CHI ’24: Overview of the Post-conference Survey

Author: Caroline Appert

Thank you to everyone who participated in the CHI post-conference survey, which was open for approximately two weeks after the conference (from May 29 to June 10). Overall, 613 people participated in the survey, representing approximately 15% of the total number of CHI attendees (3,995 in total). This compares with 17% of the total number of attendees in 2023.

This article provides a concise overview of the survey findings. The interpretations are based on a thematic analysis of the open comments (the thematic analysis was done by Caroline Appert and Simone Barbosa).

Attendees’ experiences

The following chart shows the distribution of respondents based on their type of attendance at the conference. The green bar represents respondents who attended the conference in person, while the brown bar represents those who attended virtually. Among the respondents, approximately 15% attended the conference virtually. This percentage is close to the actual virtual attendance, which was approximately 17% for CHI ‘24.

Bar chart showing number of attendees from 0 to 500 on the Y axis against attendance type. Two bars are shown. On the left, the bar for “attendance type = virtual” goes up to almost 100. On the right, the bar for “attendance type = site” is slightly above 500.

In the survey, 78 respondents shared their reasons for choosing virtual attendance over on-site attendance. These reasons are reported in the table below (respondents were allowed to select multiple reasons).

What factor(s) did influence your decision to attend the conference virtually rather than in person?
————————————————————————-
| 56.41% | My decision to attend virtually was influenced by ecological reasons. |
| 14.10% | My decision to attend virtually was influenced by cultural reasons. |
| 03.85% | My decision to attend virtually was influenced by the Covid-19 pandemic.|
| 42.31% | My decision to attend virtually was influenced by personal reasons. |
| 38.46% | My decision to attend virtually was for other reasons. |

The distribution of responses to the question “How valuable was your CHI experience?” is presented below, using a breakdown by attendance type: virtual attendance on the left and on-site attendance on the right. The chart shows that on-site attendance was highly valued, mostly because it offers opportunities for networking (connecting with people and meeting new people) and for getting up-to-date with scientific content that is timely and of high quality. In contrast, feedback from virtual attendees was less positive. In their open comments, virtual participants highlighted the lack of networking and interaction opportunities (both as attendees and presenters), the fact that there was little to participate in and insufficient synchronous/hybrid interactions for the sessions they could participate in.

Two histograms, one per attendance type, that report on the distribution of answers to the question “How valuable was your CHI experience?”. Possible values are along the x-axis with min value being “Not at all valuable” and max value being “Very valuable”. Counts per value are reported along the y-axis. The histograms show two opposite trends. The negative trend for virtual attendance on the left. Positive trend for on-site attendance on the right.

The survey included a series of questions where respondents rated various aspects of the conference, including sessions and activities (such as papers, demos, posters, panels, SIGs, student competitions, workshops, and keynotes) as well as conference events (such as receptions, exhibits, job fairs, and town hall meetings).

Overall, attendees gave mostly positive feedback about the sessions and activities, with all venues receiving a similar level of appreciation. However, some on-site attendees noted that activities in the exhibit hall could have been better. The most frequently mentioned issues included the lack of conviviality in the exhibit hall (quality of food and price of drinks during the reception), the perceived suboptimal poster layout for visibility and interactions, the limited number of demos during Interactivity, and the reported lifelessness of the Job Fair session. Virtual attendees, who did not experience the exhibit hall, gave consistently positive feedback across the sessions and activities they accessed. However, the low number of responses collected from virtual attendees makes it difficult to draw general conclusions.

Respondents also had the opportunity to evaluate how well different tools and aspects of the conference met their needs. This included the conference website, the Progressive Web App (PWA) used to navigate the program, and the different types of content offered by the conference, such as streamed content, recorded content, and the Sli.do application for handling questions and answers (QAs). The distribution of their responses is presented below.

Feedback regarding the PWA was mixed. Some attendees liked it, while others reported issues with getting the app to work on some mobile devices and found the navigation model confusing for such a large program. Attendees also suggested that the information on the conference website, especially instructions for presenters and remote attendees, should be better structured and clearer. Respondents mentioned that some information was included in specific blog posts that were not easy to find. To promote a better balance between remote and on-site participants, attendees were instructed to submit questions through the Sli.do app regardless of their mode of attendance. However, comments revealed that on-site attendees found this experience to be not fluid, and they quickly avoided using the app to ask questions during sessions. As a result, the appreciation of the Question and Answer experience was mixed. Finally, feedback regarding asynchronous presentations was notably negative, likely driven by the lack of interaction possibilities with pre-recorded videos.

Six bar charts laid out as a 2x3 matrix that report on the distribution of answers to question “How did the following meet your needs as an attendee?”. Histograms are labeled: PWA Navigation, Website, Streamed Content, Recorded Content, QAs, Asynchronous Presentations. Each bar chart shows a breakdown per possible values along the x-axis (values range from “Poor” to “Excellent”). The histograms reveal a positive trend with about 75% or more answers belonging to categories “Good” or “Excellent”.

To enable asynchronous presentations, presenters had to submit pre-recorded videos. For papers, case studies, and alt.chi presentations, the duration of these videos was capped at 15 minutes. For posters (including LBW, student competitions, and the Doctoral Consortium) and demos (Interactivity), the duration was set to 3 minutes. In the survey, we asked respondents to provide their feedback about whether these durations were suitable from both attendee and presenter viewpoints.

The charts below illustrate how well these durations aligned with what people expect either as attendees or as presenters. The large grey areas in the bars indicate that a significant proportion of attendees did not watch these videos. Regarding the duration of the videos, the trend for the 15-minute videos is that they are generally considered either appropriate or too long, while the trend for the 3-minute videos is that they are either appropriate or too short. Compared to last year, where long videos were 8–10 minutes long, a larger proportion of respondents this year mentioned that shorter videos would be preferable. In the comments, some attendees mentioned that the duration of videos should match the duration of on-site talks. As presenters, they also wish these videos to take less time to prepare and would like to receive instructions further in advance.

Four histograms, one per video type (paper, poster) x perspective (attendee, presenter), laid out horizontally that report on the distribution of answers to the question “How was your experience regarding the durations of videos?”. Possible values are “it worked well for me”, “I would prefer videos”, “I would prefer shorter videos” and “I did not watch any video presentation”. Overall, each bar has a large proportion of gray (did not watch) and then blue (worked well). Paper videos have a notabl

Accessibility

Among the respondents, approximately 2% requested an accessibility-related feature at the conference. The chart below indicates that these requests contributed to a positive experience. Very few accessibility-related issues were reported by CHI participants in the survey. Some of the issues reported included lack of food options, lack of air-conditioned spaces or difficulty using the PWA.

Histogram reporting the distribution of answers to question “Did you have negative accessibility experiences?”, using a breakdown per attendance type. Both virtual and on-site bars are predominantly “no”.

Transportation

The question about local travel in Hawai’i during the conference revealed that the majority of local travel was made on foot, and by car to a lesser extent.

How did you undertake local travel in Hawai’i while attending the conference?
———————————————————-
| 90.26% | On foot |
| 42.69% | By car |
| 03.48% | By metro (skyline) |
| 02.78% | By bike |
| 12.76% | Other |

Willingness to attend CHI ’25

In response to the final question on the likelihood of attending CHI ’25 in Yokohama, Japan, the majority of respondents will probably or definitely attend the conference physically, although some indicated that this would depend on the acceptance of papers or the availability of funds. The chart also illustrates that a very small number of respondents indicated that they would attend virtually, with a few stating that virtual attendance was not worth the effort.

Bar chart showing the distribution of answers to question “How likely is that you will attend CHI ‘25?” from 0 to 250 on the Y axis. Four bars from left to right: “Virtually” (about 30 on the Y axis), “Not attend” (about 65), “Probably physically” (about 250) and “Certainly physically” (about 75).

As a final note, we would like to thank all the respondents who took the time to provide valuable feedback. We have carefully read and analyzed all the detailed comments, categorizing them into relevant themes and summarizing the key points. The input and insights shared by the attendees are very valuable to everyone involved in SIGCHI, the CHI Steering Committee, and the overall organization of CHI.