3.1.1. Number of Articles in the Selected Database and the Growth Trend

As described above, the Web of Science database was selected for our review, and studies were searched within it according to the adopted criteria. After the initial analysis, it was discovered that none of the reviewed articles were published in 2011. The first article that met the inclusion criteria appeared in 2012. In the years 2012–2017, we observed a clear increase in interest in the issue under study. The peak period of interest was 2017, when 18 articles were published. Despite the decrease observed later, 2020 was again characterized by an increase in the number of publications compared to the previous year (see Figure 2 below).

**Figure 2.** Number of articles concerning the issues of collective intelligence and policymaking published annually, and the growth trend for the period 2012–2020.

#### 3.1.2. Concepts and Terms That Appeared in the Articles

We analyzed the content of the research articles included in the review, and created lists of the most common concepts that appeared in article titles, article abstracts, original keywords, as well as KeyWords Plus generated by the Web of Science algorithm [36]. The results are presented below in Table 1.


**Table 1.** Rankings of top 10 concepts based on: (**a**) article titles, (**b**) article abstracts, (**c**) author keywords, (**d**) KeyWords Plus.

3.1.3. Identifying Methods and Strategies of Studying CI in Policymaking

In this section, the methods and strategies of studying CI in policymaking, which were identified in the analyzed texts, are presented. As described in Section 2, 15 methods and strategies were identified in the reviewed sample, and each text was associated with a minimum of one and a maximum of five methods. In Table 2 we present a list of identified methods and strategies, ranked from the most to the least popular, and the adopted definitions, supplemented with references to theoretical literature.

As we can see, the analysis of organizational structure/design was the most popular method. Fewer studies used the analysis of created values approach. Subsequent identified methods, such as the analysis of the e-participation process, the analysis of participants' behavior or collaboration models enjoyed moderate popularity. On the other hand, the least frequently used methods included the analysis of platform usability, analysis of the impact of AI algorithms and analysis of organizational learning. The relatively rare occurrence of the analysis of impact on policymaking approach is also worth noting.


**2.**MethodsandstrategiesofstudyingCIinpolicymakingidentifiedinthereviewed

*Entropy* **2021**, *23*, 1391



