« Guild Recruitment Location | Main | Mapping Social Networks »
October 03, 2005
Grouping and Guild Recruitment
Server Sample: RP (High), PvE (Medium), PvE (High), PvP (High), PvP (High)
Sampling Period: 8/07/2005 12:00 am - 8/14/2005 12:00 am
Sampling Resolution: ~12 minutes
Parsing Method: The sample unit is each unique character. Each character was tracked across the server logs. Total playing time, lowest observed level, highest observed level, guild affiliation, and zones seen in were parsed.
Data Filter: None
Sample Size: 148,846 characters
There was another aspect of guild recruitment that we were interested in exploring. We wondered whether most guild recruitment events could be attributed to grouping. In other words, do unguilded characters get recruited into guilds primarily because they group up with someone who is in a guild? Or do most guild recruitment events occur outside of a grouping situation?
To address this question, we built upon the guild recruitment location data:
Grouping Status During Guild Recruitment Event: For each guild recruitment location, check whether the character is in a group or not when they are observed to be in a new guild.
Here are the raw percentages for each zone type. Given that most guild recruitment occurs in main cities and outdoor zones, this implies that most characters are recruited into a guild when they are not grouped.
To get a more direct number, we took the guild recruitment location data and multiplied it by the percentage grouped data. Of all guild recruitment events, about 39% occur when a character is grouped, and 61% occur when a character is ungrouped.
Posted at October 3, 2005 10:42 AM
Trackback Pings
TrackBack URL for this entry:
http://blogs.parc.com/cgi-bin/mt-tb.cgi/47
Comments
This seems like fairly meaningless data. Since you can invite someone to a guild from anywhere in the world there's no particular reason to care where their character is at the time. You may have simply captured where folks of all levels spend their time, perhaps weighted by what levels of folks join guilds more. In fact, you could probably correct for this with your other data. Does the above distribution tell us anything about guilding specifically?
Also, given administrative details with many guilds (approving new members, etc.) even if grouping is what gets someone into a guild, there's no reason to expect them to be grouped at the moment they get the invite.
Posted by: Brent Michael Krupp at October 4, 2005 09:40 AM
Brent - I think that's what we were trying to test. Do most guild invites happen on-the-fly during grouping or not? This is the hypothesis that the data might or might not falsify. The data could have turned out the other way, but we wouldn't know until we actually did the analysis.
Also, the grouping data point is separate from the location data point. We simply presented both to cut the data in different ways. What the data shows is that guilding does not seem to mostly occur on-the-fly from grouping events (which very well could have been the case).
We explore the data to test hypotheses. Hindsight is 20/20 and everything seems more obvious once the data is laid out, but to claim that the data exploration itself is meaningless is a little strange. You seem to be arguing that personal convictions should trump actual data and analysis. In empirical research, we tend to believe the opposite.
Posted by: Nick Yee at October 4, 2005 02:33 PM
The data gathering is awesome, and I love this site, don't get me wrong. I guess I just misinterpreted the meaning of the above table. Standing alone, it just didn't seem to say very much. I suppose the data will be useful when viewed in the context of other data and/or other analyses of this data.
Posted by: Brent Michael Krupp at October 5, 2005 09:58 AM
weh1f8o4ogjwo941
Posted by: Kirt Noble at November 12, 2008 02:18 PM
weh1f8o4ogjwo941
Posted by: Jeffrey Orr at November 13, 2008 12:11 AM