Tag Archives: online reviews

New Federal website shows complaints about mortgage lenders

Posted on by

Thanks to the Consumer Financial Protection Bureau, there is a new website for narratives of consumer complaints regarding mortgage lenders:

The bureau logs each complaint by category in a publicly viewable database and gives the company that is the subject of a complaint time to respond via a nonpublic online portal connecting it with the consumer through a bureau intermediary. In the past three years, according to the bureau, it has received and worked on more than 627,000 complaints. They range from alleged harassment by debt-collection attorneys, to foreclosures, student loan defaults and poor treatment of customers by loan servicers. Roughly 28 percent of all complaints filed to date have been about mortgage issues — the largest single category. What’s been missing, though, has been any real detail about the troubling circumstances that triggered the complaint in the first place expressed in the customer’s own words.

Starting in late June, that all changed. The bureau began posting what it calls “narratives” that name the bank or company involved and go into sometimes excruciating detail. Allegations get pretty serious — charges of lending fraud, violations of federal regulations and illegal overcharges. Some are heartfelt, such as one from a Virginia homebuyer whose closing was repeatedly delayed by the bank: “Who compensates us for the loss of income for the days taken off from work (to attend closings)? For the movers that have been scheduled? For the pre-move-in renovations that cannot now be done because the contractors are fully scheduled for the rest of the summer?” (To see the narratives, go to http://tinyurl.com/phnkq99)

The first batch of 7,700-plus narratives was posted June 25, including hundreds of mortgage complaints. The consumer’s name and address — other than state of residence — are redacted, as are all details the bureau or the consumer considers ?private.

Lenders are not permitted to post their own narratives, but instead must use one of several stock responses, such as “company can’t verify or dispute the facts in the complaint” or “company believes it acted appropriately as authorized by contract or law.” Lenders can also decline to participate in the narratives process by saying, “Company chooses not to provide a public response.”

The article suggests two large threads emerge from the complaints: dislike of being placed in customer service hell without getting answers from anyone and problems with escrow accounts.

Not surprisingly, lenders are not happy with this information on the website. The issue is similar to that which plagues many online reviews: how can businesses or readers be sure that the story or review is credible? Yet, this certainly puts more information on the side of consumers and this is needed in an industry that holds so much debt for so many people.

These narratives posted online would make for some good coding opportunities for social scientists…

How trust builds and then declines through online reviews

Posted on by

Two sociologists explain how trust develops and then changes over time in the online sharing economy:

For their research, Parigi and Cook examined Couchsurfing, a website that supports international travel and cultural exchange. Its members both host visitors and surf the site to find sympathetic lodging as they travel the world, all without exchanging money. Profile pages of members list Couchsurfing friends and other personal information.

The findings revealed, the researchers wrote, an interesting mechanism at the root of interpersonal trust: “The accumulation of ratings about users (whether guests or hosts) had a double-edged effect on trust and relationships: it made relationships easier to establish initially but it also weakened them after a certain threshold.”

In other words, technology boosted interpersonal trust among users at first, but it also made it more difficult to build stronger ties as users acquired more and more reviews…

Parigi and Cook explain that in an online community, interactions between people are more normalized, less open to chance. “This is because trustworthiness is promoted not by interpersonal ties, but by the monitoring of one another in a network in which reputations are posted,” they wrote…
As a result, he said, an interesting conundrum seems to be emerging: technology makes it possible for people to trust complete strangers, while at the same time it may be weakening the bonds that unite individuals.

Trust is a necessary component of human relationships; people need to have some confidence that the other person is not going to take advantage of them or let them down and deeper connections can form when mutual trust develops. Yet, trust can develop in different ways. In this particular case, it sounds like the trust is built on crowd-sourced data – posted online reviews – that contribute to confidence but don’t necessarily lead to deeper relationships.

Maybe this is all okay. We don’t expect to form deep relationships with everyone we interact with, particularly when it comes to economic transactions. (Think of interactions with cashiers or waiters or others at the lower status jobs in the service economy.) The larger issue may be when most or all of online interactions develop these qualities. This is the same sort of question that worried Georg Simmel in his piece “The Metropolis and Mental Life.” Simmel made a similar argument: when humans enter a new social space (the big city) that is built around interdependence and specialization, humans can’t hope to get to know everyone. Similar situation today: humans enter a new social space (the Internet, social media) that prioritizes individual action and choosing what connections to make.

Analysis suggests fake Twitter followers common among Washington political leaders

Posted on by

A new analysis of political leaders in Washington D.C. suggests many of them have a lot of fake or inactive Twitter followers:

Of the president’s 36.9 million Twitter followers, an astonishing 53 per cent – or 19.5 million – are fake accounts, according to a search engine at the Internet research vendor StatusPeople.com. Just 20 per cent of Obama’s Twitter buddies are real people who are active users.

Overall, the five most influential accounts linked to the Obama administration – the first lady has two – account for 23.4 million fake followers.

Biden’s nonexistent fans make up 46 per cent of his Twitter total, with 20 per cent being ‘real’ followers. The White House’s followers are 37 per cent fake and 25 per cent active; the first lady’s primary account is 36 per cent fake and 29 per cent active…

The difference between fake followers and ‘real’ ones is comprised of ‘inactive’ accounts, which may relate to real people but no longer send tweets with any regularity.

If this analysis can be trusted, this appears to be a bipartisan problem. But, it would be helpful to hear more about how inactive or fake users are determined: shouldn’t we expect that there are some people on social media platforms like Twitter and Facebook who have set up profiles but then don’t use them regularly? At least there is one infographic that helps provide more detail regarding this phenomena. Plus, you can use this app to analyze your own account.

And, once we have such numbers, we should then think through what it means: is it dishonest for politicians to have a lot of fake or inactive Twitter followers? Should the standards for having fully active followers be different for politicians as opposed to other public figures? Does having more followers really translate into a more positive public image or more votes?

UPDATE: This is not a problem just relegated to well-known figures. See this story from this morning on fines levied against companies that posted fake reviews:

The New York Attorney General has slapped 19 companies with a $350,000 fine after his office unearthed fake review writing for Google Local, Yelp, and others in a yogurt shop sting.

Eric Schneiderman revealed that a raft of search engine optimisation (SEO) companies created dummy accounts and paid writers from the Bangladesh and the Philippines $1 to $10 per review after his office set up a fake yoghurt shop in Brooklyn, New York and sought help to combat negative comments.

“Consumers rely on reviews from their peers to make daily purchasing decisions on anything from food and clothing to recreation and sightseeing,” said Schneiderman in a statement.

“This investigation into large-scale, intentional deceit across the Internet tells us that we should approach online reviews with caution.”

Plus, this is an online concern at sites like Amazon where reviews provide important information for potential buyers.

Sad: newest version of SimCity causing a large outcry

Posted on by

As a longtime fan of SimCity, I have not enjoyed how the changes to the newest version have turned off many fans:

Electronic Arts’ long-awaited release of SimCity on Tuesday should have been an occasion for a worldwide collective all-nighter of urban planning, a nonstop bacchanal of factory building, endless intricate min-maxing of grids of pavement. 12 a.m. Eastern Tuesday morning should have been SimCity’s finest hour.

Instead, the whole operation seized up and shit the bed. EA, a technology company with a market capitalization of over $5 billion, could not muster the online servers necessary to handle an influx of players looking to build their cities. This was entirely a problem of EA’s own making, as SimCity was not designed with an offline mode. Even if you don’t want to team up with others and join your cities together, you can’t just build your personal metropolitan layouts in peace: Every player must be constantly connected online, as a draconian step to crack down on piracy of this PC-only game.

Hey, launch hiccups happen, right? Everybody all tries to connect at once, servers get throttled, and you figure out a way to make it work. Trouble is, as of this writing EA hasn’t figured out a thing. SimCityis still totally busted. It’s difficult to log in: Nearly all of the servers are full, and when a player does find one that’s available, attempting to log in usually throws back an error. And you can’t try again until a 20-minute counter finishes ticking down…

In other words, SimCity is currently in the midst of a disaster that makes zombie attacks and nuclear meltdowns seem tame. Electronic Arts’ attempts to fix the problem have not only been unsuccessful, they’ve been making the SimCity blackout even worse, at least from a public relations standpoint: EA said Thursday that it would actually begin removing features from the game in an attempt to get it to run. At first it was non-core features like achievements and high score leaderboards. By the end of the day EA had ripped out the “Cheetah” gameplay mode, which speeds up the passage of time so you can develop your city more quickly…

In response to Wired’s request for comment, an EA spokesperson referred us to a blog post by SimCity senior producer Kip Katsarelis, who wrote that Electronic Arts would be adding new servers until the player base could be fully accommodated, and that it would prioritize stabilizing this situation before it turned the game’s features back on. She did not give a timeframe for the resolution.

It sounds like a bad situation from all that I have read. For example, check out the overwhelmingly negative reviews on Amazon. This is too bad as the pictures of the graphics I’ve seen look beautiful and some of the interactive elements between cities and within regions sound really interesting. But, I agree there should be a one-player mode so a city builder is not always dependent on other players.

In the end, I hope SimCity is redeemed. One thing that has been noticeable in the online complaints is the number of gamers who have enjoyed SimCity for years. It is not flashy, doesn’t involve violence (though you might count destroying your own city by disaster), and might even give you some insights into urban life. People want to play the game but under certain conditions.

Killing books with coordinated one-star reviews on Amazon

Posted on by

I’ve posted before about the wild world of Amazon reviews and here is another example: a group of Michael Jackson fans succeeded in burying a new book about the late pop star.

In the biggest, most overt and most successful of these campaigns, a group of Michael Jackson fans used Facebook and Twitter to solicit negative reviews of a new biography of the singer. They bombarded Amazon with dozens of one-star takedowns, succeeded in getting several favorable notices erased and even took credit for Amazon’s briefly removing the book from sale.

“Books used to die by being ignored, but now they can be killed — and perhaps unjustly killed,” said Trevor Pinch, a Cornell sociologist who has studied Amazon reviews. “In theory, a very good book could be killed by a group of people for malicious reasons.”…

The retailer, like other sites that depend on customer reviews, has been faced with the problem of so-called sock puppets, those people secretly commissioned by an author to produce favorable notices. In recent months, Amazon has made efforts to remove reviews by those it deemed too close to the author, especially relatives.

The issue of attack reviews, though, has received little attention. The historian Orlando Figes was revealed in 2010 to be using Amazon to anonymously vilify his rivals and secretly praise himself. The crime writer R. J. Ellory was exposed for doing the same thing last fall.

This is an interesting world where arguments are being made that people have the right (free speech) to provide harsh and even untrue Amazon reviews.

I don’t envy Amazon for having to deal with this issue where reviews would have to be more closely monitored. Even with close monitoring, people could provide excessively positive or negative reviews as long as they couldn’t be identified as being relatives or bragging out their actions on Facebook (as one member of the Michael Jackson fan group did). It puts Amazon in an unenviable position of having to play the heavy and try to crack down on people.

It would be interesting to see arguments of when these tactics might be supported or praised. Imagine a neo-Nazi writes a book; is it ethical or effective to shut down their book on Amazon? What about an obnoxious political figure on the other side that you can’t stand?

An argument for Amazon’s one-star reviews reveals the role of cultural critics

Posted on by

A professional critic praises Amazon’s one-star reviews:

About a year ago, while shopping online for holiday gifts, I became an unabashed connoisseur of the one-star amateur Amazon review. Here I found the barbed, unvarnished, angry and uncomfortably personal hatchet job very much alive. Indeed, I became so enamored of Amazon’s user-generated reviews of books, films and music that my interest expanded to the one-star notices on Goodreads, Yelp and Netflix, where, for instance, a “Moneyball” review notes the movie “did not make you feel warm and fuzzy at the end as a good sports film should.” How true! A rare opinion on a critical darling!…

But there is a visceral thrill to reading amateur reviewers on Amazon who, unlike professional critics, do not claim to be informed or even knowledgeable, who do not consider context or history or ambition, who do not claim any pretense at all. Their reviews, particularly of classics, often read as though these works had dropped out of space into their laps, and they were first to experience it. About “Moby-Dick,” one critic writes: “Essentially, they rip off the plot to ‘Jaws.'” About “Ulysses,” another critic writes: “I honestly cannot figure out the point, other than cleverness for cleverness’ sake.”

Likewise, to seriously dismiss “The Great Gatsby” as “‘Twilight’ without the vampires,” as an Amazon reviewer did, may be glib and reductive, but it’s also brilliantly spot on, the kind of comparison a more mannered critic might not dare. “Whoever made that ‘Twilight’ comparison, whether they know it, is showing their education, that they can connect new media with old works and draw fresh conclusions,” said David Raskin, chair of the art history, theory and criticism department at the School of the Art Institute of Chicago…

Speaking of honesty: It should be pointed out here that, in general, online amateur reviews are not mean but usually as forgiving as the professional sort. Bing Liu, a data-mining expert at the University of Illinois at Chicago who has studied online reviews — “partly because I was curious if they were real or just someone gaming the system” — told me that 60 percent of Amazon reviews are five-star reviews and another 20 percent are four-star. The information research firm Gartner released a study in September predicting that, within a couple of years, between 10 and 15 percent of online reviews will be paid for by companies — rigged.

It sounds like the argument is this: you can find the average American in the one-star Amazon reviews. Instead of getting the filtered, sophisticated review typically found in media sources, these reviewers give the unvarnished pop culture take. Discussed in this argument is the idea of social class and education. An approved reviewer or critic, the typical gatekeeper, is able to put a work in its context. The educated critic is trying to make the work understandable for others. The educated critic often has experience and education backing their opinions. In contrast, the Internet opens up spaces for individuals to post their own reactions and through aggregation, such as the Amazon five-star review system, have some say about how products and cultural works are perceived.

This new reality doesn’t render cultural gatekeepers completely irrelevant but it does do several things. One, it dilutes their influence or at least makes it possible for more critics to get involved. Second, it also makes more visible the opinions of average citizens. Instead of just theorizing about mass culture or pop culture, we can all see what the masses are thinking at the moment they are thinking it. (Think of the possibilities on Twitter!) Third, it provides space like in this article for reviewers to admit they don’t always want to write erudite pieces but want to have a “normal person reaction.”

Just one problem with this piece: the critic says he doesn’t really read the one-star Amazon reviews for information. Instead, he appreciates the “visceral thrill.” He quotes an academic who says such reviews reveal cultural gaps. Thus, celebrating the one-star reviews may be just another way to assert the traditional reviewer’s cultural capital. Read the one-star reviews for entertainment but continue to go back to the educated reviewer for the context and more valued perspective.

New cultural gatekeepers: paid online reviewers

Posted on by

After recently discussing buying Twitter followers, the New York Times explores another new online realm: paid online reviewers who only give extremely positive reviews.

In the fall of 2010, Mr. Rutherford started a Web site, GettingBookReviews.com. At first, he advertised that he would review a book for $99. But some clients wanted a chorus proclaiming their excellence. So, for $499, Mr. Rutherford would do 20 online reviews. A few people needed a whole orchestra. For $999, he would do 50…

“The wheels of online commerce run on positive reviews,” said Bing Liu, a data-mining expert at the University of Illinois, Chicago, whose 2008 research showed that 60 percent of the millions of product reviews on Amazon are five stars and an additional 20 percent are four stars. “But almost no one wants to write five-star reviews, so many of them have to be created.”

Consumer reviews are powerful because, unlike old-style advertising and marketing, they offer the illusion of truth. They purport to be testimonials of real people, even though some are bought and sold just like everything else on the commercial Internet.

Mr. Liu estimates that about one-third of all consumer reviews on the Internet are fake. Yet it is all but impossible to tell when reviews were written by the marketers or retailers (or by the authors themselves under pseudonyms), by customers (who might get a deal from a merchant for giving a good score) or by a hired third-party service.

I am most intrigued here by the possible change in relationship between a reviewer and an author. The article suggests there is some sort of “sacred” distance between the two: the reviewer is free to criticize the work without recrimination. Some reviewers have attained elite cultural gatekeeper status, people who guide decision-making for millions of people. Think of critics like Siskel and Ebert and Robert Christgau who are seen as authoritative figures. Hence, people are upset when they learn that a positive review they saw wasn’t an “honest” opinion but rather a business transaction.

However, let’s not forget that these reviewers also make careers out of their thoughts – they may not have sold out to a corporation or a product but they do have a financial interest. I would argue that this distance between reviewer and author/creator has never really been so sacred and there are plenty of areas where we are used to paid reviewers. If you follow a reviewer enough, you can often learn what they do or do not like. Indeed, some reviewers have become outspoken proponents of certain movements and not others. Is this based on a completely rational, detached perspective? Of course not. Don’t many reviewers interact with the people who are producing the products they are reviewing? Think of blurbs on the back of books: are these truly unsolicited comments or from people who are truly judging the merits of the book? More crassly, commercials often present “reviewers” or “real people” or people made to sell certain products. Perhaps this is simply a sign of our times and will become normal as there is clearly a market for good reviews.

It will be interesting to see how websites like Amazon, heavily dependent on user reviews, works through this issue. I always try to read both the five star and one star reviews when considering a product. Additionally, there are other issues: the ratings can be about the product itself or a particular aspect of the product or about people’s expectations for the product or the shipping or the customer service or something else. I think Amazon could include a few extra questions, as other websites do, that would help one sort through the variety of reviews. Overall, the system is not perfect and we should be aware that we may not be getting the “unvarnished truth,” but at least it is better than going off anecdotal evidence from a friend or two…right?

The (terrible?) world of “professional” Amazon reviewers

Posted on by

A recent study of some of Amazon.com’s top 1000 reviewers has PC Magazine writer John Dvorak questions the validity of their reviews:

In the first academic study of its kind, Trevor Pinch, Cornell University professor of sociology and of science and technology studies, independently surveyed 166 of Amazon’s top 1,000 reviewers, examining everything from demographics to motives. What he discovered was 85 percent of those surveyed had been approached with free merchandise from authors, agents or publishers.

Pinch, who also found the median age range of the reviewers he surveyed was 51 to 60, a surprise said Pinch, because the image of the internet is more of a young person’s thing. Amazon is encouraging reviewers to receive free products through Amazon Vine, an invitation-only program in which the top 1,000 reviewers are offered a catalog of free products to review…

This is the fraud aspect of the process that cannot be tolerated. And now to find out they are in a much older demographic makes me think they are just product hoarders who will say what they need to say to get more products. This conclusion is hinted at by the professor.

I do not like man on the street reviews. I never have, and I’ve always thought they could be easily corrupted by smart public relations folks who have already dove into what they call social media. This includes phony personas on Twitter and Facebook that are used to sway public opinion, shipping free goodies to “influential” bloggers, and things like this Amazon scandal.

Dvorak is not really arguing that reviews are not valuable but rather that because Amazon does not fully disclose how these reviewers operate, customers could be duped. The problem here is trust: Dvorak and others might assume that reviewers are doing this out of the goodness of their hearts but instead they are “professionals.” Instead, these reviewers are being “paid.” This is a classic gatekeepers problem: how do you know that a reviewer is trustworthy and giving unvarnished opinions? There are plenty of critics these days for various media outlets and websites. I suspect many average citizens have to read through multiple reviews from a single critic to see if their thoughts line up with their own or to see if they are consistent.

Of course, Amazon relies on a crowd sourcing approach, just like aggregator websites such as Rotten Tomatoes or Metacritic. Do these top reviewers really sway people’s opinions about products since there are often many others who provide reviews of the same products?

Why not ask Amazon whether critical reviewers have been kicked out of these programs? Dvorak is suggesting that these reviewers would speak positively about products just in order to receive more – couldn’t Amazon fight back against this?

My first thoughts when I saw this study a while back was that how confident could Pinch be about his findings based on 166 reviewers. Why not go for a larger sample out of the 1000 Top Reviewers?

(Side note: at the end, Dvorak applauds Pinch for tackling this topic:

By the way (and off topic), you should read my writings over the past 30 years, because I have been hounding sociologists around the world to begin to study these sorts of computer and Internet activities. Give Professor Pinch an award, will you! Maybe that will encourage more studies.

Maybe so.)

The online “mega-reviewers”

Posted on by

One of the innovations of online stores is the ability for users to rate what they like and then for other users to base decisions or comment on those previous ratings. A site like Amazon.com is amazing in this regard; within a few minutes, a reader can get a much better idea about a product.

But statistics from Netflix, another site that allows user reviews, indicate that many users don’t rate anything while there is a small percentage of people who might be called “mega-reviewers”:

About a tenth of one percent (0.07%) of Netflix users — more than 10,000 people —  have rated more than 20,000 items. And a full one percent, or nearly 150,000 Netflixers, have rated more than 5,000 movies. By contrast, only 60 percent of Netflix users rate any movies at all, and the typical person only gives out 200 starred grades.

This rating pattern might fit a Poisson or a negative binomial regression where many people rate none or very few movies while there is a smaller percentage who rate a lot. (A useful statistic in helping to figure out the shape of the curve: while there is 40% that doesn’t rate anything, of the 60 percent who rate any movies at all, what is that median?)

The Atlantic talks to two of mega-reviewers who seem to motivated by seeing what the system would recommend to them after having all of their input. Interestingly, they suggest Netflix still recommends movies to them that they don’t like after watching them.