Hey;
mittlerweile bin ich fertig mit meiner Arbeit und habe auch schon eine Bewertung erhalten. Ich wollte ungern Ergebnisse teilen bis ich eine Bewertung für meine Arbeit erhalten habe, deswegen hat es ein wenig gedauert :). Vielen Dank an dieser Stelle noch einmal für eure gute und hilfreiche Teilnahme an der Umfrage, es hat mir sehr geholfen! Besonders positives Feedback war wirklich hilfreich während der Arbeit und da bin ich diesem Forum sehr dankbar für! Da es leider nicht immer nur positiv ist und teils auch sehr destruktive Kritik im Internet gegeben hat, hatte ich schon relativ früh die Kommentar-Funktion bei den Umfragen deaktiviert.
Ich habe schon kurz nach der Abgabe ein wenig die Ergebnisse zum Teilen auf den Plattformen zusammengefasst. Die Zusammenfassung ist auf Englisch, da ich nicht nur deutsche Plattformen in meine Arbeit miteinbezogen habe. Insgesamt habe ich zwanzig individuelle Umfragen auf verschiedenen Plattformen gestartet und es fanden sich auf 15 von 20 Plattformen Teilnehmer (Teilnehmerzahl ~230). Falls ihr noch Fragen haben solltet, könnt ihr mir diese gerne stellen. Twitch war im übrigen die Plattform der Gruppe mit niedriger Nutzerbeteiligung an der Erstellung der Community-Richtlinien, die am meisten den Community-Richtlinien zugestimmt hat. Dennoch war die Zustimmung bei allen Plattformen mit Nutzerbeteiligung höher.
Here are some results of my master thesis.
A total of twenty platforms were approached to conduct the survey. Users from 15 different platforms participated. As the focus was on user participation in the policy development process, the platforms were divided into two groups according to user participation. Both groups were similar in size and structure. While the group with low user participation (LUP) was mostly 18-29 years old, the group with high user participation (HUP) included more younger and older people (under 18 and over 50 years of age). In both groups, the participants regularly consumed content on the platforms. The participants of the HUP Group created their own content more often.
Here some details regarding the two groups:
Group 1 (HUP): all platforms with high user participation in the guideline creation process
6 platforms; 107 responses
Group 2 (LUP): all platforms with low user participation in the guideline creation process
9 platforms; 116 responses
Used Methods:
Spearman-method (Correlation)
Welch-Tests (statistic significance)
Tested effects:
- Agreement to the single guidelines
- Knowledge of the community guidelines
- Conscious implementation of the community guidelines
- Maturity of the community guidelines
- Applicability of the community guidelines
- Sufficiency of the community guidelines
- Missing community guidelines
Hypotheses were made to test for differences between the two groups. Effects of user participation in other research areas were considered in order to hypothesize possible effects in the process of policy development. In particular, the involvement of users in system development and crowdsourcing processes. Mainly positiv aspects were expected by letting the user participate in the guideline creation process.
Hypotheses that were confirmed:
- H1: The user participation in the creation of the guidelines has a positive effect on the approval of the catalogue of guidelines.
- H2: The individual guidelines are more user-oriented with given user participation and should therefore receive stronger user approval than the guidelines specified by the platform owner.
- H3: The stronger user orientation should have a significant positive impact on the applicability of the Directives.
- H6: Since users are increasingly concerned with the guidelines, the guidelines should be more mature from the user's point of view.
- H10: The involvement of users in the preparation of the guidelines should have a significant positive impact on the adequacy of the rules on content production.
- H5: Due to the increased familiarity with the community guidelines as well as the stronger adaptation of the specifications to the needs of the users, these are increasingly consciously implemented by the users.
- H13: Users attach importance to quality in both consumption and content creation.
- H4: With a high level of user participation, users have a better understanding of community guidelines.
Correlation tests were performed with the spearman-method and for the hypotheses Welch-tests were used. Welch-tests are one kind of t-test that can be used for different sample sizes. The significance level was set at 1%. As tool for the calculations RStudio was used and the responses were filtered with increasing knowledge of the community guidelines. All in all, 4 Tests were performed. One with every response (not considered were the questions to the guidelines in general), on with minimal knowledge (2/7 points), one with medium knowledge (4/7 points) and one with very good knowledge of the community guidelines (6/7 points). The results were confirmed in all tests. Only in the last test the differences in applicability and agreement lost their significance. Probably due to the small sample size in the LUP group (28 remaining, 77 in the HUP group). The sample sizes in the last test also show the differences in knowledge of the community guidelines.
The first major differences between the two groups were their agreement to the Community guidelines and their conscious implementation. In the LUP group ~69% at least somehow implemented the guidelines while in the HUP group with 96% almost everyone at least somehow consciously implemented the guidelines. ~20% of the participants in the LUP group do not implement the guidelines at all.
This is also consistent with the Group's knowledge of Community directives. While the knowledge was rated with 4,052 in the LUP group it was significantly higher in the HUP group (5,916).With increasing knowledge regarding the community guidelines were rated more applicable and mature. The agreement on the Community guidelines also increased with increasing knowledge of them.