Frank Waddell Ph.D. is an assistant professor in the University of Florida’s College of Journalism and Communications. He studies new technology and online storytelling, including work related to automated news and the psychology of online comments. The interview has been edited for length and style.
Tell us about yourself
I am a journalism assistant professor in the College of Journalism and Communications. I got my Ph.D. in mass communications at Penn State. Before that, I was at Virginia Tech for six years, where I got my undergraduate and masters, also in communications. Most of my research as part of the Consortium on Trust in Media and Technology is situated at the intersection of technology and credibility. That’s sort of me in a nutshell. I am obviously doing research but also teaching graduate classes in research methods, theory and statistics as well.
Tell us about the research that you are doing for the Consortium on Trust in Media and Technology.
In general, my research frames the idea that the internet, and digital media as a whole, have radically changed the way audiences evaluate and consume information. The internet has been both a blessing and a curse, right? We have incredible access to information now, but at the same time, we have no really good ways to sift through all of it. The increasing prevalence of sources has necessitated that consumers find ways to prioritize quickly and that is something from psychology that is a well-established trait. We do not like to think very hard about every single decision that we make during the day. You would not go and get out of bed in the morning if you tried to do that.
When you go on the internet to search for something, you make a decision. We try to use the same quick rules in our everyday behaviors. These decisions are something that psychologists and communication scholars refer to using names like cognitive heuristics, or rules of thumb.
Basically, they are just these quick ways of making judgments that do not require us to think a lot. One of the classic examples in education is students, when they do not know the answer on an exam, will write a really long response because they use a heuristic that, “if it is a long response then it must be a good response.” In reality, there is nothing that really equates the length of an essay with the strength of an essay, but that is a heuristic that they hope the professor might use.
In reality, there are lots of heuristics that we use in our day-to-day behavior but one that I have done a lot of research on is something called the Bandwagon Heuristic. It is basically the tendency to trust that, if the crowd thinks that something is correct or if the audience thinks something is good or bad, then their opinion of it must be normative, it must be correct.
It is basically herd behavior. This is very common online because every type of media that we have nowadays has … likes, comments and shares. These are opportunities for people to express their opinions, but unfortunately people also rely upon these sources to make judgments about the news or whatever is being portrayed.
And this impacts how people evaluate news?
A lot of my research has shown that you can take a news article on any topic and do nothing to it except change whether readers in the comment section have said positive things or hostile things or nothing at all. What we tend to find is what others sometimes call the “nasty effect.” Basically, when other viewers or readers say nasty things about the journalist or complain about the quality of the writing, it can lead people to actually think that the news that is being shown or written about is also less credible.
This effect seems to happen because people make generalizations based on what just a few people might say online. So, the top three comments on the news article might be that this reporter has no idea what they are talking about and, unless you have a really strong pre-existing belief, you might be predisposed to simply generalize and think, “this is how most other people see it, therefore I am going to feel the same way.”
Unfortunately, something that operates across human experience is called negativity bias. Negativity bias, as the name implies, is just our tendency to be more likely to remember negative information. An example I like to use is if I give a talk. Afterwards, let’s say 30 people come up to me and 29 tell me, “hey, that was a great talk.” And one tells me, “hey, you had a misspelling in your PowerPoint.” A lot of people will ruminate on that one negative instead of the 29 positives.
Unfortunately, that is the way that people tend to remember online comments. A lot of people might be saying constructive things but just a few negatives can be what ends up making people think that the collective is mostly negative toward the content.
Maybe you can relate this back to trust. We have this herd mentality and we have people generalizing and exhibiting negativity bias. What does that tell us about trust and communication?
So, here is the bottom line: we have a crisis in trust right now. It applies to media in general and also to how people evaluate individual pieces of news. So, a natural question might arise, “what are the different things that make us trust news more or make us trust news less?” My research suggests that regardless of what a journalist is actually writing, what the crowd or other audience members are saying about them through things like online comments can undermine their trustworthiness, despite the fact that it has nothing to do with the actual journalistic quality of the writing. That is because of things like the Bandwagon Effect.
So, journalists basically are in a terrible situation. Not only do audiences not think very critically about what they are reading, because they are overwhelmed with information, but when they do make judgments, it is sometimes totally unrelated to the quality of the writing.
So, basically what you are saying is that the reporting and the facts matter less than what people are saying about those facts?
In some cases, yes. I am just saying that a journalist can do nothing wrong at all. They can have a fine piece of journalism, but if the audience turns hostile to them, that can bring down trust. Which, to me, is surprising because when journalists work a beat for years and years, they are basically the expert, and you have somebody essentially coming off the street with a reaction—and they are the ones who could bring down the trust.
Tell us a little bit about the research you have done that explains all this.
In my research, what I do is take a news article that is topical at a time when the study is done. For example, one time I did research on the opioid epidemic and so this was a very standard news article just describing some of the dangers, the challenges and how common the opioid crisis is. Everybody reads the exact same news article. Let us say we have 90 people. So, for one-third, at the end of the article, the audience says, “what a great article.” But another third sees the article with comments at the end of it are really negative. Then, the remaining third just see the article itself, with no comments. They are what is called a control, or baseline condition.
Afterward, everybody answers the same questionnaire, asking them, “Was this news article trustworthy? Was it credible? Was the issue that was described important?” What my research shows is, very consistently, those value judgments, those estimates of trust, are lower when people read news accompanied by negative comments. What is really unfortunate is that you do not see an effect going the other direction, so journalists never benefit from people saying, “what a great news article.” That does not increase trust. It just keeps it at baseline. It is almost exclusively bad news for journalists
Tell us what else you have studied.
Another thing I have been looking at is the rise of automated journalism. A lot of news articles are written nowadays automatically, using algorithms that have been programmed by journalists or other specialists. These are things like sports scores or the weather or financial reports, the typical drudgery that journalists used to have to do. What I was curious about was, how do audiences evaluate news differently when they think that automation has either written it entirely or at least contributed to it? Some researchers before me predicted that there will be heuristics saying that if a machine or an algorithm created this, then it must be more trustworthy or less biased than if a human created it. It is almost like people thought that we would trust machines more.
That was the expectation. So, the research that I have done is very similar in methodology to what I just described. Take the exact same news article, but this time instead of reading the comments, vary who readers think the author is: either a human journalist or an algorithm or “robot reporter.” To my surprise, what several of my first studies found was the literal opposite of what I was expecting. People actually dislike news articles when they think they are written by algorithms rather than humans.
There seems to be a backlash against machine-written news and what my research seems to suggest was it is like a violation of our expectations. We expect robots to be in factories or to vacuum our carpets. We do not expect them to be writing the news. So, that was surprising.
But I also realized it may be that in the pretty near future there might be something that other scholars have called a man-machine marriage, basically where journalists are using algorithms actively to write news. It is more of a collaboration. I thought maybe it would be interesting to look at that. What I found was audiences trusted news the most when it was that type of collaboration. So, they are accepting of algorithms when a journalist is still involved. When they see it is a journalist using this new interesting tool, they say, “oh, okay.”
You have done research surrounding gender as well. Tell us about that.
Something that really caught my eye not too long ago was the increasing hostility and harassment of journalists, especially journalists who are female. There are a lot of interesting statistics. Something like 50 percent of working female journalists have been harassed by readers in the past six months and it is on the rise. So, that alone just sort of rang alarm bells for me.
We have a lot of academic research on sexism, but not from a journalism perspective. A lot of us do not know journalists personally. So, how is it that we are learning what a journalist does? I think people mostly learn about it through entertainment media. Unfortunately, Hollywood executives and TV producers have no idea what journalism is like. If they have made TV shows to show what journalism really is like, it would probably get pretty boring. So, unfortunately, to make things more interesting, they sensationalize and often show a picture of reality that is nothing like what journalism is like. For women, it is especially bad.
I would say nine times out of ten, if there is a female journalist in a movie or in a TV show, it follows a super predictable pattern. The female journalists are almost always presented as incompetent, almost always presented as hyper-sexualized. Oftentimes, they are portrayed as people who would do anything to get a story. Female journalists in real life have lamented that men assume that they are hitting on them whenever they want an interview. They assume that men are learning this from TV shows, right? Where else is it coming from?
So, what I was really interested in was, can we look at this from a research perspective? Where I started was with the basic question, what type of person watches TV shows where female reporters behave unethically and says, “what I just watched on TV or in this movie is the way things actually are in real life.” That is something that researchers call perceived realism. I had people watch a video clip from either “House of Cards” or “Thank You for Smoking,” two notorious examples of Hollywood sexualizing journalism. Those video clips are situations where the female journalist offers sexual favors for a news story. Afterwards, people rated how realistic they thought the scene was.
I also measured things like how distrustful they are of the news media in general; how much they hold sexist beliefs, said things like “men are superior to women” and that type of thing; and held very specific beforehand beliefs about how common it is for women to sleep with their sources. What I found was that all three of those things increased realism. So, if you were sexist, you were especially likely to think that the trailer was realistic. If you distrusted media already, you found that the media representation was very realistic. Most strongly of all, if you already thought that female journalists sleep with their sources, then you found that specific scene more realistic.
That sounds like common sense, but the reason why it is really bad news is that it kind of creates a vicious cycle. It sounds alarm bells for Hollywood. One thing that is not great about our media environment is it is actively telling lies about the nature of journalism, and that further perpetuates the myth for the stereotype.
You took this work a bit further. Tell us about that.
The follow-up study is a little more academic in nature. I was just curious, is there actually a unique phenomenon of people holding discriminatory attitudes towards female journalists? Is this something that is distinct from sexism? Is this something that is distinct from disliking media? So, to make a long story short, I did a couple different studies where I measured their beliefs about women in the news media. What I found was…there is a unique phenomenon that is sexism towards female reporters, which goes beyond just regular sexism and that goes beyond just thinking that women are incompetent. It goes beyond thinking that the news media are not trustworthy. So, it is a unique phenomenon and it has consequences.
When you take all of this work together, what does it tell you?
All of the available statistics show trust in the news media is declining. I think my research starts to highlight just some of the reasons why. In situations where we cannot think critically about what we are reading, we rely on things that really should not bear any importance on thinking about credibility or trust—but they ultimately do.
You can think of trust as a pie chart. Every slice of the pie is a reason why we trust something more or less. Comments are a slice of the pie. Authorship is a slice of the pie. What my research has shown is a lot of these things do matter, but it is more difficult to make statements about which one might matter more or less.