Here’s how Slate recently positioned yet another study on “Internet addiction:”
“Problematic Internet Use” Is Now Officially a Thing
The original title of the exact same article on The Conversation was little better:
There’s a new addiction on campus: Problematic Internet Use (PIU)
Why are media outlets continuously pushing problematic Internet use on an unsuspecting public?
Let’s tackle the larger Slate first. The articles are duplicates, discussing the same research by the author of the study (Synder et al., 2015). Obviously Slate appears to have a headline editor who just took the crux of the claims made by the researcher at their face value. With no mental health experience (or apparently, even a little checking), they just assume that if a researcher says, “Hey, X is now an official disorder,” it must be true.
Although that’s not how conditions or disorders are considered “official.” For something to be official, it has to be sanctioned by the bodies responsible for diagnostic manuals, such as the ICD-10. In the case of a mental disorder in the U.S., that would be the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (the DSM-5). Does the claimed “problematic Internet use” disorder exist in any form in the DSM-5? Nope. How about “Internet addiction”? Nope again.1
I know this and my editors know this since we live and breath this stuff every day. Folks at big media companies haven’t a clue, so they appear to go with the most outrageous, eye-catching headline they can pull off. But factually, this headline is simply a lie. Problematic Internet Use is no more an “official thing” today as it was when it was first proposed nearly two decades ago. And this isn’t the first time a mainstream media outlet made a claim about “Internet addiction” being officially declared real.2
So I might expect this level of journalism from Slate. But The Conversation?
The Conversation claims in its mission statement: “to provide you with a reliable source of high quality, evidence-based information.” Sounds good, right?
So how do we arrive at calling it a “new addiction” (for a proposed disorder that’s been around for nearly two decades)? Even the first line of the article screams, “Fact checking on Aisle 1!”
Problematic Internet Use is now considered to be a behavioral addiction with characteristics that are similar to substance use disorders.
Considered by whom? What is the source of this official declaration? Why are The Conversation’s editors simply taking a researcher’s opinion and printing it as an established fact?
In fact, the excellent review of research into the conceptualization of problematic Internet use (PIU) by Tokunaga & Rains (2010) suggests just the opposite:
The findings provide relatively little support for the model developed from the characterization of PIU as a form of pathology. The fit indices demonstrate that the pathology model does not adequately fit the data.[…]
PIU can be understood as an indicator of one’s inability to successfully regulate his or her Internet use. […] The deficient self-regulation perspective describes PIU as lapses in effective self-regulation and moves away from the idea of PIU as patterns of media consumption consistent with an addiction or disease.
It is not a disorder or addiction in any traditional sense, so calling it a “behavioral addiction” isn’t just sloppy word choice. It’s just plain wrong. (Tokunaga, 2015 is an even better read.)
The Conversation: Academic Rigor?
I reached out to The Conversation (“Academic rigor, journalistic flair”) and spoke with Editor Maria Balinksa via email to get a better understanding of their editorial process. She replied, in part: “Authors write in their area of expertise. Each article is reviewed by two editors on our end but our editing process does not include a peer review panel.”3
I’m confused how something that can be published without critical review fits in with their aim “to promote better understanding of current affairs and complex issues.” How can we understand the complexities of something like “Internet addiction” when the only articles published on the issue by the mainstream press are those that are glowing and confirm the researcher’s own hypotheses and self-interests?4
Which is a gaping blind spot in The Conversation’s charter:
Provide a fact-based and editorially independent forum, free of commercial or political bias.
What about self-interest bias? What about confirmation bias? What about publication bias? Researchers have an interest to promote their work widely, since their entire career is based upon dissemination of positive findings. It’s even more exasperating when you realize the current study had a tiny subject pool (N=27) composed only of college students!
While I respect The Conversation’s good intentions, this article demonstrates you can find fluff pieces on there just like anywhere else online. At least they’re trying… But it also seems like it’s offering the shiny sheen of “Academia-approved” rather than making any serious attempt to vet the quality of what they’re publishing.
The Status of Internet Addiction Today
There have literally been hundreds of studies published on “Internet addiction” or “problematic Internet use.” Most of them are, quite plainly, crap. They suffer from fatal flaws or constantly-changing definitions and rely on psychometric measures that are not very good.
While “Internet addiction” researchers continue to publish poor-quality research on the topic, some high-quality reviews of the research have been published. Two studies that are of particular value are Laconi et al. (2014) and Czincz & Hechanova (2009).
Laconi found an incredible 45 different measures of “Internet addiction” — and the majority of them have problems:
Concerns regarding the solidity of the theoretical bases of the different measures can be added to those regarding the general lack of rigorous evaluation of psychometric properties (Wartberg et al., 2013). Indeed, more than 26 scales have only one study supporting their psychometric properties.
The most commonly used assessment, the Internet Addiction Test, lacks “rigorous and systematic psychometric investigations.” It also has construct validity problems — a core component of a test’s psychometrics. “The results of factor analyses revealed significant differences between studies suggested a potential lack of construct validity of the IAT, in addition to somewhat low reliability.”
This isn’t exactly good stuff to hear for such a commonly-used measure. If your measure has problems, whatever data you get back from that measure aren’t going to be reliable.
Laconi did find a problematic Internet use measure that holds the most promise: the Generalized Problematic Internet Use Scale-2 (Caplan, 2010). “The scale demonstrated good psychometric properties including a consistent 4-factor model. The brief 15-item format makes it a useful tool, at least in the area of research.” Future researchers should take note.
Czincz’s summary of what their research found is why you can see the concept of problematic Internet use is still struggling, despite two decades’ worth of research.
The three main problems with the existing research on PIU are the challenges regarding the general conceptualization of PIU, the dearth of methodologically sound studies, and the lack of a widely accepted assessment measure with adequate psychometric properties. There continues to be a lack of consensus in the research regarding the definitional and diagnostic base for PIU, which has lead to inconsistencies across studies and posed challenges for the identification of optimal treatment options. […]
Most research on PIU to date is not methodologically sound due to difficulties with sampling and research design. The majority of studies involve self-identified convenience samples of problematic users or student samples, which significantly biases the results (Byun et al., 2009; Warden et al, 2004). […]
There is no assessment measure of PIU that is both psychometrically sound and widely accepted. Most of the existing measures have adapted diagnostic criteria from other psychological disorders to PIU and lack adequate psychometric properties. […]
Nothing significant has been published in the past five years to change the overall conclusion that research into problematic Internet use remains, in a word, problematic.
The new study (Snyder et al., 2015) published in The Conversation (and on Slate) suffers from at least one of the same flaws as Czincz identified. It used a self-identified convenience sample of problematic users or student samples, which significantly bias the results. A good researcher wouldn’t imagine asking people, “Hey, if you think you have X problem, come do our study,” and then conclude with a straight face: “The conclusions come through loud and clear. PIU exists and it affects family relationships.”
Despite the flaws with Snyder’s study, it actually is interesting qualitative research. I just object to the way the researchers broadly characterize their findings. Characterizations that are then further twisted by mainstream media sites and headline editors who don’t seem to know what they’re talking about.
For further information
Slate’s version of the article: “Problematic Internet Use” Is Now Officially a Thing
The Conversation’s version of the article: There’s a new addiction on campus: Problematic Internet Use (PIU)
Caplan, S. E. (2010). Theory and measurement of generalized problematic Internet use: A two-step approach. Computers in Human Behavior, 26, 1089–1097.
Czincz, J. & Hechanova, R. (2009). Internet addiction: Debating the diagnosis. Journal of Technology in Human Services, 27.
Laconi, S., Florence Rodgers, R., & Chabrol, H. (2014). The measurement of Internet addiction: A critical review of existing scales and their psychometric properties. Computers in Human Behavior, 41.
Snyder SM, Li W, O’Brien JE, Howard MO. (2015). The Effect of U.S. University Students’ Problematic Internet Use on Family Relationships: A Mixed-Methods Investigation. PLoS ONE, 10: e0144005. doi:10.1371/journal.pone.0144005
Tokunaga, R.S. (2015). Perspectives on Internet Addiction, Problematic Internet Use, and Deficient Self-Regulation. In E. Cohen (Ed.), Communication Yearbook 39 (pp. 131–161). New York: Routledge.
Tokunaga, R.S. & Rains, S.A. (2010). An Evaluation of Two Characterizations of the Relationships Between Problematic Internet Use, Time Spent Using the Internet, and Psychosocial Problems. Human Communication Research, 36, 512–545.
- Yes, “Internet gaming disorder” disorder appears in the DSM-5, under “conditions for further study.” But that’s it, no broader “problematic Internet use” is there. This issue remains hotly contested by researchers, so it is hardly a settled matter. Furthermore, anything that falls into the “conditions for further study” cannot be used for diagnosis, billing, or reimbursement, so nobody recognizes them as “official.” [↩]
- I had a sense of deja vu while reading this article, so I didn’t have to look far to see I covered a very similar claim made back in 2012 — that Internet addiction is finally proven “real” and “officially” exists. What happened then? Someone saw a draft of the DSM-5 where a similar disorder was included in the category of “conditions for further study.”
Problematic Internet Use didn’t even make it into the final DSM-5 as a distinct disorder in any form, however. [↩]
- Her full response is reproduced below in our Comments section. [↩]
- At the very end of the article, the editors of The Conversation did request that the author point out a single limitation of their research and provide a “justification of why the results are significant.” That hardly negates the misleading headline or puts the research into any kind of broader context or perspective. [↩]