to this end as a partner’s reaction to gaining knowledge of her study participation.

Conclusions

Results suggest that the application of ESM should be considered more broadly in the field of research on violence against women. Preliminary evidence demonstrates that it is both safe and feasible for women to participate daily – even though they currently are in relationships where they experience IPV. Therefore, it is likely that women who have experienced victimization but are not in relationships in which IPV currently is occurring can also participate safely and effectively. In summary, ESM seems to hold promise as a method for collecting data among women who experience violence that can elucidate the daily dynamics of victimization and a number of its associated behaviors and experiences. The application of this method promises to advance research on violence against women to a place where daily dynamics and the phenomenology of victimization will be better understood and therefore, can better inform prevention efforts and intervention.

Acknowledgments

The research described here was supported, in part, by grants from the National Institute on Drug Abuse (K23 DA019561 and R25 DA020515).

Biographies

Tami P. Sullivan, Ph.D., is an Assistant Professor in the Division of Prevention and Community Research and Director, Family Violence Research and Programs, Department of Psychiatry, Yale University School of Medicine. Dr. Sullivan’s program of research focuses on understanding (a) precursors, correlates, and outcomes of women’s victimization and their use of aggression in intimate relationships, and (b) the co-occurrence of IPV, posttraumatic stress, and substance use with specific attention to daily processes and intensive longitudinal data. She is particularly interested in risk and protective factor research that informs the development of interventions to be implemented in community settings.

Enna Khondkaryan earned a bachelors degree in psychology with a minor in human rights from the University of Connecticut. Miss. Khondkaryan is a research assistant in Dr. Sullivan’s program of research on intimate partner violence, posttraumatic stress, and substance abuse on a National Institute of Health/National Institute on Drug Abuse funded project examining the temporal relationship between intimate partner violence and substance abuse. Enna will begin graduate school at Columbia University in New York, School of Social Work in the fall of 2010.

Nancy P. dos Santos, M.A. is a research assistant in Dr. Sullivan’s program of research on intimate partner violence, posttraumatic stress, and substance abuse on a National Institute of Health/National Institute on Drug Abuse funded project examining the temporal relationship between intimate partner violence and substance abuse. Miss dos Santos is a Psy.D. candidate from the University of Hartford. Her research interests include intimate partner violence, as well as its effects on female victims and their children. Her clinical interests include personality disorders, severe mental illness, and substance abuse.

Erica N. Peters, B.A. is a predoctoral psychology fellow in the Division of Substance Abuse at the Yale University School of Medicine and a Ph.D. candidate in Clinical Psychology at the University of Vermont. Ms. Peters’ research focuses on behavioral and pharmacological approaches to smoking cessation and marijuana cessation. Her clinical interests focus on behavioral treatments for individuals with substance use disorders.

REFERENCES

  1. Black MC, Kresnow M-J, Simon TR, Arias I, Shelley G. Telephone survey respondents' reactions to questions regarding interpersonal violence. Violence and Victims. 2006;21:445–459. [PubMed] [Google Scholar]
  2. Campbell R, Adams AE, Wasco SM, Ahrens CE, Sefl T. "What has it been like for you to talk with me today?": The impact of participating in interview research on rape survivors. Violence Against Women. 2010;16(1):60–83. doi: 10.1177/1077801209353576. [DOI] [PubMed] [Google Scholar]
  3. Carlson EB, Newman E, Daniels JW, Armstrong J, Roth D, Loewenstein R. Distress in response to and perceived usefulness of trauma research interviews. Journal of Trauma & Dissociation. 2003;4:131–142. [Google Scholar]
  4. Clements K, Holtzworth-Munroe A. Participation in laboratory research result in emotional distress with one's partner: Comparing violent and nonviolent couples. Journal of Family Violence. 2009;24(5):283–295. [10.1007/s10896-009-9229-4] [Google Scholar]
  5. Corkrey R, Parkinson L. A comparison of four computer-based telephone interviewing methods: Getting answers to sensitive questions. Behavior Research Methods, Instruments, & Computers. 2002;34(3):354–363. doi: 10.3758/bf03195463. [DOI] [PubMed] [Google Scholar]
  6. Csikszentmihalyi M, Larson R. Validity and reliability of the experience-sampling method. Journal of Nervous & Mental Disease. 1987;175(9):526–536. doi: 10.1097/00005053-198709000-00004. [DOI] [PubMed] [Google Scholar]
  7. Garcia-Moreno C. Dilemmas and opportunities for an appropriate health-service response to violence against women. Lancet. 2002;359(9316):1509. doi: 10.1016/S0140-6736(02)08417-9. [Article] [DOI] [PubMed] [Google Scholar]
  8. Griffin MG, Resick PA, Waldrop AE, Mechanic MB. Participation in trauma research: Is there evidence of harm? Journal of Traumatic Stress. 2003;16(3):221–227. doi: 10.1023/A:1023735821900. [DOI] [PubMed] [Google Scholar]
  9. Hektner JM, Schmidt JA, Csikszentmihalyi M. Experience sampling method: Measuring the quality of everyday life. Thousand Oaks: Sage Publications, Inc; 2007. [Google Scholar]
  10. Khantzian EJ. The self-medication hypothesis of substance use disorders: a reconsideration and recent applications. Harvard Review of Psychiatry. 1997;4(5):231–244. doi: 10.3109/10673229709030550. [DOI] [PubMed] [Google Scholar]
  11. Khantzian EJ. The self-medication hypothesis revisited: The dually diagnosed patient. Primary Psychiatry. 2003;10(9):47–48. 53–54. [Google Scholar]
  12. Langford DR. Developing a safety protocol in qualitative research involving battered women. Qualitative Health Research. 2000;10(1):133–142. doi: 10.1177/104973200129118192. [DOI] [PubMed] [Google Scholar]
  13. Newman E, Kaloupek DG. The risks and benefits of participating in trauma-focused research studies. Journal of Traumatic Stress. 2004;17(5):383–394. doi: 10.1023/B:JOTS.0000048951.02568.3a. [DOI] [PubMed] [Google Scholar]
  14. Newman E, Walker EA, Gefland A. Assessing the ethical costs and benefits of trauma-focused research. General Hospital Psychiatry. 1999;21(3):187–196. doi: 10.1016/s0163-8343(99)00011-0. [DOI] [PubMed] [Google Scholar]
  15. Robinson WS. Ecological correlations and the behavior of individuals. American Sociological Review. 1950;15:351–357. [Google Scholar]
  16. Rosenbaum A, Langhinrichsen-Rohling J. Meta-research on violence and victims: The impact of data collection methods on findings and participants. Violence and Victims. 2006;21:404–409. [PubMed] [Google Scholar]
  17. Rosenbaum A, Rabenhorst MM, Reddy MK, Fleming MT, Howells NL. A comparison of methods for collecting self-report data on sensitive topics. Paper presented at the 8th International Family Violence Research Conference; Portsmouth, New Hampshire. 2003. [PubMed] [Google Scholar]
  18. Searles JS, Helzer JE, Walter DE. Comparison of drinking patterns measured by daily reports and timeline follow back. Psychology of Addictive Behaviors. 2000;14(3):277–286. doi: 10.1037//0893-164x.14.3.277. [DOI] [PubMed] [Google Scholar]
  19. Searles JS, Perrine MW, Mundt JC, Helzer JE. Self-report of drinking using touch-tone telephone: Extending the limits of reliable daily contact. Journal of Studies on Alcohol. 1995;56(4):375–382. doi: 10.15288/jsa.1995.56.375. [DOI] [PubMed] [Google Scholar]
  20. Sobell LC, Sobell MB. Alcohol Timeline Followback Users' Manual. Toronto, Canada: Addiction Research Foundation; 1995. [Google Scholar]
  21. Stone AA, Shiffman S. Capturing momentary, self-report data: A proposal for reporting guidelines. Annals of Behavioral Medicine. 2002;24(3):236–243. doi: 10.1207/S15324796ABM2403_09. [DOI] [PubMed] [Google Scholar]
  22. Stone AA, Shiffman S, Alienza AA, Nebeling L, editors. The science of real-time data capture: Self-reports in health research. New York: Oxford University Press; 2007. [Google Scholar]
  23. Straus MA, Hamby SL, Boney-McCoy S, Sugarman DB. The revised Conflict Tactics Scales (CTS2): Development and preliminary psychometric data. Journal of Family Issues. 1996;17(3):283–316. [Google Scholar]
  24. Todd M, Armeli S, Tennen H, Carney MA, Affleck G. A daily diary validity test of drinking to cope measures. Psychology of Addictive Behaviors. 2003;17(4):303–311. doi: 10.1037/0893-164X.17.4.303. [DOI] [PubMed] [Google Scholar]
  25. U.S. Census Bureau. Census 2000 Summary File 1/ Connecticut. 2001. [Google Scholar]
  26. Walker EA, Newman E, Koss MP, Bernstein D. Does the study of victimization revictimize the victims? General Hospital Psychiatry. 1997;19(6):403–410. doi: 10.1016/s0163-8343(97)00061-3. [DOI] [PubMed] [Google Scholar]

RESOURCES