So this study came out recently – “Pornography Actresses: An Assessment of the Damaged Goods Hypothesis.” What to make of it..?
First and foremost – the study is published in the Journal of Sex Research, a peer-reviewed academic journal. The research team behind this work is made up of: James Griffith, Lea Adams, and Lucy Gu, all of Shippensburg University; Christian Hart of Texas Woman’s University; and Sharon Mitchell, former founder and director of the now defunct (thanks, AHF!!) Adult Industry Medical (AIM) Healthcare Foundation in Los Angeles.
This study is seated in psychology – James, Lea, and Christian are listed as psychology faculty at their respective universities (here and here), and Lucy appears to be a graduate student (maybe in Soc/Anthro, though?) at Shippensburg (here, see page 51). And Sharon Mitchell, though not a “doctor” or a “real PhD,” has extensive experience working in the adult industry, with the adult performer population, and with STIs in a prevention, mitigation, and monitoring sense. I would describe her expertise as a formal and informal amalgamation of many disciplines (psychology included), all situated around the adult community.
Hence, this is a real and true study. (so many of them aren’t)
What was found?
Basically: Compared to a matched group of women (age, ethnicity, and marital status), “porn actresses were more likely to identify as bisexual, first had sex at an earlier age, had more sexual partners, were more concerned about contracting a sexually transmitted disease (STD), and enjoyed sex more… In terms of psychological characteristics, porn actresses had higher levels of self-esteem, positive feelings, social support, sexual satisfaction, and spirituality compared to the matched group. Last, female performers were more likely to have ever used 10 different types of drugs compared to the comparison group.”
According to the researchers, these findings did NOT support the “damaged goods hypothesis,” which “posits that female performers in the adult entertainment industry have higher rates of childhood sexual abuse (CSA), psychological problems, and drug use compared to the typical woman.”
Put simply: according to the findings reported, women who work as porn performers are not “damaged goods.”
(this is always where the issues are)
Data for this study came from asking two groups of women – women porn performers (“pornography actresses”) and a matched comparison group of “typical wom[e]n” – a series of survey questions. Questions were grouped in five distinct subsections: basic demographics, sexual behaviors & attitudes, self-esteem, quality of life, and alcohol & drug use. After the data was gathered and (presumably) cleaned/normalized, each group’s respective responses were compared via a series of statistical analyses.
Let’s learn more about each respective group…
Performers: the group consisted of 177 participants, all qualified by having been paid to work in at least one “X-rated movie.” Participants had an average age of 26 years (range: 18 to 50) and an average of 3.5 years working as a performer (range: 1 month to 30 years). Most were single (44%) and “Caucasian” (63%), though a range of relationship statuses and an array of racial and ethnic groups participated in the study.
Women volunteer participants were gathered via convenience sampling over the course of four months at AIM – this means that the sample is made up entirely of women who “conveniently” walked through the door of their own accord and completed a survey. Participants who completed the survey had their names put in a lottery for one of two opportunities to win $300 worth of STI testing from AIM.
This section of the “Procedure” description makes me slightly uneasy [my comments embedded]:
Data collection from the pornography actresses occurred for four months [when exactly? during what year at least? though not telling this helps protect participants’ identities, it also withholds a hugely significant bit of contextualizing information]. Flyers detailing the study that mentioned the incentive were posted in the reception area of the waiting room of AIM. All clients that visited AIM were required to check in at the reception desk where the flyers were in full view. When clients checked in, staff at the reception area informed them of the study [everyone who came in was informed?], encouraged individuals to participate [yikes!!], and referred all interested parties to the chief medical officer [is this referring to Sharon Mitchell or to an actual physician on staff? this is a hugely significant question]. If interested, the chief medical officer provided further details of the study and gave the self-administered questionnaire [so, after consultation with the CMO, participants filled out the surveys alone..? what if they had additional questions?]. Participants were first provided with a consent form and ensured that there was no way to match their identity with their responses [if participants wrote on the pages with their own handwriting, this is not true. if participants interacted with someone they knew at any time during this process, this is extra not true]. The participants completed the questionnaire in the waiting room prior to receiving services. After completion of the questionnaire, participants gave the survey to the staff at the reception area [more identity issues].”
Things I wish I knew
I would’ve liked an average number of scenes worked at the time of data gathering, even an estimate. This impacts perspective – 10 scenes in 3.5 years is not the same as 10 scenes in 10 months or 10 scenes in 2 months. I would have also liked to have seen some cross tabs showing who was what (ie married and in adult for X years, etc).
Also, when were participants vetted? When was whether or not they were qualified to take the survey (women with at least one paid “X-rated” gig) determined? …because more ladies than just women with at least one scene under their belts tested at AIM. Was this asked at the front desk or something (because if it was already known or assumed, this is another significant identity issue)? And were qualified trans women performers allowed to participate?
Most importantly though, I wonder about identity and community… The notion that there was “no way to match their identity with their responses” is ludicrous. Maybe from the perspectives of the university-based researchers this was acceptable; but the notion that participants couldn’t be identified, in the small community that was AIM, is simply false.
And this is to say nothing of participant/AIM employee/CMO dynamics…
“Typical wom[e]n:” The comparison group was matched on age, marital status, and ethnicity. This means that the researchers sought out women until they found ones who filled the demographic requirements set by the performer group. Accordingly, the percentages across groups were (supposedly) identical to the above described demographic characteristics. To me, this claim seems impossible.
Further, “the comparison group was recruited using convenience sampling at university and community (ie airport) settings… There were 68 (38.4%) women recruited from university settings and 109 (61.6%) from an airport.”
Focusing strictly on the identity/identifiable issue described in the performer population, the procedure for gathering the “typical” sample is far less problematic – strangers in airports and at universities VS completed in the office of one central medical clinic. There is cause for concern with identity and culture here, specifically related to the marked differences between university culture, airport (like, representing the general world?) culture, and adult performer culture – each is a very distinct subcultural bubble. And though I don’t doubt the quotas were filled, I highly doubt they were filled in the exact same combinations. This seems like an impossible task, but maybe…
Things I wish I knew
People were just approached in public to take this survey? Was any incentive offered? Who explained the study to them, and was this done right at Gate Number Whatever? Huge methodological concerns…
The data was gathered and a bunch of stats were run. The study itself contains many interesting tables and charts. You can scroll through the actual text to see for yourself. I didn’t see any real problems with this – basic number crunching. (study text is here)
These data were used to examine the “damaged goods hypothesis,” which is the general public perception that “pornography actresses” 1) are victims of childhood sexual abuse (CSA), 2) are psychologically unhealthy, and 3) use drugs. The researchers determined that “pornography actresses” 1) suffered no more CSA than “typical” women, 2) were psychologically healthier, and 3) used more drugs more often. Researchers offered some speculation on each of these points – some were insightful, others seemed like they were reaching. (read it all here)
Problems Identified and Engaged by the Researchers
1. This study is not generalizable – though it gets at a lot of women working as porn performers, it also only gets at some women working as porn performers.
2. Who’s a “pornography actress”? Setting criteria to preclude participation (pornography actress = paid to have worked in at least one X-rated film… hardcore? softcore? non-sex??) had to be difficult given the current diversity present in adult content forms and performance. And though I do not agree with the criteria described in the text of the paper, some form of definition had to be set. The researchers acknowledge issues along these lines.
3. The third problem is worth quoting:
“…some of the measures were problematic. Some measures used for sexual behaviors and attitudes were single-item indicators with unknown validities and reliabilities [ie does ‘this’ really mean ‘that’ consistently to multiple people]. As an example, participants were asked if they were victims of CSA without further clarifications or definitions. Thus, it is quite possible that a given behavior in a particular situation may have been perceived as CSA by one individual but not by another [not providing clarification along these lines is a huge misstep]. Another issue with regard to measures is that education was not examined [doh!! amateur mistake]. Education has been found to be related to a variety of sex-related constructs and would have been a sound matching variable, and its inclusion should be strongly considered in future studies [especially when many of your ‘typical’ population were in college #justsayin]”
What’s my two cents?
Over the course of the past week, I have read several reactions to this study. Two extremes include “no duh/thank you for confirming porn performers are people, too” and “it’s super bad to emphasize all this porn star ‘good'” – hmmm…
In my view, the “no duh” part is right on. No kidding some women who work in adult are more adjusted in some respects relative to some “typical” women. This study never once suggests ALL of anyone about anything and goes so far as to out-and-out declare that its findings are NOT representative of ALL porn performers. This is good.
But people don’t seem to like to read further than the abstract these days; thus, pretty much every report on this study emphasized the “OMG porn stars – as in all of them – are so well adjusted, like, overall.” This is an overblown misread/misreport of what the study clearly states, which provided many haters with ample opportunity to get critical (a la “it’s super bad to emphasize all this porn star ‘good'”). To these folks I ask: isn’t reporting only the bad and none of the good just as problematic? (guess what – it is!!)
Misrepresenting data is extremely unethical, but the researchers’ reporting seems to have been done accurately on the basis of what was gathered. Unfortunately though, this was not want many people (haters) wanted to hear.
This study moves us toward a greater, more nuanced understanding of the adult industry by finally authenticating in a peer-reviewed study some commonsense knowledge we all already possess. Contrary to many studies and “studies” exploring things like “pornography actresses,” this work engages women working as adult performers and involved in the adult community at the time of data gathering. It acknowledges many of its limitations and attempts to connect participants with the wider social world. It also provides a platform by which other researchers and scholars can take the next step….
…which is good because this study has at least one major epistemological flaw – disconnect. Disconnect is a major problem with many difficult to access, contested, marginalized, and/or stigmatized populations, not just adult.
Though what we have on paper is a dream team of legit psychology researchers, a grad student to do all the grunt work, a high profile industry insider, access to participants from the target population, and some form of funding (I assume, though there is no mention in the paper how all this was paid for), what we get are big methodological flaws that stem from researchers not knowing enough about the adult industry… and the adult industry not knowing enough about research.
In spite of some flubs, the researchers working on this study clearly know how to conduct rigorous social science research. They made several mistakes and some of those mistakes point to poor research design, yes [ie not asking about education, claiming the participant populations were exactly matched, not having a clear survey instrument (definitions, etc), having someone of “authority” who is significant in the community aid with survey administration, etc]; but perhaps something happened behind the scenes that we are unaware of.
So some mistakes are clear flaws in research design, but others point to a lack of in-depth understanding of the population. The researchers clearly did not know the nature of AIM, they do not understand the close-knit nature of the adult community, they do not understand the “life course” of the average performers’ careers, they use strange language, etc. Perhaps some of the blatant methodological flaws could have been fixed had they been more familiar with the population and community…
…which is why I thought they had Sharon Mitchell. Not only could Sharon provide access, she could have let the researchers know when they were missing the mark. But that didn’t happen. I suspect this is because, though Sharon may be an industry expert, she is neither a scholar nor a researcher. Were she more research-minded, she might have suggested that having desk staff solicit participation or having a known clinic leader “help” with the surveys was a bad idea. She might have offered insight regarding common trends observed informally over her years working with performers (and as a performer herself) that could have helped develop a more nuanced analysis. Or maybe she did all of this and her insights were ignored? That’s possible too…
I would think that the goal was to come together and fill in the blanks, but I guess that didn’t happen.
Ultimately this study is a positive thing. It brings a “healthy” performer voice into a space where it’s rarely heard, and it had a great comparative framework; but the finer points of the study (including how the data was worked with and reported) should have been done very differently. Perhaps these methodological issues needed to be engaged in order for the next round to be that much better?
One can only hope.
FYI, here’s the full reference:
Griffith, James D., Sharon Mitchell, Christian L. Hart, Lea T. Adams, and Lucy L. Gu (2012). “Pornography Actresses: An Assessment of the Damaged Goods Hypothesis.” Journal of Sex Research – currently in early release (I think… as far as I can tell, for the moment, this study is only available through the publisher’s website. no issue or volume numbers have been assigned there)
Not sure how long publishers Taylor & Francis will continue to offer the full text for free, but the entire study is here… for now, at least!!
* * *
You may quote anything herein with the following attribution: “Reprinted from Porn Valley Vantage/PVVOnline, copyright © Chauntelle Anne Tibbals, PhD (www.PVVOnline.com).”