Q&A: MVP Chiayi Yen Is On A Mission To Make Data Science More Diverse
Published Apr 10 2019 08:20 AM 493 Views
Steel Contributor

First published on MSDN on Feb 15, 2018

Photo from left to right: R-Ladies Taipei Ju-yin Tang with her child, R-Ladies co-founder MVP Chiayi Yen, and R-Ladies Taipei co-founder and MVP Ning Chen with her child

MVP Chiayi Yen is a Taiwanese quantitative modelling specialist, a PhD student, and the translator of two novels on computer science for children. Now based in Mannheim, Germany, she has also dedicated herself to correcting the gender imbalance in the data science field. Three years ago she helped launch R-Ladies Taipei – a female-oriented data science community that’s hosted more than 50 sessions!

We spoke to her about what inspired her to found R-Ladies Taipei, how she grew the community, what’s challenges she’s faced –and about the time she taught Taiwan’s first female president to code. Check it out!

1) What inspired you to found R-ladies Taipei? What’s the mission of the community?

The idea of founding R-Ladies Taipei had occupied my mind since I was made an officer in the TW R User Group (TW-RUG), but it became possible when I met Ning Chen and Ju-yin Tang in 2014. We noticed that among the 80 participants attending each TW-RUG meetup, usually 5 or less of them were women – not to mention, all the speakers were male.

As female members, we were curious where this gender-imbalance came from. Was it because women didn’t excel in the data science field? Or because they can’t adapt to the current male-oriented community culture? It was not the case that women were not able to perform well; instead, it was more likely that it was comparatively harder for women to network comfortably in a male-dominated community. Therefore, we decided to do something to somehow mitigate this gender imbalance; that is, we launched R-Ladies Taipei, a female-oriented data science community.

For me, the mission of R-Ladies Taipei is to create a friendly environment where women are able to share knowledge about data science. It is not an organization that isolates women from the rest of the population in the open source community. Instead, we consider women undervalued human capital in the data science field, and we truly believe that an investment in female human capital would bring huge and sustainable dividends – providing more diverse opinions and additional insights into data science.

2) How exactly did you scale the community to this size?

We never worried about the size of the community, but women were more motivated to participate than we originally imagined. In our first meetup on November 2014, more than 80 women participated in the event. Given the fact that there were only 5 female participants on average in the TW-RUG meetup before we launched R-Ladies Taipei, the event totally exceeded our expectation!

I think one of the keys to our success in scaling the community is to leverage other existing popular communities and to encourage more inter-community collaboration. For example, we co-hosted some meetup events with TW-RUG, and therefore the event information was spread via both our Facebook and meetup pages, which enabled us to reach new potential audiences.

3) R-ladies has hosted 50+ sessions over three years. What common themes have you covered, and how do you work to keep those in the community engaged?

There are many different and diverse topics, and I think the only common point is that it is all about data. For example – web-crawling and data ETL; machine learning and data mining; open data and Kaggle competitions; cloud computing and development tools; and some case studies like security, air quality, financial trading, digital humanities, and others. There are so many!

As for how to keep participants engaged in the community, it is a universal problem for most communities around the world. The most difficult one might be that you can only find a few female tech speakers. Not being able to find female speakers could lead to a big problem for female-oriented communities like us. However, we are particularly proud that we have never lacked female speakers in the past three years, because we find a way to overcome it – that is, with featuring female speakers from our own community!

While we call for speakers just as other communities do, the main difference might be that we encourage our participants to become speakers even when they have only been learning R for a few months. We

help those volunteers to prepare their talks in advance if necessary, and most importantly, we do not criticize their performance afterward. This indeed makes a huge difference in women’s incentive to share something on the stage.

But why do we trust beginners to speak? What if she delivers a terrible talk to the audience? Well, it doesn’t matter – because when it happens the audience thinks, ‘Hey, if this lady can be a speaker, I can register to speak next time too because I can do better than her!’ Sometimes, we find that women are less confident and more afraid to make mistakes than men. Therefore, we make a more mistake-tolerant environment so that women are more willing to exchange their findings and experiences without fear of failure. And in our experience, most of the speakers are well prepared and perform amazingly well – the only thing they need is to be brave!

4) R-ladies once taught Taiwan’s President Ing-Wen Tsai to code. Can you tell us a bit about this, and why it was important for the female development community in Taiwan

In November of 2016, I was invited to be one of the female engineer representatives to teach Taiwan President Ing-Wen Tsai to code for the event “Hour of Code 2016,” held by Microsoft Taipei. “Hour of Code” is a yearly event that encourages all people, in particular minorities, to learn “computational thinking” by coding even if they are not engineers. In 2016, the main theme was women, so Microsoft Taipei invited President Tsai as the special guest. This is because she, as the first-elected female president of Taiwan, is a great spiritual figure inspiring women.

The three largest female open source communities in Taiwan – R-Ladies Taipei, PyLadies Taiwan and Girls-In-Tech Taiwan – were selected as the most appropriate candidates to be the “partners” of President Tsai while she learned to code. And we were partners not only for President Tsai – symbolically, it expressed the message that the female development community could be a great partner for all women who want to learn technology!

5) What tips would you give to others looking to start their own communities, or looking to expand the communities they already over see?

According to our experiences, I think there are two elements that are most important: one is demand and the other is supply. From the demand side, a clearly defined target audience segment might be a good start. Take us for example: “R-Ladies Taipei” defined itself as a community about “R” language for “Ladies” living in “Taipei”. This not only helps us to differentiate our position in respect to other communities, but also it is easier for participants to evaluate whether to join this community or not.

From the supply side, it is important to feature a big pool of speakers, especially in the first year. However, it is very challenging to find enough speakers in the beginning stage of a new community. Our experience is that a series of related topics can be nice warm-up events for a new community. For example, we selected a 12-Chapter book “Machine Learning for Hackers” as the main theme of our meetup events in our first year, and we called for volunteer speakers to introduce one chapter for each of the 12 monthly meetups. And after 1 year discussing and programming, even someone who had been a newbie one year before was able to construct her own model and apply it in her job; we got everybody ready for more advanced topics.

Another important thing was letting everybody get the chance to “level-up” in the community. While some communities focus on organizing events for experts, they leave out beginners; meanwhile, others focus on organizing events for beginners, and leave out more advanced members. Therefore, the strategy that R-Ladies Taipei uses is that we treat our community as an “organism” and we allow all our members to help grow it up in the way they want. For example, our beginners organize their “R-Basic” study group and summarize what they learned to some newly-joined members in the monthly meetup. For intermediate members, we organized a “Go Kaggle!” event and invited these intermediate members to attend Kaggle competitions in order to gain more practical experiences on data modelling. As for advanced members, we have a plan called “Travel across the world with R-Ladies,” where we encourage these advanced members to expand their network and challenge themselves to be speakers in other R-Ladies branches while traveling abroad. Therefore, no matter what level you are in the community, there’s always challenges to conquer.

6) Do you have anything else to add?

Finally, I am thankful for Microsoft’s efforts in supporting the female developer community, especially those support from Reneata Chang, the one who is in charge of MVP program in Taiwan, Hong Kong, and Macau. And the MVP Program rocks :)

Version history
Last update:
‎May 03 2019 04:33 PM
Updated by: