|
|
I rated opinions by 100 strangers; waited a week to see how many reciprocatedMay 15 '01 Write an essay on this topic.
Popular Products in Books
The Bottom Line Most of the time I've spent on Epinions reading and rating opinions by strangers has apparently been wasted, as far as getting any reciprocity is concerned
Ever since I came to Epinions at the beginning of March, I've felt I was giving out a lot more ratings than I was receiving in return. After two months, in which time I had put 57 people on my Web of Trust and acquired about 40 fans who put me on theirs, I decided it was time to start taking notes on just how frequently a stranger whom I rated took the time to return the favor. I defined "stranger" as someone who was not on my Web of Trust, had not placed me on his WOT, and had not rated any of my last five opinions. Thus, if that person did rate any of my opinions in the near future, it would probably be due to their noticing my name on the list of people who had rated theirs. On Sunday the 6th and Monday the 7th I read and rated opinions by precisely 100 strangers. Obviously I made it an even hundred in order to make it as easy as possible to calculate the percentages. I set up a few rules for my experiment: 1. I would only rate opinions by strangers who had posted at least one new review within the past two weeks. (Most of them posted new ones on Saturday, Sunday, or Monday, in fact, but sometimes I ended up looking at the profile of someone I didn't know who had rated someone else's review a week earlier.) 2. I would only rate opinions which were among the five most recent that a person had done; that is to say, the ones that show up in their profile when you first click on their name. I assumed that those could still be receiving regular checks from the authors, but if I rated something they had done 10 opinions ago, they might never get around to noticing my visit. 3. I would rate no more than two of those five most recent opinions, to avoid skewing my experiment by causing someone to become grateful that I had rated five of their opinions in a row as VH. 4. I would not add any of the writers I examined in this experiment to my Web of Trust until after the experiment was complete. If I found someone whose work seemed particularly interesting, I made a note in my log of people rated to consider them for the WOT in the future. Adding someone to the WOT would create a higher-than-average chance that they would take notice of me and at least read and rate one or two of my own book reviews as a way of saying thanks. I wanted to treat everyone as equally as possible. 5. I did not tell anyone I was doing this experiment. I didn't want people I told to decide to earn brownie points with me by rushing to rate me when they otherwise would not have bothered. 6. I decided that after I hit the 100 mark, I would wait for a full week before checking my most recent 20 opinions or so to see who had returned the favor after I read and rated them. After all, a lot of us probably only spend time on Epinions once or twice a week, but if someone doesn't react to my visit within the first seven days, it's not terribly likely he will do so later on. 7. I was going to keep track of what sort of ratings I gave everyone, including SH and NH. I suspected it was unlikely that anyone who received NH or SH from me would be inclined to rate one of my pieces in return, but I wanted to know if they did. And I wasn't going to only count people who received VH from me, nor was I only going to rate opinions which had already received an average rating of "VH" from other people and thus might be worth my trouble. I wanted an unbiased "representative cross-section" of the Epinions community, instead of arbitrarily excluding someone from consideration. In most cases where I gave a low rating, I left a comment explaining why I had done so and inviting the recipient to e-mail me if he fixed up his opinion and thought I'd like the new and improved version better. So I went and did it. In retrospect, I could have chosen my time better. I started doing this two days after the new Mummy movie hit the theaters, with the result that I ended up reading an awful lot of Mummy 2 reviews that popped up in the Just In section. I didn't even try to count, but I'd estimate I read over 30, possibly over 40. (Do you see what sacrifices I am prepared to make to research an essay properly?) I should mention that most of the opinions I rated were in the Books category or the Movies category, with a sprinkling of other subjects mixed in for variety. I spent most of my time at these two places: http://www.epinions.com/book-Books-All/show_~recent_content which shows several pages of the most recently posted book reviews, and http://www.epinions.com/mvie/show_~recent_content which serves the same function for recent movie reviews. Saved me the trouble of spending all my time in the general "Just In" area trying to ignore reviews of cars and cosmetics and heavy metal and other items of very little interest to me at the moment. So: Out of these 100 strangers who suddenly received ratings from me (usually two, but occasionally only one), how many noticed my surprise visit and took the time to click on my name and find something to rate within my last 20 opinions? Sixteen! Two of those sixteen actually added me to their Web of Trust, and the number of ratings which I received from individual strangers during the week-long experimental period ranged from 1 to 8, but I considered it a victory whenever any one of the hundred strangers rated at least one of my opinions, and that means we had a victory rate of 16%. This disturbs me. It suggests that 84% percent of the time when I rate a couple of opinions by someone I hadn't noticed before, I get absolutely no return on my investment. If that applies to me, I suspect it applies to the community in general. It leaves me with the nasty realization that I probably would have been better off to spend more time in these past two months just writing extra opinions beyond the number I actually did, and far less time hunting down new people to read and rate in the overly optimistic hope that a fair number of them would do the same for me later on. All those hours I wasted, except for the fraction of that time which was spent reading people who actually ended up on my Web of Trust, of course. I don't regard those particular moments as wasted, but I sifted through an awful lot of chaff to find a few grains of appetizing wheat! As an old Kingston Trio song put it, "You've got to give until you get / I'm the one who ought to know!" So here's my advice to anyone who is wondering how to increase the number of hits you receive: be sure to reciprocate every single time someone actually does read and rate one of your recent opinions! That way there is a higher-than-average chance that this person will return the favor by coming back and reading and rating another of your opinions later on, and then you can do the same for him, and eventually you've established something resembling a friendly relationship for mutual convenience. Expecting him to become a regular provider of pennies for your opinions when you are not doing a single solitary thing to benefit him is an exercise in wishful thinking, unless your writing is so brilliant that everyone who reads one of your opinions is hypnotized into instantly adding you to this Web of Trust and faithfully keeping an eye out for every new opinion you post in the future. This is so obvious that I never needed anyone to explain it to me in so many words when I arrived, but my results suggest that 84 percent of our population haven't quite grasped the point yet. Of course, there may be other factors in play. Some of those 84 percent may not even connect to the internet as frequently as once a week to check on how their opinions are doing. Some of them may have looked at my profile and failed to find anything in my five most recent opinions that appealed to them at all (although I tried to make sure I had a fairly eclectic collection of book reviews on display when I started the experiment, to broaden my potential appeal). Some of them may have read one of my reviews but failed to rate it for whatever reason, in which case I at least got a penny for my trouble (oh joy!) but without knowing who helped me earn it. But my strong suspicion is that the vast majority of those 84 strangers simply never bothered to check their lists of raters to see what new faces had popped in lately, or checked in an idle sort of way but never felt the need to do anything about it. I do not know for certain that I have always returned the favor of reading and rating when I see that a "stranger" has suddenly done that for one my opinions, but I know I have consciously tried to keep an eye out for such occurrences. If my success rate in reciprocating someone's first visit (or a return after not having rated anything of mine for at least a month) has not been one hundred percent, I am certain it has been something much, much closer to 100 than to 16. I only started here in early March, and I was not inclined to let any possible fans slip through my fingers because I ignored them the first time they dropped in. Ironically, some of the people who have not yet reciprocated my visits of last Sunday and Monday are probably going to end up on my Web of Trust anyway. After all, as I stated in a previous opinion, I put people on my WoT because I look forward to reading their future opinions, not because of anything they have already done for me or which I demand they do for me in the future. But even if I add 10 of the 84 on that basis, we still have 74 people out of the 100 who are not likely to receive any more visits from me for a long time to come. I have learned that reading and rating them doesn't give me anything for my trouble, neither by greatly impressing me with the style and content of their opinions, nor by showing signs of developing into a mutually profitable arrangement. Note: Having kept a log of all my readings and ratings of strangers for 100 people in a row, I found the habit was growing on me. Since last Tuesday, I have added at least another hundred records to the log, and I intend to continue the practice indefinitely. I will probably post an update when I hit the 500 mark, to give you all the latest statistics on how much reciprocity there is in such a large assortment of Epinionators. Hopefully the group average will be a bit higher than sixteen percent at that point, but I'm getting cynical on the subject.  |
| Read all comments (36)|Write your own comment |
