Background

Contributor Level (CL, formerly CV) is a recently revised measurement of how much a SteamGifts (SG) user has given through giveaways (GAs) to the SG population. While often criticized as a poor measure for several reasons (mainly due to not considering number of giveaways won and not differentiating public, group, and private gifts), it has arguably worked well enough for most users of the site.
As a rule, users with higher CL have given more games (in relative monetary value) compared to those with a lower CL. This also gives them access to high level GAs, with higher odds of winning, which may result in them having a higher number of games. In this study, we seek to explore whether CL can be a predictor of number of games owned by an SG user.

Hypothesis

Our hypothesis is that users with a higher CL have more games on Steam compared to those with lower CL.

Methods

For this study we used a cross-sectional model with the convenience sampling method. Several public giveaways were made within the same period of time and of the same length for levels 3, 4, and 5. From the entries, 20 names (an arbitrarily picked number – no sample size calculations were done) in order of appearance to the investigator were picked, with the following inclusion criteria:

  1. User account registered at least 6 months ago (to ensure user knows how site works)
  2. User is of same level as the giveaways (users of higher CL are ignored during sampling)
  3. User has a public Steam profile (so number of games can be viewed)

We did not create level 1 and 2 giveaways in this study due to the associated risks and difficulties; instead, 20 samples each were sequentially collected from the users list from a random starting point with the same inclusion criteria.
Samples were then listed in a data table using Microsoft Excel 2007, a graph was made and statistical and descriptive analysis were performed using the same tool, the results of which are available below.

Results and Analysis

The results of the analysis can be seen in Figure 1 at the end of this article. Initially the graph shows classic linear growth from levels 1 through 4; however, it suddenly drops off at level 5. While it is tempting to make the very likely assumption that users at level 5, such as this one, are more tastefully selective in games, generous, and better-looking in general, this result is unfortunately more likely due to inadequate sample sizes. This is also suggested by the error bars, which are rather large, and proven by statistical analysis which showed no significant differences between all groups.

Conclusion

Our study did not produce any meaningful results, but we will try to skew and manipulate with the data so that we can net a sliiightly positive result and get it published. We also will use meaningless weasel words including “trend” “leans toward” and “somewhat” to mislead you, and use the classic “this study is a good starting point for further exploration into the subject matter”. As usual, our recommendations would be to replicate the study with more data points (i.e. give us more funding).

This study will be published in four days in the African Journal of Internet Behaviour.

This study was privately funded. Competing interest(s): author is a user of SG and actively participates in giveaways.

Continuing Scientific Education points can be earned from this article by answering this QUIZ.

========================================

Puzzle ENDED! Here is the SOLUTION

View attached image.
9 years ago*

Comment has been collapsed.

How many giveaways do I get chance to win as CSE credits after quiz?

9 years ago
Permalink

Comment has been collapsed.

The CSE Quiz leads to a single giveaway and some non-convertable CSE credits.

9 years ago
Permalink

Comment has been collapsed.

Preposterous
I want REFUND!!! I want my minutes back so that I can waste it on other shit like imgur/reddit.

9 years ago
Permalink

Comment has been collapsed.

Sure, here you go. plink plink

9 years ago
Permalink

Comment has been collapsed.

thank you very much.

9 years ago
Permalink

Comment has been collapsed.

really should've spent those credits at imgur :-\

Check out the solutions.

9 years ago*
Permalink

Comment has been collapsed.

Bumping for 100% solved. :)

That was some weird science.

9 years ago
Permalink

Comment has been collapsed.

View attached image.
9 years ago
Permalink

Comment has been collapsed.

Am i an anomaly in your study?

9 years ago
Permalink

Comment has been collapsed.

Q4 is... really not that hard ;)

9 years ago
Permalink

Comment has been collapsed.

I knew it! This proves scientifically that I lost games when I moved from Level 4 to 5!

Wait, what about Level 6? Did I get my games back?

9 years ago
Permalink

Comment has been collapsed.

I resemble these results! :-p

9 years ago
Permalink

Comment has been collapsed.

I'm probably an outlier, 300 games and level 7. I'm just not much of a game collector.

9 years ago
Permalink

Comment has been collapsed.

Interesting study. Have you analyzed the numbers and % of bundled games?

9 years ago
Permalink

Comment has been collapsed.

Further exploration of that aspect of this study would take considerable effort, as it will require the matching of 100 steam libraries against the bundle list. Not to mention the fact that the bundle list is dynamic in nature, resulting in the possibility of addition to the bundle group while the data is being analyzed. As a result, our group decided not to spare more resources on this issue.

Translation to non-science speak: Meh, that sounds like too much work. And I'm lazy.

9 years ago
Permalink

Comment has been collapsed.

Convincing explanation

9 years ago
Permalink

Comment has been collapsed.

Users at level 5, such as this one, are more tastefully selective in games, generous, and better-looking in general

This line, I like it

9 years ago
Permalink

Comment has been collapsed.

Nice study, though the sample size is a bit low. On the other hand, steamgfits doesn't provide the data in a presentive manner, so I can understand the low sample size.

9 years ago
Permalink

Comment has been collapsed.

WHAT IF... cg is actually a psychology professor working on a textbook on virtual items and how people interact with them?

And when he finishes his study he will take the site down ;__;

9 years ago
Permalink

Comment has been collapsed.

The power of a good article and puzzle, and of course solved!

View attached image.
9 years ago
Permalink

Comment has been collapsed.

A small bump at the halfway point. Only two more days until publication!

My lab is celebrating with a non-stop party!
In unrelated news, we are almost out of 100% EtOH (ethanol) for our experiments.
We have called every grant provider we know of to address this shortage, but we had difficulties articulating the problem for some reason.
Our request for some orange juice for the lab has also been met with negative responses so far.

Solve rate is at 36% - the puzzle shall remain hintless. Or is it really?

9 years ago
Permalink

Comment has been collapsed.

yeah last 2 questions have got me and i've been mulling over this for a few days

9 years ago
Permalink

Comment has been collapsed.

Solutions are now posted if you are interested.

9 years ago
Permalink

Comment has been collapsed.

A small bump for less than 24 hours to go to participate in the study.

9 years ago
Permalink

Comment has been collapsed.

For a moment there I was happy that someone actually did something meaningful instead of just trolling... but then I realized it's just a...

9 years ago
Permalink

Comment has been collapsed.

I am sorry that this study did not live up to your expectations. I promise the next one will be more substantial.

9 years ago
Permalink

Comment has been collapsed.

Bump 4 $cienc€

9 years ago
Permalink

Comment has been collapsed.

Less than two hours on the clock.

Five questions, how hard can it be, really?

9 years ago
Permalink

Comment has been collapsed.

Bump for solved ... sadly, I don't qualify for level 3 CSE credits :/

9 years ago
Permalink

Comment has been collapsed.

Puzzle has ended! Here is the SOLUTIONS:

First, here are the answers to the quiz. Here’s where you can find them:

  1. Sampling method used is clearly mentioned in “Methods”. Multiple answers possible (see above)
  2. In “Methods”, 5 groups with n=20, means total n=100 OR you can hover over Figure 1 at the end of the text, which also says n=100
  3. Statistical software used is also in “Methods”. Multiple answers also possible.
  4. Ah yes, the one that stumped many of you. You will find that you can HOVER over images in the SG forums, and a comment will be visible. In this case if you hover over Figure 1, it says ”…Error bars indicate Standard Error of the Mean or SEM.”. I gave several subtle hints in the form of some popular reaction images WITHOUT their usual captions, and the caption would appear when the image is hovered. There is also another way you can solve this one, but I kinda forgot what it was... hmmm…
  5. Pretty straightforward, answer available at the end of the article.

Some end-puzzle statistics: a total of 257 people attempted the puzzle, making 2865 guesses; a total 93 people solved it, which means it has a solve rate of 36.2%. Which is pretty impressive, considering barely any hints were given.

So congratulations to all of you scientific solvers and the giveaway winner, and I’ll see you when I publish my next study!

9 years ago
Permalink

Comment has been collapsed.

…except that wasn’t really the end.

While most of you took the “congratulations!” message and directly headed for the door with a wide grin and head held up high, some of you realized that I have THREE private giveaways going on. Solving the quiz only gave you giveaway #1. So where were #2 and #3?

  • Giveaway 2: Well, if you opened the attached image in another browser tab, it will bring you to this image. Remove the .jpg, and you’ll find the imgur page, where the image comment says “Error bars indicate Standard Error of the Mean or SEM” (oh, THAT was the other way of solving Q4!) More importantly, it also says “For the original dataset, please contact the author at www.steamgifts.com/giveaway/doO4v/”, which links to THIS.
  • Giveaway 3: I provided a link to the journal where I planned to publish the study. Well, I assume you’d realize that there is no such thing as the “African Journal of Online Behavior”; instead, the final five letters of the hyperlink (wewe9) leads to THIS giveaway when posted to an SG giveaway link.

Both methods are the simplest and most common ways of hiding giveaways; you can read about these and more complex methods in Zelgh’s Puzzle Guide.

…so this gives us even MORE statistics. How many people found ALL THREE giveaways? Well:

  • 93 people solved the quiz and found Giveaway #1
  • 15 people thought the journal link looked legit, and found giveaways #1 and #2
  • 3 people thinks hovering over the image is enough, and found giveaways #1 and #3

And 9 people found all three giveaways!

Congratulations to the fabulous puzzlers, in alphabetical order: CopperUndies, m139, onk, Starlynk (who solved the entire thing within minutes of posting), dalikkus, nomecopies, Nordhbane, q00u, and Toff !!!
There may be others, but you left no trace on the giveaways so I didn’t account for you.
EDIT: we can add tevemadar to that list, making a nice, round 10 completionists!

So this is the REAL end to the puzzle. Thanks for playing and I’ll see you on the next one :)

9 years ago*
Permalink

Comment has been collapsed.

I did not comment into the 'root' of Salammbo (had it already), but have proof: http://www.steamgifts.com/go/comment/kDnp0R7

9 years ago
Permalink

Comment has been collapsed.

Of course, I just realized that post since it was a reply to m139. Changed the post to reflect this.

9 years ago
Permalink

Comment has been collapsed.

Nice statistics, so I'm on the elitist 3.89% of players who found all the giveaways. Cool.

9 years ago
Permalink

Comment has been collapsed.

See. I can only blame this on my being on an iPad.

9 years ago
Permalink

Comment has been collapsed.

On the iPad I can not even reveal the hidden/spoiler text. Ok, I have not updated to iOS 7/8 because of the looks, so the browser may have changed a lot.

9 years ago
Permalink

Comment has been collapsed.

Right, I figured out you can copy/paste those bars into notes to see what they say.

9 years ago
Permalink

Comment has been collapsed.

Then it really may have changed: on iOS 6 copying the hidden text copies spaces (as many hidden characters have been selected).

9 years ago
Permalink

Comment has been collapsed.

so what was the exact answer to the last question as i tried copy and paste methods as well as simple yes and no and variations of what you had listed.

9 years ago
Permalink

Comment has been collapsed.

Uh, "Yes"? "Y" works as well.

9 years ago
Permalink

Comment has been collapsed.

yes was my first answer maybe i had answered an incorrect answer to number 4 and then answered yes to the other one to soon but im pretty sure when i typed yes it said wrong answer, Maybe it was case sensitive.

9 years ago
Permalink

Comment has been collapsed.

Pretty sure ITH is not case sensitive. And I never changed the answer for that one.

Oh well, better luck next time.

9 years ago
Permalink

Comment has been collapsed.

Sign in through Steam to add a comment.