Motivating Quality

Abstract

This study examines the type of edit that amateur editors called Advisors used in their comments on Epinions.com product reviews and the extent to which their editing-related comments might have motivated reviewers to revise and update their reviews. Advisors made substantive-type suggestions most frequently, but for the most part, reviews that received editing-related comments were not updated more often than were those with nonediting-related comments. Unlike professional editors, Advisors lack gatekeeping control that compels writers to revise their work, but as companies recognize the value of quality user-generated content, they may use amateur editors more often, perhaps in conjunction with professional technical editors.

Keywords

amateurs credibility editing editors Epinions motivation product reviews professionals quality replication UGC user-generated content

More than ever, consumers turn to the opinions of other consumers at Web sites such as Yelp.com, Amazon.com, and Epinions.com for purchasing recommendations rather than or in addition to seeking out the advice of paid experts (e.g., experts who test products at Consumer Reports or technology and restaurant reviewers in the New York Times). In 2008, in the United States alone, over 82 million people created content for the Web, and over 115 million people used that content. By the end of 2013, over 115 million people in the United States will have created content, and over 154 million will have used it (Verna, 2009). Although these numbers include user-generated content (UGC) other than online reviews posted to sites like Epinions and Yelp, the numbers are impressive.

Despite the vast amount of content that users now produce, the quality of that content is not so clearly established. Rello and Baeza-Yates (2012), for example, examined the quality of UGC by measuring lexical quality—the frequency with which users misspelled words. They examined five types of social media sites, including opinion sites such as Epinions, and compared those results to the lexical quality of the Web in general. Opinion sites, along with the other four types of UGC sites, showed lower lexical quality than did the Web in general and, together with social network sites, showed the lowest lexical quality of the types studied. Such findings are important, given that the extant research on UGC quality suggests that quality does indeed matter. Otterbacher’s (2011) comparison of the most versus the least prominent reviews on Amazon, IMDb, and Yelp showed that the least prominent reviews used multiple punctuation marks (e.g., !!!!!) more often and contained more spelling and grammar errors. Prominent reviews used blank lines, a formatting choice that increases readability. Research on UGC quality, then, suggests that room for improvement certainly exists and that improving UGC quality is worthwhile.

Part of that lack of quality appears to stem from a dearth of editorial oversight. Discussing the effects of our ability to quickly and easily share our opinions online, Keen (2007) argued that the democratization of media in online spaces has sidelined such gatekeepers for quality as professional editors so that “instead of creating masterpieces…[we] are creating an endless digital forest of mediocrity” (pp. 2–3). Although Rice (2009), in his analysis of response in networked exchanges (discussion forums and blogs), meant to complicate the relationship between editorial policy and professional writing, he too acknowledged that “disciplinary responses at the editorial level insure stable professional identities as well as a professional writing space” (p. 304). Keen’s rather bombastic argument about mediocre content and Rice’s far more nuanced consideration of users’ responses to one another both make clear the substantial role that editors play in developing and maintaining content quality.

In this article, I report a study of the product review site Epinions, specifically, a study of the comments that the site’s amateur editors supplied to its reviewers. Several prior studies have focused on Epinions because it is a well-established site (started in 1999). Also, compared to sites that focus on particular product categories (e.g., CNet.com or TripAdvisor.com), it covers a wide range of products (see Liu et al., 2008; Mackiewicz, 2010; Pitsilus & Chia, 2010). But Epinions is an interesting UGC site to study for another reason: It identifies members who have demonstrated a superior ability to provide feedback to reviewers and enlists these members as amateur editors in order to help reviewers revise and update their reviews so that the reviews will be more helpful. Epinions grants these site members special recognition—status as Advisors—and charges them with providing “constructive feedback via comments to reviewers on how to improve [review] content quality” (“Rating Reviews,” n.d.). Advisors wield some power within their specific product categories (e.g., electronics or home and garden) in the site because their ratings of reviews (very helpful, helpful, somewhat helpful, and not helpful) carry more weight than do other members’ ratings. Thus, an Advisor’s review rating strongly affects a review’s prominence and the size of its readership.

Even though Advisors have special status within Epinions, they are not professional editors; they are amateurs who have an interest in the product category and in developing quality content about products within that category. In her study of technical communication blogs, Cleary (2012) pointed out that a main difference between professionals and amateurs is that professionals are paid for their work. Besides having “the capacity to make money,” they have formal education or training (p. 11, see also Carliner, 2012, pp. 51–53). Advisors receive no monetary compensation for their efforts, and they lack the relevant training and education that professional editors possess. In addition, unlike professional editors, Advisors have no measure of control over content and its dissemination. They are not gatekeepers to the site. Thus, they cannot block reviews that break the Epinions policy on flaming. They do not approve or authorize reviews that meet a certain quality threshold, and they cannot remove reviews that fail to meet those standards. An Advisor can ask reviewers to revise and update their reviews, but the reviewers are not obligated to do so.

My purpose here is two-fold. First and foremost, I want to report my study of Advisors’ comments to reviewers, a study that replicates,¹ approximately replicates (see Sasaki, 2012), and expands my earlier study of Advisors’ comments (Mackiewicz, 2011). Second, I want to speculate on the impact that Advisors and other amateur editors might have on technical editors’ work with UGC, particularly in relation to UGC quality as well as on their potential impact on the teaching of technical editing.

In my 2011 study, I investigated the types² of edit (Van Buren & Buehler, 1980) that Advisors used in their comments to reviewers and the extent to which their comments motivated reviewers to revise and update their reviews. Although I found that, in general, Advisors’ comments had no motivational impact on reviewers, these comments addressed a range of edit types. But they focused particularly on the substantive type (i.e., suggestions for comprehensive editing). Thus, in their concern for global-level issues such as completeness and accuracy, Advisors countered a prevailing view of editors’ work—that it involves, for the most part, making minor changes in punctuation, word choice, and syntax. Such surface-level wordsmithery exemplifies what Ames and Jensen (2004) called commodity work—work that does not make a strong and strategic contribution to an organization. In the eyes of many if not most, editing work is not symbolic–analytic work (Johnson-Eilola, 2004; Reich, 1991)—work that involves complex tasks and analysis and requires workers to have an “advanced set of knowledge and skills” that allows them to synthesize information into knowledge and to “improve communication both internally and externally” (Dicks, 2010, p. 54). By providing new findings about Advisors’ comments to reviewers, the study that I describe here responds to Geisler et al.’s (2001) call for IText research investigating the components of “effective” networked communication (pp. 280–282).

Specifically, in my 2011 study, I analyzed 76 Advisor comments on 60 reviews of technical products and found that Advisors overwhelmingly used substantive-type comments as opposed to the other four types that I found in the data—the screening, policy, format, and language types. Van Buren and Buehler’s (1980) system of levels and types of edit describes the thoroughness and complexity of editing work in order to define the scope of such work, to facilitate scheduling, and to allocate resources effectively for it. The nine types of edit in their system move from substantive editing, the most involved work, to coordination editing, the least involved. (See Table 1 for examples of Advisors’ editing-related comments at five types of edit.) I examined the types of edit that Advisors used to convey their suggestions—paying particular attention to their politeness strategies—as they advocated for users by mentoring reviewers (helping them meet the expectations of the discourse community for the category for which they were writing, such as electronics, and helping them achieve higher review ratings).

Table 1.

Examples of the Five Types of Editing-Related Advisor Comments Found in the Data.

Type of Edit	Definition	Example
Substantive	Deals with the content of the document, including the coherence and consistency of the document’s parts	“Since Epinions’ readers come to our site seeking information that will help them to make an informed purchasing decision, we look for both technical information and the reviewer’s user experience in product reviews here.”
Screening	Ensures basic correctness, including the absence of misspellings and incomplete sentences	“I would suggest that you use a spell checker and remember to capitalize your I’s.”
Policy	Ensures conformity to the organization’s policies, such as absence of derogatory comments and advertisements	“While I’m not implying anything, I am saying that the review could have been written from published specs rather than from real world experience which is the essence and a requirement of the Eps TOS.”
Format	Ensures conformity in typography, layout, and visual elements	“Just wanted to mention that dividing a review into a few paragraphs helps the eyes of those reading and also helps organize thoughts you are expressing.”
Language	Ensures clear and effective presentation of text, including grammar and syntax, usage, conciseness, abbreviations	“I can write a very informative review of a camera without saying camera more than once, maybe twice. Using the same word repetitiously is not only distracting it can and will turn readers away.”

The following study expands my 2011 study by analyzing a larger data set of 142 Advisor comments on 105 product reviews and by comparing Advisors’ editing-related and nonediting-related comments on updated reviews (reviews that reviewers had edited) and nonupdated reviews in order to determine whether these amateur editors’ comments might have played a role in motivating reviewers to edit and update their reviews. This new study also examines the extent to which Advisors’ praise (e.g., “Very nicely done: lots of detail backed by commentary on your user experiences with the product’s features. Fine job!”) might have influenced reviewers’ motivation to update their reviews. It replicates and approximately replicates the 2011 study by examining the type of edit of Advisors’ editing-related comments and the affective (relationship) work that Advisors did in their nonediting-related comments. The findings I report here complicate the findings and conclusions of that earlier study. Before I present the results of this study, I will briefly discuss the research on product reviewers and reviews and explain my research questions and methods.

Research on Product Reviewers and Reviews

In light of the increasing amount of UGC in general and consumers’ increasing use of online reviews of products and services in particular, a growing number of business and other communication researchers have begun to examine reviewers’ motivations for contributing reviews (Hennig-Thurau, Gwinner, Walsh, & Gremler, 2004; Wasko & Faraj, 2005; Yoo & Gretzel, 2008). This body of research suggests a number of potential reasons for reviewers’ expenditure of their time and effort. One important motivator is altruism, “the act of doing something for others without anticipating any reward in return” (Sundaram, Mitra, & Webster, 1998, p. 527). Another potential motivator, according to Hennig-Thurau, Gwinner, Walsh, and Gremler (2004), is the satisfaction that people receive from communicating experiences and evaluations to a real audience. In other words, people might be motivated to contribute content to a site because doing so provides them with a social benefit: “Such behavior signifies their participation in and presence with the virtual community of platform users and enables them to receive social benefits from this community membership” (p. 42). As I will discuss later, Advisors’ comments to reviewers draw on and appeal to these motivations.

Research Questions

This study investigates the extent to which reviewers who had received editing-related comments from Advisors also updated their reviews—an action suggesting that they had followed Advisors’ suggestions. Thus, my first research question (Research Question 1) asks the following: Were editing-related Advisor comments attached more frequently to updated reviews than to nonupdated reviews?

My second research question relates to the first in that it examines the effect of Advisors’ praise—a positive politeness strategy (see Brown & Levinson, 1987) that can act as a counterbalance or buffer to suggestions (e.g., Hyland & Hyland, 2001)—on the extent to which reviewers updated their reviews. Researchers have examined the role of praise in writing conferences (e.g., Thonus, 1999; Walker & Elias, 1987) as well as in editor–writer sessions (e.g., Mackiewicz & Riley, 2003). Such research has shown that besides buffering suggestions, praise (i.e., identifying writers’ accomplishments and encouraging them to continue their efforts) builds rapport (i.e., affective relationship work) and, in the case of tutor–student writer conferences, promotes return visits. In light of this research suggesting that praise can motivate writers to revise and improve their work, I examine Research Question 2: Is praise more frequently found in Advisors’ editing-related and nonediting-related comments on updated reviews than it is in their editing-related and nonediting-related comments on nonupdated reviews?

My third research question also examines the role of affect. My 2011 study indicated that Advisors used nonediting-related comments in two ways: to welcome reviewers who were new to the site or, at least, the product category and to share their own experiences with the product or similar products. Both of these discourse moves—welcoming and relating experiences—build rapport and thus serve a relational purpose much like the positive politeness strategy of praise. In building rapport, such discourse moves facilitate effective communication—a competency that falls squarely within the realm of symbolic–analytic work (Dicks, 2010). To replicate the 2011 study’s examination of Advisors’ comments that are unrelated to offering feedback for improving writing, I investigate Research Question 3: Do Advisors’ nonediting-related comments serve any purposes besides welcoming reviewers and sharing personal experiences? That is, do these comments serve purposes that fall outside Advisors’ stated charge to suggest ways that reviewers can improve their reviews in order to make them more helpful?

In my 2011 study, I found that the majority of the Advisors’ editing-related comments (24 of the 35, or 69%) focused on substantive editing. Substantive editing includes editing content for accuracy (e.g., pointing out to a reviewer that the digital camera under review does indeed have an automatic flash) as well as making suggestions about adding, deleting, or reorganizing content. I also found that Advisors’ comments far less frequently addressed the screening, policy, format, and language types of edit; thus, Advisors less often concerned themselves with matters such as spelling and capitalization, adherence to Epinions policy, formatting, and grammar and word choice. Just 11 (31%) of the 35 editing-related comments focused on these types. These findings suggested that tasks often viewed as commodity work occurred with far less frequency than did tasks that were more in line with knowledge-producing, symbolic–analytic work—work that addresses substantive matters such as accuracy and completeness. To test these findings from my earlier study, I examine Research Question 4: Does a larger sample of Advisors’ editing-related comments reveal the same frequency of substantive-type comments and the same infrequency of comments of other types? Finally, expanding beyond testing the prior study’s findings in order to determine whether reviewers were more likely to make changes relating to one type of edit more than other types, I examine Research Question 5: Were editing-related comments that suggested a particular type of edit attached more frequently to updated reviews than to nonupdated reviews?

Method

Research assistants and I collected a sample of 105 reviews related to the following technical search products: digital cameras and lenses, stand mixers, video cameras, vacuums, garden tools, microwaves, televisions, toasters, CD players, GPS systems, refrigerators, and laptop computers. As Mudambi and Schuff (2010) explained, search products are products that consumers can obtain information on before purchasing (p. 187). Because the utility of such products stems from “tangible, seemingly objective criteria,” consumers can evaluate and compare them and “feel rather comfortable relying on other consumers’ evaluations” when making purchasing decisions (Sen & Lerman, 2007, p. 79). In contrast, experience products—goods such as wine and music—are products that consumers must purchase (or sample) in order to evaluate them (Nelson, 1970). An evaluation of their quality would more likely be based on subjective criteria, so individual consumers would need to purchase and use the products in order to form an opinion about them (Nakayama, Sutcliffe, & Wan, 2010, p. 252). To collect these reviews, we followed the same procedure that I used in my 2011 study, searching from the most recent reviews (from 2012) to the older reviews (from 2002). We gathered reviews of products through all price ranges. We gathered only reviews that had received an Advisor’s comment.

These 105 product reviews generated 142 Advisor comments: 71 comments attached to 49 reviews that reviewers had updated and 71 comments attached to 56 reviews that reviewers had not updated. As in my 2011 study, I then separated the Advisor comments into two groups: comments that suggested one or more ways to change the review according to some type of edit (e.g., substantive) and those that did not suggest a way to edit the review. Also, as in the previous study, I coded the comments’ types of edit according to the appearance of certain words or phrases that are relevant to each type. For example, I coded comments that included content, details, or information as substantive; misspelled or capitalize as screening; Epinions policy or Review Writing Guide as policy; white space, paragraphs, or bold as format; and repetition as language. In the previous study, a trained coder and I separately coded a set of comments taken from the data set, achieving 93% agreement. As in the previous study, we coded comments with every type-of-edit code that applied; however, in this study, only seven comments (5%) addressed two or three rather than one type.

Then we identified both editing-related and nonediting-related comments that contained both formulaic praise (e.g., “Nice review”) and nonformulaic praise (e.g., “I like the balance you strike between product specs & features and personal experience. The humor you bring to your reviews makes them all the more inviting and readable”). Thus, a comment could be coded both for a type of edit and for praise. To code for praise, we relied on my (Mackiewicz, 2006) description and delineation of compliment types (based on Manes & Wolfson, 1981) in writing conferences between tutors and technical writing students.

Results

In this section, I discuss the results of my research questions for this study and compare these results to the findings of my 2011 study.

Frequencies of Comments on Updated Versus Nonupdated Reviews

My findings for Research Question 1, which examined whether editing-related Advisor comments (of any type of edit) were attached more frequently to updated reviews than to nonupdated reviews, show that reviews that were updated did not receive more editing-related comments than did reviews that were not updated (see Table 2). Fewer than half (46%) of the comments on updated reviews were editing related whereas more than half (53%) of the comments attached to nonupdated reviews were editing related. These results suggest that receiving an editing-related comment from an Advisor is no more of a motivator for reviewers to edit and update their reviews than is receiving a nonediting-related comment. Although instances in which reviewers responded to Advisors’ editing-related comments with explanations of ways that they had improved their reviews suggest that on an individual level, Advisors can connect to and influence reviewers, the quantitative results show that the simple presence of an editing-related suggestion is not in itself enough to get reviewers to lend more of their time and effort toward improving the review for other Epinions members.

Table 2.

Frequencies of Editing-Related and Nonediting-Related Comments on Updated and Nonupdated Reviews.

Comment Type	Number Attached to Updated Reviews (%)	Number Attached to Nonupdated Reviews (%)	Total Number of Comments (%)
Editing related	31 (46)	40 (53)	71 (50)
Nonediting related	36 (54)	35 (47)	71 (50)
Total	67 (100)	75 (100)	142 (100)

Praise as a Motivator

Drawing on research examining student writers working with more expert others (e.g., writing center tutors) and on studies suggesting that praise can motivate writers to revise and improve their work, this study posed Research Question 2, which examines whether praise is more frequently found in Advisors’ editing-related and nonediting-related comments on updated reviews than it is in their editing-related and nonediting-related comments on nonupdated reviews. Of the 142 Advisor comments, 47 (33%) contained praise for reviewers’ work. The results for Research Question 2 show that the praise was quite evenly split between comments attached to updated reviews and comments attached to nonupdated reviews (see Table 3). This finding suggests that, in itself, praise does not play a substantial role in a reviewer’s decision to edit. Also, in editing-related comments, praise had little effect. Only six reviews with editing-related comments containing praise were updated whereas eight reviews with editing-related comments containing praise were not. These findings suggest that praise’s motivating effect on writers may be weaker than what previous research has suggested, at least in contexts in which the editor—or amateur editor—cannot mandate changes, so the writer is not compelled to comply.

Table 3.

Advisors’ Use of Praise in Nonediting-Related and Editing-Related Comments in Updated and Nonupdated Reviews.

Praise + Comment	Updated Reviews (%)	Nonupdated Reviews (%)	Total (%)
Praise + nonediting-related comment	18 (75)	15 (65)	33 (70)
Praise + editing-related comment
+ Substantive	4 (17)	6 (26)	10 (21)
+ Format	1 (4)	2 (9)	3 (6)
+ Policy	1 (4)	0 (0)	1 (2)
+ Screening	0 (0)	0 (0)	0 (0)
+ Language	0 (0)	0 (0)	0 (0)
Subtotal	6 (25)	8 (35)	14 (30)
Total	24 (100)	23 (100)	47 (100)

Once again though, on a case-by-case basis, praise appeared to make an Advisor’s suggestions about editing more palatable to reviewers. In particular, Advisors who used specific, individualized praise to balance their suggestions—as opposed to general praise, such as “Great job here”—showed reviewers that they had closely read their work and had found that work to be of quality. For example, Advisor k1drm (2010) praised vladhed for using specific content, noting an odd characteristic of the garden tiller:

I liked how you pointed out the oddity of the 40:1 fuel mix ratio. Sounds like you need to keep two separate 2-cycle gas cans. That could be a pain.

After offering this specific praise, k1rdm made a substantive-type suggestion for editing, asking vladhed to add more detail (specifications) about the tiller:

I wanted to rate this review helpful or better, but I think our Epinion readers would benefit from a little more information on the product. It would help if you could add some additional detail such as weight, brand of engine, fuel capacity, how long it runs on a tank of fuel, etc.

In this case, using praise to balance out a suggestion paid off: vladhed not only edited and updated the review but also wrote a comment (attached to the original review) that thanked k1rdm for the “constructive comment.”

Advisors used praise in nonediting-related comments as well; 33 (70%) of the 47 occurrences of praise did not accompany a suggestion for editing. This finding signals that Advisors were thinking not just about ways that reviewers could improve their reviews and ways to motivate them to do so but also about reviewers’ experience as they participated within the Epinions discourse community in general or a particular category community in particular. Thus, such praise appeared to serve two main purposes. With it, Advisors could show reviewers goodwill that might lead them toward reviewing additional products and even becoming regular contributors to the category and site. And especially with specific as opposed to general praise, Advisors could mentor reviewers in what constitutes valued writing within the community by suggesting their criteria for evaluating reviews and their reasons for rating reviews as they did. For example, Advisor rudixeno (2009) left a comment conveying characteristics that readers appreciate in a review:

Glad to see you writing in the Home & Garden Section. I like the balance you strike between product specs & features and personal experience. The humor you bring to your reviews makes them all the more inviting and readable.

The praise that rudixeno used suggested Epinions users’ preference for reviews that not only provide technical specifications and descriptions—the kind of information often available on manufacturers’ Web sites—but also relate the reviewer’s personal experiences using the product. Through the very act of noticing and noting a reviewer’s effort, then, Advisors can use such specific praise both as a mentoring tool and as an extension of goodwill.

Thus, although the findings for Research Question 2 suggest that praise in itself did not play a substantial role in motivating reviewers to edit their reviews, they also suggest that praise signaled Advisors’ attention to reviewers’ work and made the Advisors’ suggestions for improving the reviews more tolerable and less face threatening for reviewers.

Other Purposes for Advisors’ Comments

Just as praise served affective purposes (even if it did not seem to motivate reviewers to edit), so too did Advisors’ welcoming statements and their statements relating their own experience with the product. Those were the two purposes of Advisors’ nonediting-related comments that I found in my 2011 study. So in this study, I wanted to see whether Advisors’ nonediting-related comments served other purposes. Research Question 3, then, examines whether Advisors’ nonediting-related comments serve any purposes besides welcoming reviewers and sharing personal experiences. My analysis of the 71 nonediting-related comments showed that, as in the 2011 study, Advisors used such comments to welcome reviewers and share similar experiences. For example, Advisor njchicaa (2009) made this welcoming comment on momoften’s review of a stand mixer: “Just wanted to say welcome to the site.” Such short salutations acknowledged the reviewers’ presence in the category community and showed them that they had indeed connected with an audience. While such welcomes tended to be minimal, they let reviewers know that they were not reviewing in a vacuum.

More interesting were those comments in which Advisors shared their own experiences and, in doing so, laid the groundwork for an online conversation. For example, sweeper (2008) compared his own experience with the product—a Nikon camera lens—with that of the reviewer:

I had considered the Nikon 12-24 myself.…I had the D70 at the time and it focused albeit but using the camera motor. I also use it with my D300.…I borrowed the Nikon version and used it in the field for comparison. The Nikon feels better built but you simply can’t tell the difference between the two on the images. Someone swears the distortion is a bit more controlled in the Tokina model but that was pushing.

In such cases, Advisors addressed reviewers as peers with a common interest. By responding with a conversational turn that made room for a reviewer’s response, Advisors further engaged reviewers in the category’s discourse. In doing so, Advisors helped to initiate and integrate reviewers into the site’s community of content creators and thus worked to increase the likelihood that those reviewers would contribute content to the site again.

But this study also revealed a third purpose to Advisors’ nonediting-related comments: thanking reviewers. Advisors thanked reviewers for their original reviews and for their edited and updated reviews. In so doing, Advisors acknowledged reviewers’ (sometimes substantial) efforts, such as when cntaur5 (2009) thanked myord for an update: “I deleted my initial comment. Thanks for the clarification on your review.” By acknowledging the difference between myord’s original review and the updated version, cntaur5 made explicit to myord and to other reviewers that extra effort gains notice, gets original (suggestion-containing) comments from Advisors removed, and potentially brings about the reward of a higher review rating.

Why do Advisors bother acknowledging reviewers’ efforts by showing gratitude? Clearly, as representatives for Epinions, Advisors use thanking to build goodwill. Indeed, as Raggio and Folse (2009) pointed out, failing to offer thanks creates the risk of being perceived as ungrateful. More important, thanking acts as a “moral reinforcer,” as a social reward that generates more of the actions for which a person receives thanks (p. 456). So thanking provokes altruistic behavior; it encourages reviewers to give more time and effort in the future for the benefit of others. The speech act of thanking, then, has a prosocial benefit. It profits the community as a whole.

This finding adds to the list of affective work that Advisors do when they are not making editing-related suggestions. Such relationship work, like the relationship work accomplished with praise, reveals the complexity and breadth of these amateur editors’ work. By writing comments that welcomed reviewers, shared personal stories, and thanked reviewers, Advisors generated goodwill and supported effective interactions; thus, such nonediting-related comments exemplified the kind of communicative skill involved in symbolic–analytic work.

Frequencies of Types of Edits in Comments

Moving from inquiring about Advisors’ relationship work to checking the main result of my 2011 study, Research Question 4 examines whether a larger sample of Advisors’ editing-related comments reveals the same focus on substantive-type editing. Based on my previous finding, I argued that by focusing their comments on the substantive editing concerns of completeness and accuracy, Advisors attempted to ensure that reviewers fully addressed the concerns and met the expectations of the category discourse community. My findings here confirm the prior result: Of the 71 editing-related comments, 53 (75%) addressed substantive concerns. And, as in the 2011 study, Advisor comments addressing format, screening, language, and policy were far less frequent (see Table 4).

Table 4.

Frequencies of Types of Edit in Advisors’ Editing-Related Comments.

Type of Edit	Frequency (%)
Substantive	53 (75)
Format	11 (15)
Screening	9 (13)
Language	5 (7)
Policy	3 (4)

Note. Seven comments addressed two or three levels of edit (n = 71).

Advisors’ low proportion of language-type editing comments reveals the extent to which their work extended beyond the commodity work often associated with editing. In the 2011 study, just 1 (3%) comment of the 35 editing-related comments addressed language issues; similarly, in this study, just 5 (7%) of the 71 editing-related comments addressed language issues. Advisors’ lack of attention to the language type of editing issues perhaps stemmed from a lack of relevant education and professional training in grammar and usage. As amateurs, Advisors might have lacked the necessary expertise to diagnose and recommend solutions for problems requiring language-type editing.

Even though the findings show that Advisors did not focus on grammar and other language-type concerns, a different narrative plays out online when Epinions members and Advisors themselves discuss the scope of Advisors’ work and power within the site. To many Epinions users and reviewers, it seems, Advisors are sticklers for grammatically correct writing. For example, golferinfr (2011) made the following comments about Epinions on another review site:

Epinons.com [sic] is one of the most stringent yet nurturing web site[s] for new writers and old writers like myself. You better have your pencil sharpen[ed] when writing on this site, not like other sites that basically just have online web editors. When you are finished with your [review] it goes under a micro-scope of real Advisors or category leads for the section you are writing about.…Very Helpful means you covered all the basics of the subject that you are writing on, base[d] on presentation, grammar, hands on experience with such product.

Such comments suggest that Advisors have a reputation for focusing on language correctness even when quantitative analyses of the distribution of their comments across types of edit strongly suggest otherwise. Comments like golferinfr’s also signal that, to many, editing equals commodity work—so much so that users believe language-type comments exist where they, for the most part, do not.

Indeed, in the Epinions discussion boards, Advisors and other site members debated whether Advisors are too powerful (e.g., whether Advisors should be able to delete reviews that are poorly written and not helpful) and whether their feedback is arrogant and nitpicky—particularly whether they criticize reviewers too harshly for poor grammar. For example, Tina C. (2011), a reviewer who abandoned Epinions in favor of the newer, localized site Yelp, balked at the idea of having her review rated by an Advisor, citing bad experiences:

Reviewers are labeled as “Advisors” with a high index of unfathomably nasty attitude.…[They are] full of hot bombastic air to boast their greatness yet bares [sic] no solid substance to hold any water.

More interesting than site members’ umbrage at what they perceive to be Advisors’ smug grammar checking, forum discussions like the one that generated Tina C.’s comment revealed that some Advisors share Tina C.’s perception: They feel that Advisors focus too often on language-type editing. In their forum discussions, Advisors articulated the importance of balancing an imperative to help reviewers improve review quality with the goal of fostering and mentoring reviewers, especially new reviewers. They feel that Advisors should tend to relationships and sustain reviewers’ motivation to contribute rather than police grammar. As Advisor sleeper54 (2011) puts it, “I feel strongly the idea that Advisors need to be ‘grammer’ and ‘riting’ instructors to do their job well is pernicious and damaging to new writers and to the site as a whole.” Similarly, Advisor Stephen_Murray (2011), in a discussion about providing “grammar refreshers” to reviewers, pointed out that when it comes to grammar advice, “most posters resent it and do not take it.” Such comments suggest a belief that suggestions for language-type editing are not only damaging to relationships but also pointless in that they are typically ignored.

Frequencies of Types of Edit in Comments on Updated Versus Nonupdated Reviews

While examining Research Question 4 meant replicating the 2011 study with a larger data set, it also meant analyzing the larger data set in terms of types of edit in comments on updated versus nonupdated reviews. Thus, Research Question 5 examines whether editing-related comments of a particular type of edit were attached more frequently to updated reviews than to nonupdated reviews in order to see whether reviewers suggested changes relating to one type of edit more than other types. The results for Research Question 5 show that of the five types of edit that appeared in Advisors’ comments, only comments focused on screening were associated with updated reviews substantially more than with nonupdated reviews, appearing in 11% of the updated reviews and only 1% of the nonupdated reviews (see Table 5).

Table 5.

Frequencies of Types of Edit in Advisors’ Comments on Updated Versus Nonupdated Reviews.

Type of Edit	Updated Reviews (%)	Nonupdated Reviews (%)
Substantive	21 (30)	32 (45)
Screening	8 (11)	1 (1)
Format	4 (6)	7 (10)
Policy	3 (4)	0 (0)
Language	3 (4)	2 (3)

Note. Seven comments addressed two or three levels of edit (n = 71).

These results also show that amateur editors’ suggestions to edit and update did not spur reviewers into action, with the possible exception of comments related to surface changes such as capitalization, spelling, and punctuation. The following comment from Advisor elzora (2011) exemplifies the screening-related comments associated with updated reviews:

Presentation means a lot, and unfortunately this review has several problems. You have misspelled words such as “substnadard” and “mattered” (matted?). You [referred] to the “on and off switch” as the “one and off switch”. There are many other misspelled words (wil, mush). You can fix these easily by running the spell checker and taking a moment to proof read your review. Your review has a good start but I can’t rate higher than SH [somewhat helpful] due to the presentation. We don’t expect perfection, and your reviews are always welcome here. If you update please drop me an email and I will be glad to take another look.

In this comment, elzora explained the problems that kept the reviewer from receiving a higher review rating and offered suggestions for improving the review (run the spell-checker and proofread). In making the revising task seem less burdensome, elzora pointed out that “we”—a pronoun that she likely intended to mean Advisors in addition to readers of the product category’s reviews—expect to see an error now and again. In short, elzora articulates the level of quality (and the tasks involved in attaining it) that the reviewer must reach in order to justify a higher review rating, but she also conveys that the intense effort to perfect the review that the reviewer might think necessary, in reality, is not.

Screening-type comments from Advisors were associated mostly with updated reviews and thus likely played a role in motivating reviewers to update. Elzora’s comment prompted a response from the reviewer, setufe (2011), who wanted to explain that the errors were not indicative of the reviewer’s usual writing standard: “Sorry for all the typos. That’s certainly not my usual. I was frantically typing the review for the third time as I kept getting an html error.” After elzora articulated the problems with the original review and pointed out that the work involved in fixing the problems was rather minimal, setufe updated the review, replacing it with a new, spell-checked version. The association between Advisors’ screening-type comments and updated reviews suggests that reviewers will do further work on their reviews to improve review quality if that work demands minimal time and effort.

Comparing the Two Studies

Table 6 summarizes the findings from the 2011 study and this study and shows the findings that correspond and those that do not. The components of this study that replicate the 2011 study confirm the earlier findings. Thus, I can say with some certainty that Advisors’ comments often (at least half the time) do not focus at all on editing for improved quality. Rather, they focus on welcoming and thanking reviewers as well as on sharing product-related experiences with them. Their editing-related comments focus more on substantive than on other types of editing concerns. In contrast, findings from a larger sample of Advisor comments fail to support the 2011 finding that Advisors’ editing-related comments are associated with updated reviews. The findings from the current study suggest that Advisors’ comments have little influence on whether reviewers edit and update their reviews. But this study does find that screening-type comments are strongly associated with updated reviews. Finally, the current study has systematically examined reviewers’ praise in relation to updated and nonupdated reviews. Praise does not appear to make a difference in whether reviewers update their reviews.

Table 6.

A Comparison of Findings from the 2011 Study with Findings from the Current Study’s Replication, Approximate Replication, and Expansion of the Prior Study.

Purpose of Study	Prior Finding (Mackiewicz, 2011)	Current Finding	Comparison
Replication (Research Question 1)	Majority (54%) of comments are nonediting related.	Half (50%) of comments are nonediting related.	Prior finding is confirmed.
Replication (Research Question 3)	Nonediting-related comments focus on welcoming and sharing experiences.	Nonediting-related comments focus on thanking, in addition to welcoming and sharing experiences.	Prior finding is confirmed.
Replication (Research Question 4)	The majority (69%) of editing-related comments focus on the substantive type.	The majority (75%) of editing-related comments focus on the substantive type.	Prior finding is confirmed.
Replication (Research Question 4)	One (3%) editing-related comment focuses on the language type.	Few (7%) editing-related comments focus on the language type.	Prior finding is confirmed.
Approximate replication (Research Question 1)	Compared reviews with editing-related Advisor comments to reviews with no Advisor comments at all: 40% of reviews with editing-related comments are updated; only 20% of reviews with no Advisor comments are updated.	Compared updated and nonupdated reviews, all with Advisor comments: Updated reviews do not have more editing-related comments.	Prior finding is not supported. Advisors’ editing-related comments are not associated with updated reviews.
Expansion (Research Question 2)	NA	Compared frequency of praise in editing-related comments attached to updated and nonupdated reviews: There is no difference in frequency of praise between updated and nonupdated reviews.	NA
Expansion (Research Question 5)	NA	Compared levels of edit in comments attached to updated and nonupdated reviews: Only screening-level comments are associated with updated reviews substantially more than with nonupdated reviews.	NA

Conclusion and Implications

This replication, partial replication, and expansion study revealed that the comments of Epinions.com Advisors—amateur editors—failed to motivate reviewers to revise and update their reviews. Advisors’ editing-related comments were no more associated with reviews that product reviewers had updated than were their nonediting-related comments. Advisors lack professional status—status that stems, as Cleary (2012) pointed out, largely from being paid for work and from having relevant education. Advisors lack the efficacy that professional editors enjoy: the right and responsibility not to publish any content that they consider unsatisfactory. Thus, reviewers might see acting on Advisors’ feedback—the feedback that they do not consider nitpicky—as useful but ultimately optional.

Adding to this challenge for Advisors is another: Reviewers receive little incentive to revise. The possibility of receiving a higher Advisor rating on the review helpfulness scale and the greater visibility that a higher rating brings might motivate the subset of reviewers who strongly affiliate with the site and want to cultivate a reputation within the discourse community. But many reviewers use the site sporadically. Such reviewers log on, review one or two products, and then fall silent again—sometimes for months or even years. For these intermittent content generators, the incentives of achieving a better rating and greater review visibility likely generate insufficient motivation to update.

In editing-related comments, Advisors suggested changes to reviews across five types of edit, but by far they made substantive-type suggestions most often. Updating reviews in response to these substantive-type suggestions—advice related to filling in omissions and checking facts—would require, in many cases, substantial time and effort. In contrast, updating reviews in response to format, policy, and language types of suggestions would involve little time and effort. For example, adding headings to a review to organize the content and make the review more searchable takes little time or effort. But these comment types were rarely associated with updated reviews, even when such comments contained the motivating politeness strategy of praise. Of course, the fact that reviewers did not take up Advisors’ suggestions does not mean that the advice had no value. Advisors’ suggestions were available not just to the individual reviewers to whom they addressed their feedback but also to all Epinions users and, thus, to potential reviewers. These users received guidance in writing helpful reviews. So Advisors’ editing work had a broader impact on existing and potential content contributors than the findings of this study indicate.

More and more, technical communicators are working with UGC (e.g., Cleary, 2012; Rauch, Morrison, & Goetz, 2010)—architecting environments for it, coding and classifying it, and performing other skilled, symbolic–analytic tasks. As more studies reveal that the quality of UGC substantially affects consumers’ purchasing decisions (Chevalier & Mayzlin, 2006; Ghose & Ipeirotis, 2011; Senecal & Nantel, 2004; Zhang & Dellarocas, 2006), technical communicators—technical editors in particular—will continue to play a role in developing and improving UGC. Companies and other organizations that rely on UGC for word-of-mouth advertising will increasingly see the value in highlighting quality content and, concomitantly, motivating and helping users to develop quality content. And to some degree, technical editors’ work with UGC will relate to how much review sites (e.g., Epinions.com), retailer sites (e.g., Amazon.com), and perhaps even brand sites (e.g., Patagonia.com) use Advisor-like amateur editors to develop and curate site content. If companies find that using amateur editors is a cost-effective and reliable way to help maintain and monitor UGC, the use of amateur editors as advocates of users and mentors of reviewers may spread. And professional technical editors could find themselves responsible for overseeing teams of amateur editors. These realities and possibilities mean that understanding the ways that amateur editors such as Advisors approach their work on a site will become an important part of developing quality UGC.

The findings of this study also have implications for teaching technical editing. In technical editing classrooms, instructors devote substantial time to both the editor–writer relationship and the types of edit. Students looking at amateur editors’ comments will get a broader view of the editor–writer relationship by seeing an amateur variation on it enacted in a different, nonprofessional rhetorical situation. They will also get another view of editing priorities. While professional technical editors commonly work on types of edit besides the substantive type, amateur editors prioritized substantive-type editing, a focus that arose from their drive to get reviewers—writers who have no obligation to do more work—to make their reviews more thorough and accurate.

Reviewers did edit and update their reviews after receiving screening-type comments, and this finding likely stems from the ease with which reviewers could satisfy screening-type suggestions. In their willingness to make screening-type changes, reviewers evoke novice writers such as many first-year composition students who are willing to focus on surface-level issues (e.g., spelling and punctuation) as they revise but, without instructors’ persistent encouragement and even prodding, are reluctant to grapple with complex changes that affect meaning. As Kosma (1991) put it, novice writers “are text-bound, reluctant to jump from surface or word levels to more global decisions, such as ‘What do I want to say?’” (p. 33; see also Barkaoui, 2007). Like such writing students, reviewers focused on local changes that they could accomplish quickly and easily. Given that reviewers had little incentive to give more time and effort to developing and improving their review content, this finding is not so surprising.

To close, this replication, approximate replication, and expansion study revealed a paradox of sorts. Editing work is often seen as commodity work (i.e., rote and mechanical work) as opposed to symbolic–analytic work. Forum comments from reviewers reflect Advisors’ reputation as grammar police; indeed, Advisors’ own forum comments sometimes acknowledge this perception. But Advisors’ editing-related comments on reviews—particularly the extent of their substantive-type comments—reflected the symbolic–analytic side of editing work, making salient the fact that technical editing involves far more than tinkering with writers’ syntax and certainly more than correcting their capitalization and punctuation errors. Their substantive-type comments, however, in addition to their language, policy, and format types of comments, appeared to have no special motivational effect. Yet, their screening-type comments—the sort of editing concerns typically seen as commodity work—did indeed appear to be related to reviewers’ likelihood of updating their reviews.

Amateur editors, like professional technical editors (and writing instructors too), need a measure of influence to move writers to revise and improve their work. So empowered, amateur editors—perhaps working in conjunction with professional technical editors—may be able to raise the level of UGC quality as they change misperceptions about editing.

Footnotes

Acknowledgment

I would like to thank my research assistant, Myles Cryer, for his help in gathering product review data and coding types of edit.

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Ames

A. L.

Jensen

S. M

. (2004). Transforming your career: Contributing strategically to your company or client. In 51st Annual STC conference proceedings (pp. 119–122). Baltimore, MD: Society for Technical Communication.

Barkaoui

(2007). Revision in second language writing: What teachers need to know. TESL Canada Journal, 25, 81–92.

Benoit

W. L.

Holbert

R. L.

(2008). Empirical intersections in communication research: Replication, multiple quantitative methods, and bridging the quantitative–qualitative divide. Journal of Communication, 58, 615–628.

Brown

Levinson

S. C.

(1987). Politeness: Some universals in language use. Cambridge, England: Cambridge University Press.

Carliner

(2012). The three approaches to professionalization in technical communication. Technical Communication, 59, 49–65.

Chevalier

J. A.

Mayzlin

(2006). The effect of word of mouth on sales: Online book reviews. Journal of Marketing Research, 43, 345–354.

Cleary

(2012). Discussions about the technical communication profession: Perspectives from the blogosphere. Technical Communication, 59, 8–28.

cntaur5. (2009, January 18). Retrieved from http://www.epinions.com/review/Canon_EOS_50D_Body_Only_Digital_Camera/content_458023931524/show_~allcom

Collins

H. M.

(1985). Changing order: Replication and induction in scientific practice. Beverly Hills, CA: Sage.

10.

Dicks

R. S.

(2010). The effects of digital literacy on the nature of technical communication work. In Spilka

(Ed.), Digital literacy for technical communication: 21st Century theory and practice (pp. 50–81). New York, NY: Routledge.

11.

elzora. (2011, August 29). Retrieved from http://www.epinions.com/review/Euro_Pro_NV22_Shark_Navigator_Vacuum/content_562243341956/show_~allcom

12.

Gaddis

J. L.

(2002). The landscape of history: How historians map the past. Oxford, England: Oxford University Press.

13.

Geisler

Bazerman

Doheny-Farina

Gurak

Haas

Johnson-Eilola

… Yates

(2001). IText: Future directions for research on the relationship between information technology and writing. Journal of Business and Technical Communication, 15, 269–308.

14.

Ghose

Ipeirotis

P. G.

(2011). Estimating the helpfulness and economic impact of product reviews: Mining test and reviewer characteristics. IEEE Transactions on Knowledge and Data Engineering, 23, 1498–1512.

15.

golferinfr. (2011, May 1). Retrieved from http://www.dooyoo.co.uk/internet-sites/epinions-com/1525924/

16.

Haswell

R. H.

(2005). NCTE/CCCC’s recent war on scholarship. Written Communication, 22, 198–223.

17.

Hennig-Thurau

Gwinner

K. P.

Walsh

Gremler

D. D.

(2004). Electronic word-of-mouth via consumer-opinion platforms: What motivates consumers to articulate themselves on the Internet? Journal of Interactive Marketing, 18, 38–52.

18.

Hyland

(2001). “Sugaring the pill”: Praise and criticism in written feedback. Journal of Second Language Writing, 10, 185–212.

19.

Johnson-Eilola

(2004). Datacloud: Toward a new theory of online work. Cresskill, NJ: Hampton Press.

20.

k1drm. (2010, July 4). Retrieved from http://www.epinions.com/review/Craftsman_2_cycle_Mini_Tiller/content_516794650244/show_~allcom

21.

Keen

(2007). The cult of the amateur: How today’s Internet is killing our culture. New York, NY: Doubleday.

22.

Kosma

R. B.

(1991). Computer-based writing tools and the cognitive needs of novice writers. Computers and Composition, 8, 31–45.

23.

Lazonder

H. W.

van der Meij

(1993). The minimal manual. Is less really more? International Journal of Man Machine Studies, 39, 729–752.

24.

Liu

Lim

E.-P.

Lauw

H. W.

M.-T.

Sun

Srivastava

Kim

Y. A.

(2008). Predicting trusts among users of online communities: An Epinions case study. In Proceedings of the 9th ACM conference on electronic commerce (pp. 310–319). New York, NY: ACM.

25.

Mackiewicz

(2006). The functions of formulaic and non-formulaic compliments in interactions about technical writing. IEEE Transactions on Professional Communication, 49, 12–27.

26.

Mackiewicz

(2010). Assertions of expertise in online reviews. Journal of Business and Technical Communication, 24, 3–28.

27.

Mackiewicz

(2011). Epinions Advisors as technical editors: Using politeness to motivate comprehensive editing. Journal of Business and Technical Communication, 25, 421–448.

28.

Mackiewicz

Riley

(2003). The technical editor as diplomat: Linguistic strategies for balancing clarity and politeness. Technical Communication, 50, 83–94.

29.

Manes

Wolfson

(1981). The compliment formula. In Coulmas

(Ed.), Conversational routine: Explorations in standardized communication and prepatterned speech (pp. 115–132). The Hague, the Netherlands: Mouton.

30.

Mudambi

S. M.

Schuff

(2010). What makes a helpful online review? A study of customer reviews on Amazon.com. MIS Quarterly, 34, 185–200.

31.

Nakayama

Sutcliffe

Wan

(2010). Has the Web transformed experience goods into search goods? Electronic Markets, 20, 251–262.

32.

Nelson

(1970). Information and consumer behavior. Journal of Political Economy, 78, 311–329.

33.

njchicaa. (2009, April 29). Retrieved from http://www.epinions.com/review/pr-KitchenAid_K5SS-Multiple_Colors_Mixer/content_467907743364/show_~allcom

34.

Otterbacher

(2011). Being heard in review communities: Communication tactics and review prominence. Journal of Computer-Mediated Communication, 16, 424–444.

35.

Pitsilus

Chia

P. H.

(2010). Does trust matter for user preferences? A study on Epinions ratings. In Nishigaki

Murayama

M. Y.

Marsh

(Eds.), Trust management IV (pp. 232–247). New York, NY: Springer.

36.

Porte

Richards

. (2012). Focus article: Replication in second language writing research. Journal of Second Language Writing, 21, 284–293.

37.

Raggio

R. D.

Folse

J. A. G.

(2009). Gratitude works: Its impact and the mediating role of affective commitment in driving positive outcomes. Journal of the Academy of Marketing Science, 37, 455–469.

38.

Rating reviews on Epinions.com. (n.d.). Retrieved from http://www1.epinions.com/help/faq/show_~faq_rating

39.

Rauch

Morrison

Goetz

(2010). Are we there yet? An examination of where we’ve been and where we’re headed as technical communicators. Proceedings of the International Professional Communication Conference (pp. 297–309). Mahwah, NJ: IEEE.

40.

Reich

(1991). The work of nations: Preparing ourselves for 21st-century capitalism. New York, NY: Knopf.

41.

Rello

Baeza-Yates

(2012). Social media is NOT that bad! The lexical quality of social media. In Proceedings of the Sixth International AAAI Conference on Weblogs and Social Media (pp. 559–562). Retrieved from http://www.aaai.org/ocs/index.php/ICWSM/ICWSM12/paper/…/4711/5057

42.

Rice

(2009). Networked exchanges, identity, writing. Journal of Business and Technical Communication, 23, 294–317.

43.

rudixeno. (2009, December 8). Retrieved from http://www.epinions.com/review/pr-Hoover_SteamVac_F5914-900_Vacuum/content_494468566660/show_~allcom

44.

Sasaki

(2012). An alternative approach to replication studies in second language writing: An ecological perspective. Journal of Second Language Writing, 21, 303–305.

45.

Sen

Lerman

(2007). Why are you telling me this? An examination into negative consumer reviews on the Web. Journal of Interactive Marketing, 21, 76–94.

46.

Senecal

Nantel

(2004). The influence of online product recommendations on consumers’ online choices. Journal of Retailing, 80, 159–169.

47.

setufe. (2011, August 30). Retrieved from http://www.epinions.com/review/Euro_Pro_NV22_Shark_Navigator_Vacuum/content_562243341956/show_~allcom

48.

sleeper54. (2011, January 20). Retrieved from http://www.epinions.com/msg/show_~threads/cat_id_~121/id_~28238/forum_id_~441/pp_~2#270829

49.

Stephen_Murray. (2011, January 20). Retrieved from http://www.epinions.com/msg/show_~threads/cat_id_~121/id_~28238/forum_id_~441/pp_~2#270886

50.

Sundaram

D. S.

Mitra

Webster

(1998). Word-of-mouth communications: A motivational analysis. Advances in Consumer Research, 25, 527–531.

51.

sweeper. (2008, June 2). Retrieved from http://www.epinions.com/review/pr-Nikon_Zoom_Super_Wide_Angle_AF_12-24mm_f_4G_ED-IF_DX_AF-S_Autofocus_Lens/content_432389590660/show_~allcom/pp_~1/sort_~date/sort_dir_~asc/sec_~comment_list

52.

Thonus

(1999). How to communicate politely and be a tutor, too: NS-NNS interaction and writing center practice. Text, 19, 253–279.

53.

Tina

(2011, April 13). Retrieved from http://www.yelp.com/biz/epinions-brisbane#hrid:xUbRmy5SFgOFCpg8Oq1JBA

54.

Van Buren

Buehler

M. F.

(1980). The levels of edit (2nd ed.). Arlington, VA: Society for Technical Communication.

55.

Verna

(2009, February 19). A spotlight on UGC participants. Retrieved from http://www.emarketer.com/(S(zcjol0454gw0plmaiciprejw))/Article.aspx?R=1006914

56.

Walker

C. P.

Elias

(1987). Writing conference talk: Factors associated with high- and low-rated writing conferences. Research in the Teaching of English, 21, 266–285.

57.

Wasko

M. M.

Faraj

(2005). “Why should I share?” Examining social capital and knowledge contribution in electronic networks of practice. MIS Quarterly, 29, 35–57.

58.

Yoo

K. H.

Gretzel

(2008). What motivates consumers to write online travel reviews? Information, Technology, and Tourism, 10, 283–295.

59.

Zhang

Dellarocas

. (2006). The lord of the ratings: How a movie’s fate is influenced by reviews. In Proceedings of the 27th International Conference on Information Systems (pp. 1959–1978). Milwaukee, WI: Association for Information Systems.