#chinesecharacterhelper

So far in my life I've endeavored to learn Chinese four times. Once via an audio-based language course, once via a continuing education course, once via a tutor, and once in Chengdu China where I had both a tutor and eight dozen flashcards (each with an image on one side and a Chinese character and pronunciation on the other).

During those attempts I learned to comprehend and produce phrases, words, and sentences. At my peak I had a Chinese listening and speaking vocabulary of maybe 350 words. I could say hello, talk about family, order food, pay for things, describe my hobbies, count, and ask if the person spoke English, Spanish, or Norwegian.

When it came to reading characters however, I topped out at about 25. I could write 10, maybe.

Learning how to read and write Chinese is (obviously) different from learning to read and write English, Spanish, or Norwegian. Reading and writing Chinese felt like memorizing 100s of complex symbols, if felt like I was learning the language all over again--only at a much slower pace than learning to speak Chinese. Despite the presence of pronunciation or meaning-based clues embedded in many Chinese words (called radicals) when I looked at the characters I didn't feel like I was getting any clues about meaning or pronunciation. Thus, when given the choice of spending time listening and speaking Chinese or using that time to learn how to write and read Chinese words I rarely opted to work on memorizing characters--as it was hard, solitary, and tedious. While I should have been more diligent, my experience with learning/not learning to read and write Chinese is fairly common (Tse, Marton, Ki & Ka, 2007). In fact, the process of learning to read and write Chinese was difficult enough that it led to the creation of a pronunciation-based way of writing Chinese called Pinyin. This approach was so successful that it is used in most Chinese language schools and Chinese elementary schools as a step toward learning to write characters. The problem is, many learners would much prefer to use Pinyin and don't end up learning to read and write characters.

It was a review of the literature related to acquiring Chinese literacy that led Xianquan Liu, Shawn Hellwege, Katie Bieber, Christopher Heselton, Asha Srivastava, Juana Paramo Reyes, and I to design and develop a working alpha version of the DaZiBao Chinese writing web app.

DaZiBao uses students’ speaking and listening skills as a way to help them communicate via Chinese characters. Users type the phonetic spelling of a word. For example, the Chinese word for 'horse' (马) is spelled 'ma' and pronounced 'mǎ'. The most common way for all Chinese speakers to write Chinese in digital formats is to use an add-on input method that enables them to spell 'ma' to access their intended Chinese character '马' out of a list of other possible matches that are also spelled 'ma' (马 吗 妈 嘛 骂 码 麻).

Image 2: The process of producing Chinese words within digital mediums uses phonetic spelling and visual identification of intended characters (from Olmanson & Liu, 2017).

Digital writing via this input method for native speakers and language learners is fairly straightforward as long as the user has memorized character-spelling-meaning groupings (马-ma-horse). The problem for language learners whose oral vocabulary is much larger than their written capacity is that they often know the pronunciation-meaning grouping (ma-horse) but not the character (马). We designed DaZiBao to use pronunciation-meaning knowledge as a gateway to character learning through character selection and use.

As described in detail in a recent paper (Olmanson & Liu, 2017), our design uses a series of learning supports to enable learners to express themselves via characters before they know how to write or even recognize them. We do this in a gradual way, offering first the characters alone, then the pronunciation of each character, followed by images of each word, and finally an English translation.

In Image 3 above, by staggering the display of supports for learners we encourage the user to focus on the characters. If they can recognize their intended character by sight they can click it and continue composing without waiting for or relying on a pronunciation, an image, or a translation as a crutch.

In Image 4 above, if after three seconds a selection hasn't been made, audio icons appear under each character. Users can hear how each character is pronounced by clicking the speaker icon associated with a character.

Nine seconds later if learners still have not been able to use the audio pronunciations to identify their intended character a column of 7-10 images appear below each character--giving visual clues to character meaning. Images are identified via the Flickr API (one based on Chinese character-image search and one based on translation-image search). The results are displayed in alternating fashion as a way to offer both home and target-culture connections (see Image 5 above). See the project code repository for details.

Finally, ten seconds later and twenty-two seconds after the learner typed 'ma' English translations are displayed (see Image 6 above). We opted for this significant delay so that users would not rely too much on a simple translation and would instead associate the character, pronunciation, and multiple images with the word they intended to write and were learning.

With a partnership in place with the Confucius Institute of Nebraska, we were ready to implement the application with students enrolled in K-12 Chinese language courses in Nebraska and conduct research to understand the impact this technology could have on learners’ experiences of becoming literate in Chinese (as well as the impact DaZiBao could have on instructional practice). However there was a problem. In testing the application we found that sometimes images would be displayed that were inappropriate for K-12 settings. Before we could pilot the application in schools we needed to filter images (continued in part two).

Xianquan Liu is a PhD candidate in the department of Teaching, Learning and Teacher Education in the College of Human Sciences at the University of Nebraska Lincoln.

Shawn Hellwege, earned his Master's of Education degree in the department of Teaching, Learning and Teacher Education in the College of Human Sciences at the University of Nebraska Lincoln.

Katie Bieber is a Master’s student in the department of Teaching, Learning and Teacher Education in the College of Human Sciences at the University of Nebraska Lincoln.

Dr. Christopher Heselton, is Associate Director of the Confucius Institute at the University of Nebraska Lincoln.

Asha Srivastava, is a visiting scholar in the department of Teaching, Learning and Teacher Education in the College of Human Sciences at the University of Nebraska Lincoln.

Juana Paramo Reyes, is a participant in the Undergraduate Creative Activities and Research Experience (UCARE) program at the University of Nebraska Lincoln.

Dr. Justin Olmanson is an Assistant Professor of Instructional Technology specializing in the use of multimodality, artificial intelligence, and data in the design of learning experiences. He is in the department of Teaching, Learning and Teacher Education in the College of Human Sciences at the University of Nebraska Lincoln and leads the work on DaZiBao/Chinese Character Helper.