Chrissy Clary

ponder. conspire. digitize.

Who and what do we trust?

We as a society are still placing trust in large brands and big names. Maybe that trust is wavering or changing a bit, but in a society where alternative facts and echo chambers are real, it is becoming increasingly difficult to recognize authority and know what to trust.

When I googled the phrase “what makes an authority,” I didn’t receive the results I had hoped for. Instead of handily presenting me with a psychological paper, my search returned a list of resources designed to teach me how to become an authority myself.

The Launch Coach wants to help me become an authority, not simply an expert. The Goins Writer can help me get 100K readers in 18 months, and the Authority Blog Starter Kit shows what an Authority website really is.

Before I conducted this search, I thought we had to earn authority. From this results list, it looks as if we can purchase it at a reasonable rate. To understand the difficulty in knowing exactly who to trust, or consider an expert, think about how many people out there are calling themselves experts. “In 1970, there were about two dozen think tanks. Today, there are over 3,500 worldwide” (Weinberg, 2011).

To dig into this idea, I ran a few studies through Amazon Mechanical Turk—a service that allows anyone access to a massive number of users at a low cost. I ran two tests. In the first test, I surveyed 200 people, asking them to select the tweet that was most likely true.

The tweets that appeared to garner the most trust were a posting by NBC about celebrities raising money for semolina victims (31 percent) and a tweet from Embry-Riddle Aeronautical University’s head of athletics with recent stats from a game (27 percent). Coming in a close third was a posting by Business Insider Magazine about Ikea’s effort to build shelters for homeless people that can be assembled quickly (21 percent).

The final two tweets identified as trustworthy were retweets by people who carried no brand recognition. One claimed a black man had been killed by white supremacists in New York City (11 percent); the other accused Congressman Devin Nunes of protecting President Trump (10 percent).

I also asked survey participants why they chose as they did.

Here are some of my favorite responses from my survey: “Don’t think anyone would make it up;” “This tweet has most number of hearts and retweets;” “Nunes is corrupt;” “Because it is from a reputed news channel with blue tick, which means it is authentic;” “It is from a reputable Twitter account and the story is non-clickbaitey;” “Ikea often does things like this, it is totally believable. As it is the one most easily proven, I chose it over the others;” “It seems factual enough and people don’t usually lie over scores.”

In the second test, I asked Amazon Turk workers to review a website and tell me which sites of the ones presented were most likely to be reputable and which were least likely to be reputable. I had 50 responders to this test.

The five sites selected for this test originated from a Google search for the phrase “raising children” and are presented in the table below.

Website % who selected not reliable % who selected reliable
Raising Special Kids 14% 14%
Wikihow 35% 41%
Washington Post 19% 21%
American Psychological Association 14% 12%
Up Worthy 18% 12%

A few reasons users gave for identifying a website as disreputable:

  • Raising Special Kids: “The site design looks cheap and dated (I’m on a desktop). There’s not really any helpful information on the front page. It looks like I’m going to have to click around to find out what sort of site it is.”
  • Wikihow: “I think the article is good. But I am not sure if the person who wrote the article is trustworthy.”
  • Washington Post: “Too long.”
  • American Psychological Association: “Pretty disorganized format with poor choice of fonts and bad front-end code upon checking the page’s development features.”
  • UpWorthy: “It’s a site known for click-bait and low quality content”

Reasons participants offered for identifying a website as a reputable source:

  • Raising Special Kids: “The website looks legit and is followed by .org.”
  • Wikihow: “It seems to incorporate the most different sources, rather than just relying on one person’s point of view.”
  • Washington Post: “It is a news article and has a Harvard psychologist offering tips.”
  • American Psychological Association: “APA is a trusted source.”
  • UpWorthy: “It seems relevant to what I am looking for but also interesting to read.”

From my anecdotal surveys, I can begin to make a couple of assumptions.

First, we still have trust in branded organizations. In many instances, survey participants indicated that they made their selections based on brand recognition.

Second, our personal experiences greatly affect what we find believable. In the end, it appears our confirmation biases—a tendency for look for information that confirms previously held ideas—heavily influence our judgments about what is true and who or what we should believe.

Third, design plays an important role in how we decide if information is true.

Next Read: Is There a Conversation Around Ethics and Digital Marketing Happening Today? 

References:  

Weinberger, D. (2012). Too Big to Know: Rethinking Knowledge Not That the Facts Aren’t the Facts, Experts Are Everwhere, and the Smartest Person in the Room is the Room. Basic Books. New York.

2 Comments

  1. Interesting! But, I’m glad you refer to these as “anecdotal surveys.” Doing original research shows initiative, for sure. But I’m curious how you chose your Mechanical Turkers, because that obviously will have a huge effect on your results.

    • claryc

      May 16, 2017 at 5:29 pm

      I did not narrow down the survey group by any demographics, so it was first come first serve. I paid ten cent each for the first survey, and a nickel each for the second. The decrease in payment made it harder to get the survey completed. I did find that I can begin to narrow my audience and target differently. It is really cool. For $14 I can get all kinds of questions answered.

      I be some of that turk data is being used to power machine learning.

Leave a Reply

Your email address will not be published.

*

© 2018 Chrissy Clary

Theme by Anders NorenUp ↑