PopGnosis Virtual Group Psychology

Recent readings

I struck gold at my local library with an anthology titled The Handbook of Group Research and Practice, edited by Susan A. Wheelan. The entries intersect with my research perfectly. For instance, I have just started reading the chapter Virtual Teams (Poole and Zhang, 2005), following an equally useful chapter titled Group Identity and Self-Definition (Abrams, Frings, & Randsley de Moura, 2005). That chapter might very well have been the basis of my thesis introduction.

It outlines the concepts, citing research evidence for a psychology of groups as distinct from a psychology of individuals. The evidence for an evolution of group psychology, originating in the personality psychology of the individual is well argued. For example, it begins to answer key questions that distinguish “group think” as cognitions: The determinants of whether people behave as individuals or members of the group; how people prioritize the group over themselves; the basis of social influence within the group; explanations of intergroup prejudice; changes in self-concept in conditions of social identity; the relationship between intergroup and intragroup behavior; and motivations of group behavior.

I don’t think I can study the Abrams, et. al chapter enough, as it relates to my own work. The authors get into the Social Identity Theory (SIT) of Tajfel and Turner (1979), whom I have cited independently in my own thesis. The premise of SIT is that personality and social identity exist on a continuum. Part of SIT rests on the acceptance that a cognition of membership is an identifying characteristic of the group. Historically, this aspect of group syntality is traceable to the pioneering metrics of Cattell (1948).

The varieties of group research are sufficient to constitute a definitive text on the subject of the emerging science of group psychology, but its groundbreaking content is that which addresses the virtual group. At least, that is my prejudice.

Virtual teams are defined, as are group structures in the context of groups comprising members who never actually meet each other face to face. In addition to tasks and goals, motivators that can be studied in the traditional group setting, technology is a necessary feature of the virtual group. Internet technology, in particular, is the catalyst that makes the virtual group possible, and without it this study would not exist.

References

Abrams, D., Frings, D., & Randsley de Moura, G. (2005). Group identity and self-definition. In S.A. Wheelan (Ed.),
The handbook of group research and practice. (pp. 329-350). Thousand Oaks, CA: Sage.

Cattell, R. B. (1948, January). Concepts and methods in the measurement of group syntality. Psychological Review, 55(1), 48-63.

Poole, M.S. & Zhang, H. (2005). Virtual teams. In S.A. Wheelan (Ed.),
The handbook of group research and practice. (pp. 363-384). Thousand Oaks, CA: Sage.

Tajfel, H. & Turner, J.C. (1979). An integrative theory of intergroup conflict. In W.G. Austin & S. Worchel (Eds.), The social psychology of intergroup relations (pp. 33-47). Monterey, CA: Brooks/Cole.

Wheelan, S. A. (2005). The handbook of group research and practice. Thousand Oaks, CA: Sage.

Virtual Group Psychology: Mechanisms of Herding.

What is virtual group psychology, and why should you be interested? Virtual group psychology is the same as group psychology, which is a subset of social psychology, but extended to the domain of electronic communities (the web). In most cases the principles that apply to traditional forms of group psychology are extensible to virtual group psychology.

One of the sets of principles that work well to explain how individuals behave on the web is the informational social influence model, which now has some representation in virtual group behavior, thanks to Chen (2011). In this paper we obtain a wealth of information explaining herding behavior as group behavior that is related to the interactions of individuals, especially as they may be classified as peers. This is most readily demonstrated in the online auction context, but we should be able to generalize Chen (2011) to explain other groups forms on the web.

Unlike the traditional vis-à-vis group where we can interact with each other on a personal level, the way we interact with each other on the web is necessarily constraint, even funneled through narrow channels of functionality, all made possible with Web 2.0 technology. When you want to leverage Web 2.0 technology to exercise herding and conformist controls, remember that there are only a few basics involved. we are not saying how much is possible, only that in observation, three major material operations reveal themselves.

First, we can see that individuals can express their sympathy or antipathy through mechanisms like the “Like” button, familiar to Facebook subscribers, and other buttons that work similarly. We can include the thumbs-up buttons, for example, and their functional equivalents. Through these very binary state mechanisms, how much information can we really derive about the particular comment being rated? I would say, at best, there are two things we can determine: Does an individual approve of the content, and what ratio of the population approves (or disapproves) the content. Some minor pieces of information may also be analyzable, such as the level of social conformity within the population that drives individuals to follow or be herded into decisions with respect to the content they are viewing. These approval mechanisms are found almost everywhere now, and are as ubiquitous as the next kind of mechanism, the comment.

The comment, and its functional equivalents the review and consumer feedback is of a higher order of information. Through them, we can get a more nuanced sense of sentiment. The longer and more eloquent the comment, the better we can understand the emotional involvement of the individual to the subject under discussion. There is much untapped wealth at present on many dimensions of the commentary form. Understandably, content analysis is fraught with difficulty. There are languages and dialects and grammar rules, there are word counts and phrases and figuring out whether the commenter, on the balance, likes this piece of content or hates it, or is just lukewarm on it. Furthermore, we may also analyze the cumulative effect of multiples of comments and reviews. There may be ways to simplify this process, such as by rating each comment on simple dimensions of “approve” / “disapprove”, for instance, and simply counting the numbers of pro-comments versus against-comments. More elaborate content analysis schemes might reveal who our reviewer base is, whether they are educated or uneducated, for instance, or their nationality or socio-economic class.

A third mechanism of virtual informational social influence is a rating system. Rating systems are very similar in function to the approval / disapproval button, but they provide a more scalar level of information. The rating system is recognizable on most web sites as a star rating, where content consumers can vote for their relative like / dislike of the content by giving it a star rating on a scale of from 1 to 5, or sometimes 1 to 10. These ratings are not binary like the thumbs up or high five systems. We can get the consumer ratings of movies, for example. Interestingly, star ratings are usually displayed as means of the overall ratings, and all this information is usually presented somewhere in the visible vicinity of the overall rating. For example, we may see a bar of 10 stars, where seven stars are shaded, one star partially shaded, and the remaining two stars unshaded, and next to it in parentheses will be some set of numbers like “Ratings: 7.8/10 from 7,220 users” (see: Top Hat on IMDB).

These are the three fundamental mechanisms offered by almost all Web 2.0 enriched websites today: Approval gestures, comments and other compound textual units, and the rating. Their emotion type correspondences are approval, levels of sentiment, and levels of sympathy, in that respective order.

References

Chen, Y. (2011). Auction Fever: Exploring Informational Social Influences on Bidder Choices. Cyberpsychology, Behavior & Social Networking, 14(7/8), 411-416. doi:10.1089/cyber.2009.0355

Herding consumers online.

This morning I found myself saying, “medium” when the woman serving my breakfast burrito asked if I wanted medium or hot picante sauce. If I really know what I want, I will say it unequivocally. But when I’m new to an experience, I find myself opting for the first choice offered, and hardly wait to hear all options. OK, I admit I’m a highly suggestive, conformist kind of person. Perhaps that explains something of my interest in the subject.

I have found myself being herded in a variety of purchasing situations, responsive to proven sets of social stimuli proven to get people to buy things. For instance, I bought a condo once, and at the time I believed I sufficiently weighed the pros and cons. I was also making pretty good money at the time, enough to afford a Florida timeshare. To tell you the truth, I found the congratulatory hoopla embarrassing, and would have preferred the sales team not to announce my decision to the other people sitting around.

So, you’re siting in the quiet privacy of your home, perusing some object on Ebay. A smart phone, let’s say. Having thought about it for some time, you know you want an Android-type phone, but beyond that you don’t really have an informed opinion as to which brand is the better one. You sort by price, and see numerous brand names. Some brands are clearly more expensive than others. The pictures make two different brands look good, but thinking back to the time you bought that PDA that was reputed to have many more functions than the other brand, you are more cautious this time around. After all, the vendor failed to mention that documents created on you PDA were not compatible with Microsoft Word, and were pretty much locked into your PDA without a way to export them to your computer.

Here is a Nokia smart phone, and there is another kind you never heard about called Toplux (these were the brand names used in Li-Fen’s (2011) study, where Nokia had a high mean brand recognition of 4.73 to Toplux with a mean of 1.65). What do you do next? You will probably read the reviews on the cheaper phone you never heard about. You want to read only the positive reviews, but then again, you know that would only be fooling yourself. So you find yourself giving the negative reviews a fair reading, before looking at the glowing reports.

Who do you trust? Are the positive reviewers sincere, or are they plants? Sometimes you can get a sense of who’s who from the kind of language being used. The paid reviewers have a certain style of writing, don’t they? Or do they? To be honest, I’ve read many, many paid-for product reviews by people who obviously speak English as a second language, and were perhaps not-so literate in their own language. Another way of telling might be to follow that one person’s comment history. Hello, what’s this? For some strange reason this reviewer has also written reviews on this brand or vendor on other product pages.

You also look at product ratings, and overall, 5,000 ratings indicate the Nokia phone is pretty reliable. Toplux, however, only has a few dozen votes, and they average out to about a point below the Nokia phone. Surely 5,000 votes could not all be stooges, even if a few might be. Your mind is made up, and you bid on the Nokia phone. Better safe than sorry.

In online bidding studies Yi-Fen (2011) has cautioned that people can sense when they are being manipulated into a purchasing decision by artificially inflated positive rating numbers and product reviews. The key is in numbers that deviate too unrealistically from what consumers expect. For example, if the Toplux phone had 5,000 positive reviews, but the Nokia phone only a handful the consumer might be wary of Toplux. People like honest-looking reporting, and might purchase Toplux if it had just the right amount of positive reviews and positive rating. We might not expect Brand X to be perfect, but a 7 out of 10 for a 20% savings? Hmmm, we might buy it.

A few well placed negative reviews can tank a sale. For example, when searching for a replacement battery for my notebook, I knew that an after market brand would be cheaper. But are all lithium-ion battery packs the same? This is certainly not an expert opinion, but it looks like they are not the same. You have rebuilds, and even with factory new you have different life expectancies. Should I pay $50 for a battery, or $80? If it is the same battery, I should buy the $50 one. But what if they are not the same battery? Consider two reviews. The first is a 4-star review, saying the battery was shipped promptly, and arrived in good shape. The second review, though, while acknowledging the prompt shipment, was written a month after purchase, and the poor consumer discovered that her battery only lasted one month.

Many of us have heard of the restaurant review site Zagat (rhymes with “Cat in the Hat”). Zagat is an excellent example of crowd sourcing, where amateurs contribute to a body of knowledge. In this case, patrons of various levels of expertise will write reviews on restaurants they have visited. Some reviews are of exceptional quality. For instance, I have never heard of a Sommalier, and to see it used in connection with a server of wine gave me the impression that this patron must really know what he’s talking about. Equally, we hardly trust obsequious sycophancy when it comes to restaurant reviews. Don’t the pros find some fault with even the most expensive restaurants? So, too, do we expect to read something negative, even if we may not understand it (“The wine pairings were mediocre, and the Sommalier tried to outsnob us with obscurata”).

References

Yi-Fen, C. (2011). Auction Fever: Exploring Informational Social Influences on Bidder Choices. Cyberpsychology, Behavior & Social Networking, 14(7/8), 411-416. doi:10.1089/cyber.2009.0355

Herding and informational social influence

The term herding is titillating to the marketing industry, and comprehensible even without studying the phenomenon. Herding evokes images of the Old West, perhaps cattle being driven in unison to market across the plains, or the American Indian driving bison over a cliff, before the days of the railroad. Informational social influence is a bit more complicated, but we may think of it as the engine that drives herding and its related phenomenon, mimicking.

Herding may be thought of as manipulating visual online cues to coax populations into doing some desirable behavior, especially making some kind of product approval or purchase. Even where a single item may be on sale, the ability to control herding may realize greater gains to the merchant if it is the difference between few bids and many bids.

Yi-Fen (2011) operationalized what is intuitive about herding by testing bidding behavior against brand recognition. By a statistically significant margin ( F(2, 117)=6.355, p=0.002 ) 6 groups of 40 subjects clearly favored bidding on better known brand names, as opposed to unknown brands. When it came to bidding on a product where a high number of bids were present, the subjects tended to think critically. For instance, the subjects were not easily misled into believing that a lesser known product (a cell phone) earned a higher numbers of bids than its better recognized counterpart.

What makes the difference in experiments like Yi-Fen (2011) is informational social influence. Informational social influence is what makes brand recognition, after all. It is the sum effect of artifact: peer product reviews, ratings, comments, and the quantity of information shared about the product. Lesser known products—a cell phone, for instance, will not have as much social pressure, and so the bidder may be confused or immediately on guard against unrealistic bid numbers.

We might think of mimicking as a factor of herding. Control the tendency of people to mimic their peers, and you control herding to an extent. Mimicking behavior also works under the principle of informational social influence, where precedence is given to the decisions peers make for us. If it seems that different forms of informational social influence can potentially work against each other, they do. Yi-Fen (2011) found that a critical number of bids in a “fixed” auction had to appear organic, or they wouldn’t be convincing.

The control of perceived risk in online auctions mediates mimicking behavior. In addition to informational social influence, Huang and Min (2007) identified two more factors that mediate online bidding behavior: perceived risk and product involvement. The hypothesis that perceived risk would influence online bidder’s conformity was supported by a bidding experiment, which also used differing cell phone brands. It is easier to induce mimicking behavior in a subject who is more highly involved with a product (Huang & Min, 2007).

References

Huang, C.-H., & Min, J. C. (2007, January 1). A research note of online bidders' conformity. Social Behavior and Personality, 35(8), 1033-1034. doi:10.2224/
sbp.2007.35.8.1033

Yi-Fen, C. (2011). Auction Fever: Exploring Informational Social Influences on Bidder Choices. Cyberpsychology, Behavior & Social Networking, 14(7/8), 411-416. doi:10.1089/cyber.2009.0355

Group cohesion: from pandemonium to uniformity.

I recently found out that a term used to describe a flock of parrots is a pandemonium. Living in south Florida I occasionally get to see the North American wild parrots, called Quaker parrots. They like to invade the large sea grape trees, where they become all but invisible. If it wasn’t for their raucous cacophony, which is indeed pandemonium, you would not know they were there. I have tried to photograph them with my little compact camera, but my photos are not nearly as good as the one below. As they fly, the pandemonium of parrots is more like a ball of birds than an organized v-formation. They are loosely grouped together.

At what point, I have asked myself, does a loosely confederated ball of birds become a group? What are its parameters? Thinking about this in an informal way, we can say that there must be some mathematical cohesion, because all the birds light from the tree around the same time when spooked, they launch into a similar trajectory, they occupy a confined region of space in flight, and they end up in the same place, the sea grape tree. Not to mention they belong to the same species, and very likely constitute a single extended family.

On the other hand, look at the organized, geometric structure of a flock of pelicans in flight (which I have also tried to photograph, but my photos are not as good as this one). We can discern that the pandemonium of parrots and the pod of pelicans are both groups, having characteristics that all groups must have. Clearly, though, the two groups are structured differently. One is an amorphous blob, the other a linear stratification.

photo by auntjojo on Flickr. “This photo was taken on August 20, 2007 in Oceanside, California, US, using a Nikon E4600. ”

This is hardly a lecture in bird psychology, but rather the structure of flocks of birds should used as a metaphor to illustrate the continuum of group variety and homogeneity. Can we generalize the visible organization of these images to people groups?

Further investigation into three parameters of the virtual group

As the present research is mostly a theoretical, top-down discussion of virtual group behavior, little emphasis has been given to empirical research. This area, though, should be fertile ground for numerous ways of establishing relationships, proving or disproving hypotheses, and quantifying virtual group phenomena.

Some experiments have yielded interesting personality effects, such as the online disinhibition effect (Suler, 2004). This is a phenomenon that many who engage in discussions with others in the online world experience, where people let loose and say whatever they want, irrespective of the protocols and polite conventions normally found in the vis-à-vis context.

Much of the research into conformity and online (or virtual) group behavior comes from the world of marketing. There, understanding and predicting how people will behave can translate to big capital profits or losses. Two terms stand out in discussions of online behavior. One is crowdsourcing, originally conceived by Jeff Howe (2006) in Wired Magazine, herding, and mimicking.

We notice transitions in research that predate the World Wide Web. Long before Internet buying organizations have been looking for ways to liberate money from populations. Tactics often incorporated perceived compliance to group norms, as in Reingen’s (1982) experiment which used the informational social influence hypothesis to induce people to donate money to some cause.

Social networks are enabling new paradigms in how businesses interact with their customers and prospects. The European grocery giant Tesco has engaged in research to find out the usefulness of Twitter as a product support or company-consumer communication platform (Davies, 2010). the public sector is not behind the eight ball on the power of social networking, either, as is attested by the 200,000 citizens who participated in a White House town meeting in 2010 (). With crowdsourcing we have all kinds of related behaviors, such as online helping behavior. Here, manufacturers and retailers want to draw upon the informal expertise of consumers. According to Howe, the amateur status of crowds is one aspect of crowdsourcing; another is the willingness of large numbers to pool their expertise without consideration for recompense. These characteristics of the free and willing resource inherent in crowdsourcing have been amply observed by Chu (2009), and show promise to the future of product support, albeit with limitations and caveats. Chu’s (2004) observations really work within the context of online communities, where personalities join and impart something of themselves, as well as abide by some rules of engagement, as people do in traditional vis-à-vis communities.

In addition to helpful behavior, our online peers also influence us in other ways which beg for a certain caution and cynicism. Mimicking behavior is where we see decisions our peers have made online, and we react similarly. Marketers like mimicking because they want to know if they can get us to mimic purchasing behavior. Huang and Min (2007) experimented with online purchasing and discovered that purchases of electronic and apparel commodities can indeed by controlled and predicted by visuals cues that show us whether or not our peers have been buying Product A or Product B.

References

Chu, K.-M. (2009, March). A study of members' helping behaviors in online community. Internet Research, 19(3), 279-292. doi:10.1108/10662240910965351

Davies, J. (2010, April 22) Tesco turns to Twitter to talk to customers. The New Media Age. 5.

Howe, J. (2006, June). The rise of crowdsourcing. In Wired. Retrieved July 1, 2010, from http://www.wired.com/wired/archive/14.06/crowds.html

Huang, C.-H., & Min, J. C. (2007, January 1). A research note of online bidders' conformity. Social Behavior and Personality, 35(8), 1033-1034. doi:10.2224/
sbp.2007.35.8.1033

Reingen, P.H. (1982). Test of a list procedure for inducing compliance with a request to donate money. Journal of Applied Psychology, 67(1). 110-118.

Suler, J. (2004). The online disinhibition effect. CyberPsychology & Behavior, 7(3). Mary Ann Liebert

Twitter presents Townhall @ the White House. (2011, July). Retrieved July 19, 2011, from Twitter website: http://askobama.twitter.com/

Virtual groups learn. Comparison of group and individual learning, and is brainstorming meaningful?

30 years after Cattell’s (1948) introduction of group syntality (syntality is to groups what personality is to the individual), Shaw (1976) investigated group learning phenomena, and compared it to individual learning. As remarkable as it is to consider that groups learn, it is equally remarkable that groups learn at a faster rate than individuals.

The ability of groups to learn is compounded by a few factors, and possibly the most important is that group learning is generally constrained by the most knowledgeable member of the group. That is not, however, written in stone. What is different about the group from the individual (and the steady, incremental advances of civilization are an attestation of this) is that group members bring knowledge to the group, and as it is acquired it is kept. Even the least knowledgeable group member may bring some novel piece of information that furthers the group objective. This is kept, adding to the group summum bonum (to borrow one of Jung’s favorite phrases) or “total good.”

One should be able to see that multiple contributors to the common good increase the net quanta of group knowledge because the group benefits by multiple individuals, whereas the single individual must bring knowledge to himself, and by himself, before he can impart that knowledge to the group to which he is a member. The group benefits by a progressive and directional collection of information, much resembling the growth of the “common wealth” or crown jewels, as it were.

One of the most taken for granted, yet as it turns out suspect methods of group development is the brainstorming session. According to Osborn (1957, in Shaw, 1976) brainstorming traditionally follows three rules: 1) Ideas are expressed freely and without regard to quality; 2) Before any idea is evaluated all ideas must be expressed; and 3) Elaboration of any one’s ideas is permitted and encouraged.

Some studies of Shaw’s time showed that brainstorming did produce significant group gains, while other studies cast reservations on brainstorming as any more useful to group genius as other methods. Sawyer’s (2007) exploration of group genius seemed to agree with the latter finding.

Sawyer proposed a concept of group genius, and this quality peculiar to groups was the source of innovation. To underscore that some innovations do not arise outside of group collaboration, Sawyer pointed to jazz improvisations. such improvisations were something of a metaphor for other types of groups, as well. Even if one had an insight while alone, it could usually be traced to some prior collaborative event. All historically significant contributions were a product of collaboration. To Sawyer, lone or singular genius is a myth, and genius of individuals is relative to a larger social context.

Sawyer offered seven characteristics of creative teams that vector with Osborn’s (1957) three rules of brainstorming on at least one of these rules: Innovation emerges over time; deep listening by collaborative teams; group members build upon each others’ ideas; ideas become meaningful only after all are put on the table; surprising questions emerge; the inefficiency of innovation; and innovation emerges as a bottom-up process.

Aside from any technique of collaboration, what we might want to consider at this point is the simple recognition that groups have a life and personality of their own, and that this life comprises a function of the composition of membership, group purpose and goals, and members’ specific contributions rather than individual members’ gross global attributes. For example, as stated in a previous paper, we don’t care so much that group members come from diverse philosophical or ideological backgrounds, but are more interested, in terms of the life of the group, the net effect each member has upon the development and strengthening of the group.

References

Cattell, R. B. (1948). Concepts and methods in the measurement of group syntality. Psychological Review, 55(1), 48-63. doi:10.1037/h0055921

Osborne, A.F. (1957). Applied imagination. New York, NY: Scribner.

Sawyer, K. (2007). Group genius: The creative power of collaboration. New York: NY: Basic Books.

Shaw, M. E. (1976). Group dynamics: The psychology of small group behavior (2nd ed.). New York: NY: McGraw-Hill.

Explanation of Cattell’s Syntality with Respect to the Virtual Group

Raymond B. Cattell innovated the science of social psychology by formulating measurable parameters of personality. Cattell took the mytho-poeticism of the psychoanalytic schools out of the discussion and focused on what was measurable about the personality. In fact, Cattell identified 16 factors of personality, and the 16PF Questionnaire continues to be regarded as an important instrument in personality profiling.

Cattell reasoned that groups can also be defined according to factors, much like the personality can. Groups, then, have personality. The term Cattell used to describe the group’s personality was syntality. As Cattell sought a meaningful framework from which predictions of group behavior could be reliably drawn, The present researcher extends that rationale to virtual group phenomena. Therefore, it is reasonable to generalize Cattell’s principles of group dynamics to virtual groups. After all, virtual group creation, purpose, and development have many of the same bases as the traditional groups. Whether or not a culinary group, for example, exists after hours in the kitchen of a school cafeteria or as an online entity really doesn’t change its fundamental purpose. The differences are in its construct and mechanics. Whatever it is that makes the virtual group different from the traditional group, that is what we want to explore here.

Cattell said of personality, “That which will predict behavior in any given situation.” Since syntality was the personality of the group, a strong knowledge of syntality should give group behavior predictability. The reasons why we should want to be able to predict group behavior are numerous, and they cover everything from marketing to national security. The 2010 Egyptian Revolution was coordinated through the virtual group environment of social networking, for instance. It is known that anti-social and terrorist groups reinforce their goals by means of websites, among other resources, and with the ability to create product pages on social networks groups may form around commercial organizations as well as products.

To predict behavior, said Cattell, we define the group as a profile of indices, while another set of indices defines the situation, and as syntality vectors with situation, and voila, we should be able to predict what the group will do next. Of course we’ve skipped over the important details that prove this can actually be done, but in principle the more factors we identify belonging to the group and its developmental trajectory, the more likely we are to be right about what that group will do next.

References

Cattell, R. B. (1948). Concepts and methods in the measurement of group syntality. Psychological Review, 55(1), 48-63. doi:10.1037/h0055921

Virtual Groups and their Distinction from Traditional Groups

“Many crowds are not fortuitous gatherings, but are brought together by the common interest of their members in some object or topic.” – Wm. McDougal, The Group Mind.

As McDougal noted in 1927, a group is not merely a dissociated collection of people. when you put multiple bodies into a room, can you call it a group? McDougal might have said no, but group boundaries are dynamic enough that we can “dial” them in or out, depending on what level of analysis we are looking for. It has to do with commonalities. We can say that if a bunch of otherwise unacquainted individuals are collocated within a room, they might be considered a low order group on the basis of their common geo-spatial proximity. If that is all they share, then the group identity is rather weak.

Group identity is strengthened as we find more commonalities among them. Perhaps they live in the same rural countryside, and gather together once a years to decide on fiscal matters. After the town meeting, each goes his and her separate way, never having to see each other until next year. Not only is the group identity strong in this case, but it is transient. We can say that a common purpose, location, and time define this group. Take away the temporal connection, you take away the collocation, and there is really no group identity.

Now, let’s look at a nearly complimentary situation on the internet. Let’s take the recent Libyan Revolution, and we have many sympathetic parties in the U.S., and also in European countries. Many who have been following such Facebook groups as the Libyan Youth Movement may also share other interests, such as Americans Against Islamophobia. or Democracy Now!, or even perhaps have conservative affiliations. A large number of individuals with incompatible group identities elsewhere may vector at a common group objective here. If they have nothing else to talk about, and perhaps it is well that they don’t they can each further the common goals of the group by sharing and propagating news items relating to the subject of the group; or leaving positive, affirmative comments to each other or on news items that reinforce group values; or by exhibiting approval of what other group members have to say in some way, either through a Facebook Like button, or a thumbs-up or high-5 button.

One can even exhibit disapproval, if it is perceived that a particular message is antithetical to the group’s common goals. In fact, group cohesion can be reinforced with both approval and disapproval mechanisms.

Three Factors that Define the Virtual Group

Part of the excitement in virtual group research at the time of this writing is actually a deficiency in the research, that its many explorations to date have been peripheral to the central questions of just what constitutes a virtual group. We have an established body of literature on such subjects as crowdsourcing and herding, which are important in commercial circles. At the time of this writing, this blog entry represents the seminal paper on the three factors that describe a virtual group.

Without further ado, let’s introduce the three factors that can be used to determine the presence of the virtual group.

The first is content propagation. Virtual groups, i.e., groups that form around content found on the World Wide Web will be known because members of the group will propagate content among each other, distributing and redistributing certain content such as videos and news stories. It should occur to the reader that certain social networking systems, especially Facebook figure prominently in the facilitation of virtual groups.

The second defining factor of the virtual group is content approval. This can be operantly defined in terms of “Like” buttons, thumbs-up and “high five” buttons. Again, Facebook comes to mind, but we see these features on many web sites where content consumers can express their approval or disapproval of content. Multiple factors are at play when an individual decides to click an approval or like button, and you can read a more extensive treatment of the phenomenon of social conformity’s role in content approval here.

The third defining factor of the virtual group is sentiment. the way we operationally define sentiment is through a web page’s comment system. Comments are a more difficult analysis because of the obvious complexity of language. Textual comments, and much more difficult to objectify video comments must be analyzed for content, though we can make effective rudimentary categorizations of either “positive” or “negative” affect. In short, this means the person likes or doesn’t like the article they just read, or the video they just watched.

That’s it, three parameters of the virtual group: Content propagation; content approval; and content sentiment.

Now it is fair to be skeptical that these parameters and no others define the virtual group. But, this researcher isn’t saying a group is all about these things. The claim that is being made is that we can know when we are looking at a virtual group, and indeed we can detect the emergence of the virtual group (a subject for another paper) because these three factors will be found.

Traditionally we have all kinds of groups that are so ubiquitous as to need no deeper explanation. But traditional groups can be known by some key physical features that don’t exist in the virtual context. for one thing, people have to get together in the same physical space at the same time. Traditional groups, then, exist synchronously, here people gather together in the same geophysical space contemporaneously. For example, try to imagine a municipal meeting, church meeting, or PTA meeting where people assemble in various disparate locations, some at 8 o’clock, some later, some perhaps the next day. We might say about that scenario that no such group exists.

These same constrictions don’t apply to the virtual group. Virtual groups can exist independent of time and space. Colocation is not a requisite, and group members may assemble asynchronously, i.e., at separate times from each other.

Although certain definitions of the traditional group are generalizable to the virtual group, e.g. the reinforcing of perceived group ideals and objectives, these can be done without the necessity of its member getting together at the same time and place. Therefore, when we normalize the definition of group to include what traditional vis-à-vis groups and non-traditional virtual groups share, what is left over are the determining factors of the virtual group. That leaves the three definitions of the virtual group, as explained above.

How Social Conformity Shapes the World Wide Web

Web content, as we view it from the viewport of our browser
has a certain orderliness to it. When we go to Amazon.com and view a book, we
form an opinion of that book based on the reviews of that book. But, what is it
that determines which reviews we get to see first?

The superficial answer is that each review and opinion has a
rating, and those reviews and opinions with the most positive ratings are the
ones that appear near the top of Page 1 of any product. We see this formula on
many prominent websites besides Amazon. We see it on Zaggat restaurant reviews,
and we see it on YouTube videos (see example screenshots at end of article).

It is fair to reason that consumer reviews get the high
ratings they do based on a perception of the knowledge writers have of the
subject of their review. But this researcher suspects there is more to high
ratings than a skill with word smithery and product knowledge. To some degree,
and perhaps to a high degree the number of high-fives, thumbs-up, and five star
reviews on any product web page is a function of a phenomenon called social
conformity, of which there are actually several kinds (Kelman, 1958). The type
we are concerned with here is “informational social influence,” which comprises
covert group perception influences on our behavior.

In other words, we may read and like the product review, and
rate it accordingly. There is a component of the population, however, that will
give a rating for other reasons, and those in varying degrees:

Read review, liked review, approved or
disapproved this review according to internal criteria.

Read review, ambiguous opinion about review,
relied on internal and external criteria (e.g. numbers of positive review
approvals) to approve or disapprove this review.

Didn’t read review, no self-opinion about review, but strongly identifies with perceived group that approved or disapproved this review, relied wholly on external criteria to approve this review.

There is a dynamic continuum of factors that mediate the
approvals or disapprovals on any comment, opinion, or review. These range from
completely internalized decisions, where we genuinely like and approve one
consumer’s review, or we are unable to form our own opinion, and so rely on
other people to make that decision for us. If we see that 10 out of 15 people
before us gave this five-star review a thumbs up, for instance, then we will
give it a thumbs up.

Now, let’s say this review is for an expensive New York City
restaurant. You should start to be able to see how the effects of conformity
can generalize to enormous economic consequences. A little bit of social
conformity can result in one particular consumer’s experience being the most
prominent opinion on a product information site.

When we multiply this effect across the Internet to all
kinds of products and services, it becomes apparent that social conformity
drives Web content, and shapes the World Wide Web as we know it today.

Example 1: YouTube videos with high consumer approval numbers are more prominently positioned in video lists.

Example 2: Zaggat restaurant reviews are listed according to how many readers "like" or approve of the review. The more people that approve a review, the more likely a product consumer visiting the restaurant page is likely to see that review.

Example 3: On Amazon, book and other product reviews with the highest numbers of reader approvals will appear more prominently in review lists.

References

Kelman, H. C. (1958, January). Compliance, identification, and internalization: Three processes of attitude change. Journal of Conflict Resolution, 2(1), 51-60. doi:10.1177/002200275800200106

Toward A Psychology of Virtual Groups

Traditional Vis-à-vis group psychology is well established in the literature. Marvin E. Shaw's 1971 book Group Dynamics: The Psychology of Small Group Behavior is a seminal text in the formation and maintenance of small groups. My thesis of virtual group mechanics expands on Shaw, generalizing certain basic principals to the realm of virtual, i.e., Web groups. How do people coalesce on the Web? Why do they group together along delineations of politics, religion, aesthetics, and general interests? How do you know when a group has formed, and what are the steps involved in group formation? These are the introductory questions of this new psychology.