News:

The AARoads Wiki is live! Come check it out!

Main Menu

Total time logged in:

Started by thenetwork, January 25, 2018, 11:58:16 PM

Previous topic - Next topic

Thing 342

#50
Someone mentioned statistics, so I decided to run a sample analysis based on the information posted to this thread:

First I calculated a time-per-post metric `tpp` that measured a user's average number of minutes of forum browsing per post:

              username posts time        tpp
22                 bm7     4   12 180.000000
21             DeaconG   655  947  86.748092
20                   1  4594 6227  81.327819
19            slorydn1   907 1016  67.210584
18            Thing342  1112 1096  59.136691
17       CNGL-Leudimin  1910 1772  55.664921
16              Buck87   613  544  53.246330
15    adventurenumber1  1107  958  51.924119
14           busman_49   237  162  41.012658
13     Max Rockatansky  5796 2604  26.956522
12          thenetwork  2085  768  22.100719
11             webny99   484  163  20.206612
10           Scott5114  7150 2323  19.493706
9      jp the roadgeek  2017  609  18.116014
8  TheArkansasRoadgeek   646  192  17.832817
7              roadman  3061  881  17.268866
6              1995hoo  8678 2333  16.130445
5               vdeane  8581 1955  13.669735
4        jeffandnicole  7912 1762  13.361982
3              brandon  9039 1866  12.386326
2        Hurricane Rex   677  105   9.305761
1                Henry  3810  444   6.992126

(The formula is simply the number of hours spent on the forum divided by number of posts)

Then, I wondered what variables (other than the obvious) might influence this metric, so I gathered up people's listed ages and their user ID number (serving as a rough proxy for how long they've been here, lower -> older). Using that, I calculated a correlation matrix:
Quote
            userid         age               posts            time              tpp
userid  1.0000000 -0.69243768 -0.7023296 -0.11343862  0.4295641
age     -0.6924377  1.00000000  0.3383106 -0.09978523 -0.2951400
posts  -0.7023296  0.33831064  1.0000000  0.51151575 -0.4057392
time    -0.1134386 -0.09978523  0.5115158  1.00000000  0.4184179
tpp       0.4295641 -0.29514000 -0.4057392  0.41841791  1.0000000
The results of this experiment weren't that surprising:

  • User ID is highly negatively correlated with number of posts and age; meaning that people who've been here longer tend to have more posts (obviously) and tend to be older (also fairly intuitive). There's also weak positive correlation between account age and time-per-post, but it's not strong enough to draw any strong conclusion from.
  • Age doesn't appear to be a very significant variable otherwise, only barely negatively correlated with TPP.
  • Total number of posts has a mild positive correlation with time spent on the forum, which should also be obvious as to why.
Finally, here's a basic report on the data:

     userid           age            posts             time             tpp         
Min.   :   33   Min.   :17.00   Min.   :   4.0   Min.   :  12.0   Min.   :  6.992 
1st Qu.: 2052   1st Qu.:19.25   1st Qu.: 660.5   1st Qu.: 469.0   1st Qu.: 16.415 
Median : 8566   Median :31.50   Median :1963.5   Median : 952.5   Median : 21.154 
Mean   : 7397   Mean   :32.22   Mean   :3230.7   Mean   :1306.3   Mean   : 40.459 
3rd Qu.:12658   3rd Qu.:42.75   3rd Qu.:5495.5   3rd Qu.:1842.5   3rd Qu.: 55.060 
Max.   :14144   Max.   :56.00   Max.   :9039.0   Max.   :6227.0   Max.   :180.000 


In summary, the only real conclusion that I can draw is that tenured members of the forum maybe tend to spend more time researching the boards before composing a post. However, this is beset by the fact that the total login time variable is somewhat of an unreliable measurement. It's also a small sample size (N = 22), maybe if I had a larger dataset the results would be more concrete. I could always scrape the members list, but I suspect the admins wouldn't be too pleased about that.


hotdogPi

User IDs are weird. From June 10, 2011 to September 6, 2013 (I have no idea why those dates), many non-members were counted as being registered when they were not. This only affects two things: this page and user IDs; user IDs increased for these non-members during this time period. This means that user IDs between 2652 and 12350 are mostly skipped, as they are assigned to these "non-members".
Clinched, minus I-93 (I'm missing a few miles and my file is incorrect)

Traveled, plus US 13, 44, and 50, and several state routes

I will be in Burlington VT for the eclipse.

US 89

Not that I care at all, but apparently I've spent 8 days, 19 hours, and 55 minutes on here.

slorydn1

#53
Thing 342:

Sorry I didn't want to quote your post and then do all the work to delete all the unnecessary stuff for my answer (I just woke up so I'm feeling lazy).

Registration date can be problematic as well. To illustrate I'll use my own experience as an example:

It is true that I registered on March 26th 2012, a Monday after a long weekend of night shift at work. I was researching information about US-70 becoming a full freeway between Raleigh and Morehead City NC and one of the links on Google was a thread in the Fictional section of this forum. Non Members can't (couldn't ?) see the threads in the Fictional section so I registered. It took a full day or 2 before my registration was accepted. By then I found what I had been looking for and kinda forgot about the forum for a while. I was still checking out various sections of AA Roads (pictures, exit guides (etc), but I had basically forgotten that the forum existed. In my defense at the time I was moderating 2 non road related forums so most online "forum time" was spent attending to those duties.

I had rediscovered the forum sometime late 2013, early 2014 but I didn't spend a lot of time here, just lurking in whatever threads non members could read (I wasn't logging in). In fact I had forgotten that I had already registered in 2012. I tried to register in late January 2015 and was told my username already existed. With some correspondence back and forth with Rickmastfan67 I was set up and could log in, and I made my first post the first day I logged in on February 3rd 2015 . That truly is my start date on the forum.


While I am pretty sure I am the only one with that kind of gap between registration date and first meaningful forum activity, there has to be others who have some sort of gap to where their registration date becomes meaningless from a statistical standpoint.


*Note: With my caffeine deficiency I somehow misunderstood the "time" column to mean since registration and not the hours online displayed. Not sure how I could have done that since the whole point of this thread is "time online" but it is what it is. I'm going to leave the post up just in case someone else has a crazy idea to do something in relation to registration date, and I am now going to go drink some damn coffee.....


Please Note: All posts represent my personal opinions and do not represent those of any governmental agency, non-governmental agency, quasi-governmental agency or wanna be governmental agency

Counties: Counties Visited

Big John

Quote from: oscar on January 26, 2018, 11:38:40 AM
My total is embarrassingly high, as if to guilt-trip me into spending less time here.

Why is the total even displayed when we're logged in? I can see administrators making some technical use of that statistic, but how it helps users is to me unclear at best.
mine too.  Too much time on my hands.

hbelkins

My total time might seem impressive; I don't know, and like I said upthread, I'm not a big fan of these penis measurement contests, so I'm not going to post my statistics.

But I wonder if they're skewed because anytime I'm forced to do a new login (which sometimes happens after a browser crash), I always choose the "Stay Logged In Forever" option.


Government would be tolerable if not for politicians and bureaucrats.

US71

Quote from: hbelkins on January 26, 2018, 08:34:27 PM
My total time might seem impressive; I don't know, and like I said upthread, I'm not a big fan of these penis measurement contests, so I'm not going to post my statistics.


OMG! We agree on something.  :o  :-o

If someone really wanted to get anal-retentive,  one could compare time on-line vs number of posts.
Like Alice I Try To Believe Three Impossible Things Before Breakfast

adventurernumber1

#57
Quote from: Thing 342 on January 26, 2018, 06:32:10 PM
Someone mentioned statistics, so I decided to run a sample analysis based on the information posted to this thread:

First I calculated a time-per-post metric `tpp` that measured a user's average number of minutes of forum browsing per post:

              username posts time        tpp
22                 bm7     4   12 180.000000
21             DeaconG   655  947  86.748092
20                   1  4594 6227  81.327819
19            slorydn1   907 1016  67.210584
18            Thing342  1112 1096  59.136691
17       CNGL-Leudimin  1910 1772  55.664921
16              Buck87   613  544  53.246330
15    adventurenumber1  1107  958  51.924119
14           busman_49   237  162  41.012658
13     Max Rockatansky  5796 2604  26.956522
12          thenetwork  2085  768  22.100719
11             webny99   484  163  20.206612
10           Scott5114  7150 2323  19.493706
9      jp the roadgeek  2017  609  18.116014
8  TheArkansasRoadgeek   646  192  17.832817
7              roadman  3061  881  17.268866
6              1995hoo  8678 2333  16.130445
5               vdeane  8581 1955  13.669735
4        jeffandnicole  7912 1762  13.361982
3              brandon  9039 1866  12.386326
2        Hurricane Rex   677  105   9.305761
1                Henry  3810  444   6.992126

(The formula is simply the number of hours spent on the forum divided by number of posts)

Then, I wondered what variables (other than the obvious) might influence this metric, so I gathered up people's listed ages and their user ID number (serving as a rough proxy for how long they've been here, lower -> older). Using that, I calculated a correlation matrix:
Quote
            userid         age               posts            time              tpp
userid  1.0000000 -0.69243768 -0.7023296 -0.11343862  0.4295641
age     -0.6924377  1.00000000  0.3383106 -0.09978523 -0.2951400
posts  -0.7023296  0.33831064  1.0000000  0.51151575 -0.4057392
time    -0.1134386 -0.09978523  0.5115158  1.00000000  0.4184179
tpp       0.4295641 -0.29514000 -0.4057392  0.41841791  1.0000000
The results of this experiment weren't that surprising:

  • User ID is highly negatively correlated with number of posts and age; meaning that people who've been here longer tend to have more posts (obviously) and tend to be older (also fairly intuitive). There's also weak positive correlation between account age and time-per-post, but it's not strong enough to draw any strong conclusion from.
  • Age doesn't appear to be a very significant variable otherwise, only barely negatively correlated with TPP.
  • Total number of posts has a mild positive correlation with time spent on the forum, which should also be obvious as to why.
Finally, here's a basic report on the data:

     userid           age            posts             time             tpp         
Min.   :   33   Min.   :17.00   Min.   :   4.0   Min.   :  12.0   Min.   :  6.992 
1st Qu.: 2052   1st Qu.:19.25   1st Qu.: 660.5   1st Qu.: 469.0   1st Qu.: 16.415 
Median : 8566   Median :31.50   Median :1963.5   Median : 952.5   Median : 21.154 
Mean   : 7397   Mean   :32.22   Mean   :3230.7   Mean   :1306.3   Mean   : 40.459 
3rd Qu.:12658   3rd Qu.:42.75   3rd Qu.:5495.5   3rd Qu.:1842.5   3rd Qu.: 55.060 
Max.   :14144   Max.   :56.00   Max.   :9039.0   Max.   :6227.0   Max.   :180.000 


In summary, the only real conclusion that I can draw is that tenured members of the forum maybe tend to spend more time researching the boards before composing a post. However, this is beset by the fact that the total login time variable is somewhat of an unreliable measurement, plus the lurking factor thatiser with what they post, anyways. It's also a small sample size (N = 22), maybe if I had a larger dataset the results would be more concrete. I could always scrape the members list, but I suspect the admins wouldn't be too pleased about that.

That is some incredibly high-quality work that you have done there my friend, and I applaud you for your hard work.  :clap:  :thumbsup:

That is very interesting indeed. I'm sure that will be a treat for anyone intrigued by statistics.  :sombrero:






And just to clarify, I did post my own statistics, but that does not mean I am taking part in the dick-measuring contest. I simply volunteered the information for anyone who is interested in these statistics.

If I was being competitive, I wouldn't have that much to offer against several other people in this thread who have their collective time online in days in the hundreds.  :-D  :D


Now alternating between different highway shields for my avatar - my previous highway shield avatar for the last few years was US 76.

Flickr: https://www.flickr.com/photos/127322363@N08/

YouTube: https://www.youtube.com/channel/UC-vJ3qa8R-cc44Cv6ohio1g

DaBigE

Quote from: hbelkins on January 26, 2018, 11:48:40 AM
Oh goody, another "longest penis" contest, not dissimilar to "who has the most posts?"

LOL. Similar to what most women will tell you about dicks. In the case of the forums it's not the length of time you spend on the forum that matters, it's how you use that time.

In the interest of science and data points: 31 days, 12:09 (+/-3 days of logged-off browsing).
"We gotta find this road, it's like Bob's road!" - Rabbit, Twister

wxfree

I don't care for contests, but I don't see this as a contest, merely as an aggregation of data.  I'm up to nearly a fortnight, at 13 days, 23 hours, and 15 minutes.  I don't sign in just to read, and don't post a lot, but when I do sign in I usually stay signed in well after I need to, so I haven't actually spent that much time in non-lurking mode.
I'd like to buy a vowel, Alex.  What is E?

webny99

Quote from: US71 on January 26, 2018, 09:53:02 PM
Quote from: hbelkins on January 26, 2018, 08:34:27 PM
My total time might seem impressive; I don't know, and like I said upthread, I'm not a big fan of these penis measurement contests, so I'm not going to post my statistics.

OMG! We agree on something.  :o  :-o

If someone really wanted to get anal-retentive,  one could compare time on-line vs number of posts.

Quote from: wxfree on January 26, 2018, 10:50:43 PM
I don't care for contests, but I don't see this as a contest, merely as an aggregation of data.

Yeah, this. It's only a contest to the extent you make it one, and those of us who have answered the question aren't treating it like a contest at all. To the contrary, its quite an interesting discussion, not at all a bad one to be having.

ETA: Another thing I just noticed is that time you spend with the "post reply" window open does not count towards your time spent online.  :hmmm:

webny99

Quote from: adventurernumber1 on January 26, 2018, 10:25:04 PM
Quote from: Thing 342 on January 26, 2018, 06:32:10 PM
[stats snipped]

That is some incredibly high-quality work that you have done there my friend, and I applaud you for your hard work.  :clap:  :thumbsup:

That is very interesting indeed. I'm sure that will be a treat for anyone intrigued by statistics.  :sombrero:

Yeah, I second that. That was extremely interesting (whether it has any wider-reaching significance or not :spin:), and I'd love to see a graph like that for all of the forum users. Factoring in some of our even younger users, such as RGT, may also have an impact.

rickmastfan67

Quote from: webny99 on January 26, 2018, 11:26:51 PM
ETA: Another thing I just noticed is that time you spend with the "post reply" window open does not count towards your time spent online.  :hmmm:

It does, trust me.  However, if it takes you 16m to make the post, you don't have any time counted, but if you post within the 15m window, it does.

Hurricane Rex

Quote from: Thing 342 on January 26, 2018, 06:32:10 PM
Someone mentioned statistics, so I decided to run a sample analysis based on the information posted to this thread:

First I calculated a time-per-post metric `tpp` that measured a user's average number of minutes of forum browsing per post:

              username posts time        tpp
22                 bm7     4   12 180.000000
21             DeaconG   655  947  86.748092
20                   1  4594 6227  81.327819
19            slorydn1   907 1016  67.210584
18            Thing342  1112 1096  59.136691
17       CNGL-Leudimin  1910 1772  55.664921
16              Buck87   613  544  53.246330
15    adventurenumber1  1107  958  51.924119
14           busman_49   237  162  41.012658
13     Max Rockatansky  5796 2604  26.956522
12          thenetwork  2085  768  22.100719
11             webny99   484  163  20.206612
10           Scott5114  7150 2323  19.493706
9      jp the roadgeek  2017  609  18.116014
8  TheArkansasRoadgeek   646  192  17.832817
7              roadman  3061  881  17.268866
6              1995hoo  8678 2333  16.130445
5               vdeane  8581 1955  13.669735
4        jeffandnicole  7912 1762  13.361982
3              brandon  9039 1866  12.386326
2        Hurricane Rex   677  105   9.305761
1                Henry  3810  444   6.992126


I have a tendency to look at the posts while not logged in before logging in to post, hence the artificially low ratio. However, on the off-topic, weather and the challenge boards, I tend to post the first thing I can think of however not everything because that would make me look like an idiot. I feel like when I was in my first weeks it was a higher post per hour ratio because I had about 200 posts in my first day online (time logged in, about 7.2 minutes per post).

Also, thank you for your time finding these statistics.
ODOT, raise the speed limit and fix our traffic problems.

Road and weather geek for life.

Running till I die.

webny99

Quote from: rickmastfan67 on January 27, 2018, 01:01:05 AM
Quote from: webny99 on January 26, 2018, 11:26:51 PM
ETA: Another thing I just noticed is that time you spend with the "post reply" window open does not count towards your time spent online.  :hmmm:

It does, trust me.  However, if it takes you 16m to make the post, you don't have any time counted, but if you post within the 15m window, it does.

It should not have taken me more than 15 miutes to make that post. But I guess it did, apparently  :pan:

freebrickproductions

I'm at 26 days, 22 hours and 48 minutes as of this post.
It's all fun & games until someone summons Cthulhu and brings about the end of the world.

I also collect traffic lights, road signs, fans, and railroad crossing equipment.

(They/Them)

KEVIN_224


US71

Like Alice I Try To Believe Three Impossible Things Before Breakfast

webny99


7/8

58 days, 17 hours, and 47 minutes

ET21

Total time logged in: 8 days, 20 hours and 17 minutes.
The local weatherman, trust me I can be 99.9% right!
"Show where you're going, without forgetting where you're from"

Clinched:
IL: I-88, I-180, I-190, I-290, I-294, I-355, IL-390
IN: I-80, I-94
SD: I-190
WI: I-90, I-94
MI: I-94, I-196
MN: I-90

adventurernumber1

#71
Quote from: webny99 on February 01, 2018, 10:08:21 AM
7 days, 7 hours, 7 minutes  :clap: :bigass:

Since you really like statistics/numbers oddities like that, I have a treat for you:  :sombrero:  :bigass:







Now alternating between different highway shields for my avatar - my previous highway shield avatar for the last few years was US 76.

Flickr: https://www.flickr.com/photos/127322363@N08/

YouTube: https://www.youtube.com/channel/UC-vJ3qa8R-cc44Cv6ohio1g

Roadgeekteen

8 days, 15 hours and 24 minutes
God-emperor of Alanland, king of all the goats and goat-like creatures

Current Interstate map I am making:

https://www.google.com/maps/d/u/0/edit?hl=en&mid=1PEDVyNb1skhnkPkgXi8JMaaudM2zI-Y&ll=29.05778059819179%2C-82.48856825&z=5

hotdogPi

Quote from: webny99 on February 01, 2018, 08:37:12 PM
Quote from: Roadgeekteen on February 01, 2018, 06:12:26 PM
8 days, 15 hours and 24 minutes
Similar time to me, strangely enough. I'll let everyone draw their own conclusions :meh:

Comparing Roadgeekteen's time and your time gives a ratio of 1.172. Not noteworthy. That would be equivalent to me saying someone with 224 days or 306 days (actual: 261d 9h 17m) is similar to my time online.
Clinched, minus I-93 (I'm missing a few miles and my file is incorrect)

Traveled, plus US 13, 44, and 50, and several state routes

I will be in Burlington VT for the eclipse.

US71

What about quality of posts vs time logged in?
Like Alice I Try To Believe Three Impossible Things Before Breakfast



Opinions expressed here on belong solely to the poster and do not represent or reflect the opinions or beliefs of AARoads, its creators and/or associates.