• ## Summary

Search form
• 00:00

[MUSIC PLAYING]Hello, my name is Jane Palmer.I am a faculty in the School of Public Affairs at AmericanUniversity in Washington DC.I teach classes on research methods and social policy

• 00:22

issues.What I research is gender-based violence,issues related to victimization, perpetration,help seeking for victims.And I'm also really interested in research relatedto methodological issues when doing researchin the area of gender-based violence.I'm actually a former social worker,so I worked for about a decade in Chicago and Saint Louis

• 00:43

with children and families.And a lot of my work in those experiencesinforms the research and the teaching that I now do.So today, I'll be talking about conducting surveyson sensitive topics, such as sexual assaultor other forms of victimization on college campuses.I'm going to talk a little bit about whywe did this type of survey, how wedid the survey, some challenges that we faced when conducting

• 01:05

the survey, and the lessons that welearned in the experience of conducting the survey.So in about 2010, I was talking to some colleagues of minewho are sexual assault advocates on college campuses.And they had expressed to me frustration

• 01:25

that they didn't have data about the studentsthat they were working with.I mean, they knew the students that they saw personally,but they didn't have data about what the extent of the problemwas at their specific campus.There is some national data.There was a study conducted in academic year 1996-1997 thatfound that between one and four and one in five college women

• 01:47

will experience sexual assault while they're a collegestudent.But at this point, that data is getting pretty old,and it's hard to rely upon since it's a national estimate.But the only other source of data that collegeshave is clear react data.But this data is really limited to sexual assaults thatwere reported, and they have to have beenreported to certain people.

• 02:08

It doesn't cover all sexual assaults,so it doesn't cover a majority of sexual assaultsthat happen off campus.So I threw out the idea that maybe weshould do a survey of students.Maybe we should figure out how to do this.I was a second year doctoral studentnot really knowing what I was getting myself into.But I read a lot about it, and I reached outto some important people in the field to try

• 02:28

and figure out how to do this right and well at the collegecampuses where these advocates worked.Surveys are really important, because youcan capture instances of victimization thataren't reported to the police or a counseling centeror a sexual assault advocate.And there's been some recent national attentionin this area in that the Obama Administration has released

• 02:51

a report encouraging campuses to conduct these types of surveys.And even at the state level, Maryland recentlypassed a law requiring all public and private institutionsto conduct these types of surveys.Next, I had to figure out how to do the survey nowthat I have this great idea to try and help

• 03:12

the advocates in this way.So I contacted a researcher named Dr. Bonnie Fisher.She's at the University Cincinnati.She actually was the lead principal investigatorfor the study that I mentioned that was conductedin the mid-90's.And I talked to her, and I told her about my idea,and I asked if she had a survey that maybe she'dbe willing to share, that I could adapt or use

• 03:33

at these college campuses.As luck would have it, she had a recent surveythat she and some colleagues at some other universitieshad developed.And they agreed to share the everybody with me.The survey covered more than I could imagine.It covered sexual victimization, which of course wewere concerned about.It also covered sexual harassment, cyberharassment,

• 03:53

dating violence, and then a bunch of questionsaround bystander intervention, whichis an important aspect of this in figuring outhow people could intervene or whether people are interveningto prevent these types of victimizations.And then it also asked important questionsabout the university's response to sexual assault.So it was important to me that weuse an already existing survey.

• 04:14

I felt under prepared to create a survey myself.So I was very pleased that we are able to utilize the survey.By the time that all of these details worked out,it was early January 2011.And we wanted to conduct a survey by a lateMarch 2011, before Sexual Assault Awareness

• 04:35

Month in April.We didn't want that to affect howpeople responded to the survey.So I had a short time frame and a lotto do in that frame to be able to conduct these surveys.So we decided to replicate the surveythat was being done at these three other universitiesby Bonnie Fisher and Coker, Suzanne Swan,these other researchers that I had been in touch with.

• 04:56

And so we were, by replicating their study,utilizing a method that they usedcalled the Dillman Tailored DesignMethod that has very specific waysthat you conduct the survey.You have a questionnaire that's really user friendly.You provide an incentive in advanceof taking the survey to all people,whether they take the survey or not.

• 05:17

You personalize all content, and you contact the respondentsmultiple times during the survey period.We also had decided that we were goingto be doing, just like they had, a stratified random sample.So this was important to me, because random samplinghadn't been used very much in victimization or bystanderintervention surveys on college campuses.

• 05:38

Other studies have used more convenient samples.So all first-year students at a new studentorientation, or all members of fraternities,or that sort of thing.And those are really important groupsof people to serve on these issues,but it is important to me to really understandthe scope of the issue at the collegeto do a random sample of all students.And we stratified by class year so

• 05:60

that we invited equal numbers of studentsfrom first year class, the sophomore class,the junior class, and the senior class to participate.Another important part of this design, as I mentioned,is to provide an incentive.This really helps your response rate.It really helps me and get additional peopleto participate that wouldn't participate otherwise.

• 06:20

There's some people that would immediatelyrespond to a survey on sexual assault preventionand want to participate.And then there's others that an incentive would help themto also agree to participate.This design required a $2 cash incentive.And I will note that that has to be two single dollar bills.Some researchers did some studiesand found out that some college students don't actually • 06:43 recognize a$2 bill as US currency.So it was important that we use two single dollar bills.It was also really important that we protectedthe confidentiality of the studentsby going through Institutional Review Board Approvaland making sure we de-identified the data, after we collectedthe data, so that no one could connect a student's responses

• 07:04

to their actual identity.And then, because we weren't doing these surveys in person--so I wasn't sitting there interviewinga student about their experiences,we did a web-based survey-- it was importantfor us to provide community and national resources to them,in case they had experienced some unwanted sexual activitiesor other types of victimization, or were

• 07:26

struggling because of academic issues or other issuesin their lives.We provided a list of local and communityand national resources for them if theydid need to reach out and get some help.And that was provided in our initial letterwhen we invited them to take the survey,and then also at the end of the survey,in case they experienced any distress during the survey.

• 07:50

So as I mentioned, it was really importantthat-- any survey that you're goingto do with human subjects, you haveto go through Institutional Review Board Approval.And so, as I mentioned, we submitted this in early Januaryfor a March survey date.And sometimes it's hard to tell how long that approval willtake.There's a board of individuals thatwill review your application.

• 08:11

At the first institution, we received approvalwithin a month, which was great, because we stillhad plenty of time to get everything together.At that point, we didn't have the final survey instrument,so the researchers I was working withhad pilot tested the instruments in Spring, 2010.And they were still analyzing the dataand making some changes to the instrumentfor use in Spring, 2011.So I was waiting for that final instrument.

• 08:32

But that first institution was OK with my submitting a draftinstrument and just as long as theyreviewed the final instrument priorto doing the actual survey.However, at the second institutionthat I was working with, they would notaccept a draft institution.So IRBs are run at each institution a littledifferently.And so we ran into an obstacle in that we couldn't even

• 08:53

submit the IRB until the final instrument was ready.It ended up being that we weren'table to proceed at that second institution that year,because we didn't get the final instrument in time.But I was able to do that survey at that second institutionthe following year in Spring, 2012.So we were able to learn from what I learned in the 2011

• 09:14

survey and apply those lessons learnedto conducting it at the second institution in 2012.So back at the first institution,once we had IRB approval, I couldstart talking to administrators aboutdifferent logistical details that needed to be worked out.It turned out there were several other surveys that were beingdone that spring with students.

• 09:36

And rightly so, the administratorswere worried about something called survey fatigue.So if you're asked to take surveys too often,it may result in fatigue that may make it so that peoplewon't respond to the survey.And I didn't want surveys going on simultaneously.So I was given a window of time, from March 29 to April 10,to do this survey, because it was

• 09:57

the only time available that other surveys weren'tbeing fielded.And so this was 10 days into Sexual Assault Awareness Month,which I was trying to avoid.But I ended up deciding to go with it,because it was really important to usthat we did the survey that spring.So this meant that I had to get allthe other logistical details together by Friday, March 25

• 10:20

so that we could spend the weekend on putting togetherthe mailing, mail it out on Mondaymorning, and start the survey and on March 29.So I had a few challenges in order to meet that deadline.I ended up acquiring the survey instrument on Monday, March 21.And then I could send it to the IRB to get that final approval.Because I couldn't move to the next stepof acquiring the random sample until I

• 10:42

had that final IRB Approval.So that survey came just in time for meto get in touch with the Registrar'sOffice about the random sample, once I have formal IRBapproval.So we put that request in right away.IRB actually got back to us in 20 minutesto formally approve the study, which was such a relief to me.It was unknown, at that point, how long it would take.

• 11:02

And then I was able to get the requestinto the registrar for the stratified random sample.The next issue that I guess I hadn't totallythought through how challenging it would bewas acquiring the $2,200 in cash.So our random sample is going to be1,100 students, which meant we needed 2,200 single dollarbills.I had a colleague that was working with me • 11:23 to help acquire the cash.And we ended up having to go to three banks on oneday to get as many single dollar bills as we needed.In future years, we learned to puta request in about a week in advanceto make sure that those dollar bills will be ready the daywe needed them.But that was quite a fun road trip, going from bank to bankto try to acquire that those single dollar bills. • 11:43 And then we spend the weekend, as I mentioned,of March 26 and March 27 putting together the mailing, just twoof us locked in an office, streaming movies,and eating take-out, just getting that mailing togetherfor 1,100 people that we mailed out first thing Monday morning.So we learned several lessons in this process. • 12:06 I learned that this method works.We had a response rate of 56%, whichis pretty incredible for this type of survey.Surveys they don't use these methodstend to have a 10% to 20% response rate,so we were really pleased with that response rate.And I learned that it was really important throughoutthe process to do your homework, and justbe prepared for anything that might come your way. • 12:27 I was a second year doctoral student, so I knew some stuff,but I didn't know a lot.But I'd read a lot of literature about survey methodology,and doing victimization surveys right, and incentivesand the importance of incentives.So when someone would challenge me and say, you know what?That$2 thing, it sounds really great,

• 12:48

but it's really going to be too logisticallychallenging, I could say, this is why this is important.So I recommend, if you're considering somethinglike this, to learn that literatureand be able to tell campus administrators, or advocates,or students, the importance of why you're doing things the wayyou are.I also learned that it was reallyimportant to build rapport and maintain strong relationships.

• 13:08

I counted about 20 people that help me through this process.It wouldn't have been possible without each person involved.And I knew about half of them prior to the process,and I got to know the other half.And I have still maintained relationshipswith all of those people throughout.And those strong relationships helped sort of grease

• 13:28

the wheels and make sure that things happenedand that everything got done in time.My motto is to be patiently persistent,so not to irritate people but to makesure that they don't forget that youhave a request for something that they need.And then, I also learned to proceeddespite disappointments.I was really looking forward to doing the study of both sites

• 13:48

at the same time to have a comparisonand to help those advocates out, as I mentioned.And I just had to know when to fold and when to not go forwardwith that second institution, eventhough I was able to go for the following year.And then, finally, I would say have a sense of humorand stay positive.It was very challenging during this two-

• 14:09

or three-month period.I was working 20 hours a week.I was taking three doctoral-level courses,and I was trying to get the survey off the ground.And there are some long days and sleepless nights,but I was able to have a sense of humor about itand just stay positive throughout the process.And I would recommend that you do the same.If you're thinking about doing something like this,

• 14:30

I would encourage you to reflect on a couple of things.What kind of sampling is important for your institution?Do you want to get a targeted group, like those involvedin Greek life or those involved in athletics,or do you want to do the more random sample of studentsin general?I would encourage you to think about the incentives.Some studies might offer a raffle for an iPad,

• 14:51

and that can also be really useful.And that might be more cost effective,depending on your budget and your sample size.So I was working with a smaller campus where1,100 students that we invited was an adequate samplefor that site.But I've worked with colleagues at very large institutions.And I'll tell you, if you have 40,000 people,you're not going to send them $2 each. • 15:12 So there might be budget constraints,and you'll have to consider what makes the mostsense for your setting.There's just a few things to consider when you're doingsurveys about sensitive topics.I'll just review a couple of them.One is around survey wording and making surethat the survey's been tested somewhere else, or even • 15:33 at your campus before you use it.There's been about three decades of researchon how to word these surveys right and well, howto word victimization questions sensitivelyin a way that really elicits the most accurate estimateof victimization.I mentioned that there's considerationsrelated to the selection of the sample with the incentives. • 15:54 And then, it's really important that confidentialityis first and foremost your priorityfor the protection of the human subjectsand the protection of your students and their experiences.And providing those community resourceswill really help the students seethat you are not only interested in hearingabout their experiences but also wantto make sure that they know where they could get help • 16:15 if they would like to seek help.I learned-- first and foremost, I'm very much a perfectionist.But I've learned as I've become a researcher that no study isperfect, and that's OK.There absolutely does not exist, a perfect study.There's always challenges, and each studyis a stepping stone to the next study.And so, since I did this study in 2011,I ended up to writing my dissertation on it. • 16:37 I've done a 2012 survey, 2013 survey, and a 2015 survey.And I've learned something each way.And each year it gets easier, and each year it gets better.So in conclusion, I'd like to present a question.If you're considering this type of survey for you're setting,your college setting, or even another setting that you'reconsidering doing a survey of this type,what methods make the most sense for you? • 16:59 Is it important for you to have a random sample,or does it make more sense to have a targeted sample,such as all incoming students or all students involvedin Greek life or athletics?What incentive structure will work for you?I was lucky enough to work at a small campus.If you're at a large university, would itmake more sense to have a raffle of an iPad or a TV, • 17:20 or that sort of thing.It may also help your response rateif it's cost-prohibitive to do the$2 per student.So there may be some tweaks that you'dneed to make for your setting.But think about what those might beand how you could proceed in a study like thisin your own setting.[MUSIC PLAYING]

### Video Info

Publisher: SAGE Publications Ltd.

Publication Year: 2017

Video Type:Video Case

### Segment Info

Segment Num.: 1

Persons Discussed:

Events Discussed:

Keywords:

## Abstract

Dr. Jane Palmer discusses the best ways to conduct survey research on sensitive topics, specifically victimization. As an example, Palmer presents a survey that she conducted on sexual assault on a college campus.