Preamble

We have attempted to grasp the scale of international engagement with the K-Wave to date looking at such proxy measures as a census of online fan organisations, the amount of Korean content consumed on a global streaming platform, and changes in the quantity of coverage devoted to K-Wave related topics in English-lanugage journalism (here, here, and here, respectively).

Here, we do so again, but relying upon survey data which simply asks people about the level of engagement they believe is enjoyed by a selection of concepts related to the K-Wave in their country. While we are visualising the K-Wave, it may also be instructive to devote some specific attention to visualisation. Thus, alongside simply reporting the perception of engagment with the K-Wave across different countries and regions, we also explore a novel way of visualising this using Sankey diagrams.

The Data

As last month, we continue to explore the Korea Foundation surveys on the state of the Korean Wave. Specifically, we will look at the most recent survey, dated 2024 but actually carried out during November 2023. 25,000 people from 26 countries participated.

Respondents were asked the following question:

현재 귀하의 나라에서 한류의 인기가 어느 정도라고 생각하시나요? 귀하의 생각과 일치하는 정도를 응답해 주십시오.

While we do not have access to the translations actually used for each country or language in which the survey was administered, taking the context in which it was asked into consideration, we freely render it as:

How popular are the following K-Wave concepts in your country? Please select the option which corresponds with your belief.

The “options” alluded to in the above translation were five statements which represented points on an ordinal scale. They are reproduced as they appear in the report below:

1 - 이용하는 사람이 거의 없다

2 - 일부 마니아층에서 인기가 있는 상태

3 - 일부 마니아층뿐만 아니라 일반인들도 상당수 인지

4 - 일반인에게 널리 알려져 있고 관련 상품이 판매됨

5 - 일반인에게 널리 인기 있고 관련 상품의 판매가 원활

And we freely translate them as follows:

1 - Almost no-one engages with it

2 - It is popular among a few devoted fans

3 - A reasonable number of people among the genral public are aware of it, not just a few devoted fans

4 - It is widely known among the general public and you can buy related goods

5 - It is popular among the general public and related goods are easily accessible

Thus, higher proportions of respondents providing more highly ranked responses represents greater perceived engagment with a given K-Wave concept. The K-Wave concepts that were rated will become apparent from the visualisation which follow.

Global Awareness of K-Wave Concepts

We can visualise the proportion of all respondents to rate each concept at each point on the scale presented above, as below.

As striking as this visualisation may be, interpreting it is challenging at best. Sankey diagrams are typically associated with changes in flow over time, but here we attempt to use one to represent proportion. In this case, we consider the information better presented in a stacked or dodged bar chart, as below.

Of all of the above visualisations, we consider the ‘dodged’ barchart the most easily interpretable for determining the proportion of respondents to assign each concept a particular score. There may, though, be other aspects of the data that could usefully be highlighted using different visualisations. Sankey diagrams may, thus, be added to and modified to make them not only easier to interpret, but to enable us to draw insights that the presentation of the data in a bar chart might obscure.

Turning first to making the original Sankey diagram easier to interpret, we note that not all data need be included in every visualisation. When we filter the data to include just ratings that exceed a certain threshold of the proportion of respondents, we are left with a diagram which is easier to interpret, albeit not wholly unproblematic.

We are also able to draw additional insights from the above visualisation. For example, the absence of a node for the score ‘1’ demonstrates that there is not a single concept regarded as so little engaged with that 25% of respondents gave it such a low score. We can further see from this visualisation the scores associated with each concept by the largest proportions of respondents. For most concepts (seven out of eleven), only a single response was chosen by more than a quarter of respondents. This observation, in turn, raises questions about broader patterns of perceived engagement with the K-Wave. For example, why does such consensus about perceived engagement not obtain for Korean food, which was rated at points 3, 4, and 5 by in excess of 25% of respondents at each of those scale points?

Sankey diagrams are amenable to modification not only by filtering data, but by adding to it. This enables us to, for example, add qualitatively established categories to our visualisation. This addition, along with its positioning, grants us impressionistic insights that many other types of visualisation would struggle to capture.

The visualisation on the left simply introduces a higher-level categorisation of the concepts. It shows the concepts membership in each of these mutually exclusive categories, showing that the majority of the concepts under discussion relate to screen media. Interpreting the rating of each concept, though, remains challenging.

Changing the order of the nodes in this diagram, as in the visualisation on the right, retains the information about each concept’s membership in the high-level categories. It also adds something to our analysis by aggregating the ratings of individual concepts to show patterns of perceived engagement by category. Whether or not this categorisation tells us anything about the K-Wave is debatable, but it does demonstrate an interesting affordance of this style of visualisation.

To conclude this section, we put matters of visualisation to one side and returning to the question of how engagement with the K-Wave is perceived globally. None of the eleven concepts investigated in this survey were considered known to only a restricted group of K-Wave fans in their respective countries by more than 50% of respondents, although literature came close. Literature received the largest proportion of responses rating it at 1 on the scale (18.3%), followed by webtoons (16.5%), then games (11.6%). At the other end of the scale, K-Pop (25.9%), Korean Food (25.3%), and K-Beauty (24.6%) were rated at 5 on the scale by the largest proportion of respondents. Why we should observe this distribution in the global, aggregated set of responses remains a matter for speculation at this point.

Regional Variation in Awareness of K-Wave Concepts

Regional variation may provide some clues as to the underlying reasons for particular concepts being more or less popular on the global scale. For example, concepts that were considered very well known in some regions could have met our threshold for inclusion in the above visualisations while those which enjoy only moderate levels of engagement over all of the surveyed countries may not. Thus, we continue our exploration of this data, and this style of visualisation, by turning to regional variation in the perception of local engagement on the national level with K-Wave concepts. The regions we will be looking at will be defined using the most restrictive understanding of two spheres of influence, specifically the Sinosphere and the Anglosphere. In this case it is interesting to note that the countries of the Anglosphere are more widely dispersed over the globe than those of the Sinosphere, but all Anglosphere countries are further from Korea than all Sinosphere countries.

We will restrict our investigation here to determining whether there is any variation in either the distribution or number of K-Wave concepts perceived as well-known in the countries of these regions. In this context, we are operationalising ‘perceived as well-known’ to mean rated at 4 or 5 on the scale introduced above by at least 40% of respondents in that country.

The Sankey diagram visualising the K-Wave concepts perceived as well-known in Sinosphere countries is below.

We now turn to the Sankey diagram for Anglosphere countries, once more below.

From these visualisations, it is clear that a much larger range of K-Wave concepts are considered to be well-known in the Sinosphere. Of the concepts considered well-known in the Anglosphere, most (three out of five) are well-known only in the USA. In the Anglosphere, UK respondents are unusual for only identifying a single concept, Korean Food, as well-known. Returning to the Sinosphere, it is striking that only four of the five the concepts considered well-known by any number of Anglosphere countries are considered well-known in all Sinosphere countries, the exception being animation. Concepts considered well-known in any number of Sinosphere countries that do not share such a perception in the Anglosphere include: Variety Show, Games, Films, and Drama. This is a striking difference. It appears that diverse types of K-Wave screen media are considered well known across the Sinosphere, while they are not considered well-known, by our definition, in the Anglosphere. This is the case even given the astounding success of Korean screen media on streaming services in 2023 (discussed in greater detail for the first half of the year here and the second half here) and and as reported in English-language media over the course of the early 2020s (for example, here), the perception that it is not well-known persists. Finally, we note that the responses from Japan reveal a regionally unusual pattern of perceived engagement with the K-Wave among Sinosphere countries. The only screen media perceived as well-known there is Drama. Otherwise, the concepts which respondents identified as having high levels of engagement were those which were well-known in the Anglosphere, again, excluding Animation.

We can further compare these findings with the globally aggregated results. In so doing, we may draw the impressionistic conclusion that proximity to Korea may be a factor in higher perceived engagement across a greater range of K-Wave concepts. The globally aggregated results showed only Korean Food, K-Pop, Beauty, and Fashion rated at levels 4 or 5 by at least 25% of respondents. This is in broad alignment with the geographically disparate Anglosphere countries. The Sinosphere countries, however, buck this trend due to the greater perceived engagement with screen media.

Conclusion

We have examined the perception of engagement with K-Wave concepts across 26 countries and visualised it using Sankey diagrams. The level of perceived engagement around the world was established to be high most especially for Korean Food, K-Pop, Beauty, and Fashion. This was supplemented by a case study focusing on two specific regions, which suggested that Sinosphere countries differ from the others surveyed in that respondents from the region perceive engagement with screen media to be markedly higher.

While the data required considerable manipulation to produce Sankey diagrams which were amenable to interpretation, these insights were enabled by their simple visual inspection. Whether such insights could have been gathered from alternative visualisations is a matter of speculation. We may, however, consider this report a fruitful examination of a novel type of visualiation and an object demonstration the contribution visualisation can make to achieving a high-level understandings datasets too large for one individual to read closely.

Acknowledgement
This work was supported by the Core University Program for Korean Studies of the Ministry of Education of the Republic of Korea and Korean Studies Promotion Service at the Academy of Korean Studies (AKS-2021-OLU-2250004)