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.