Appendices
Appendix 1: College Students’ In-Person Prototype Testing Transcripts
Due to the large amount data captured by the in-person interview performed for testing with 32 people at 6 universities, I’ve chosen to exclude the full dataset from the PDF version of my thesis. The data is available upon request - and has also been published on Hugging Face: https://huggingface.co/datasets/krishaamer/taiwanese-college-students
Below is a small example of the data gathered.
Example Transcript 2024-11-13 - Tainan (STUST) - 1YEDC
Source Files: * 2024-11-13 - Tainan (STUST) - 1YEDC.json *
2024-11-13 - Tainan (STUST) - 1YEDC.txt
Ziran ID Code: 1YEDC
| Traditional Chinese Transcript | English Translation |
|---|---|
| Me: 好，我再問一次，這個是我的論文的一部分的一個App，希望妳幫幫我測試。我會錄音妳的聲音，但是妳的回覆都是匿名的，可以嗎？ | Me: Okay, let me ask again. This is an app that’s part of my thesis. I hope you can help me test it. I will record your voice, but your responses will be anonymous. Is that okay? |
| Interviewee: 可以，可以。 | Interviewee: Yes, yes. |
| Me: 好，謝謝。那可以開始了。好。呃，妳之前有用過Momo嗎？ | Me: Okay, thank you. We can start then. Okay. Uh, have you used Momo before? |
| Interviewee: 有，可是用蝦皮比較多。 | Interviewee: Yes, but I use Shopee more often. |
| Me: 好。那妳平常買什麼？ | Me: Okay. What do you usually buy? |
| Interviewee: 平常嗎？衣服跟化妝品，還有動漫的娃娃。 | Interviewee: Usually? Clothes and cosmetics, and also anime dolls. |
| Me: 嗯。那妳可以幫我找一個妳有興趣的產品嗎？ | Me: Hmm. Can you help me find a product you’re interested in? |
| Interviewee: (Searches) …有興趣的…產品…面膜。 | Interviewee: (Searches) …Interested in… product… facial mask. |
| Me: 面膜？這個？好。這樣。這個品牌妳有之前買過嗎？ | Me: Facial mask? This one? Okay. Like this. Have you bought this brand before? |
| Interviewee: 哦，有，有買過。 | Interviewee: Oh, yes, I’ve bought it before. |
| Me: 哇哇，好酷喔。買東西基本上，買東西也是一個投資…喔，這個是可以幫忙分析的。 | Me: Wow, wow, so cool. Buying things basically, buying things is also an investment… Oh, this can help analyze. |
| Interviewee: 嗯，哇，酷。 | Interviewee: Hmm, wow, cool. |
| Me: 那妳覺得這個內容裡面有什麼對妳來說有用、用途嗎？有用，就是妳覺得有趣的資料，還是妳覺得不錯的材料？ | Me: What do you think in this content is useful for you, or serves a purpose? Useful, meaning interesting information, or materials you think are good? |
| Interviewee: 可持續性跟這個歷史問題。平常在逛的時候只會看到就是品牌國家跟那個使用材料，很少看到有這個「材料可持續性」跟「歷史問題」，所以買的時候就會覺得說，就是，哦，買完了可能就，就只是用完了也就只是垃圾，但是我沒有想到說，就是，哦，我沒有想到說就是可能面膜它裡面會有一些不明的植物萃取物跟化學保存劑。應該說是知道，可是不知道說就是它可能，嗯，會有存在風險之類的。 | Interviewee: Sustainability and this history issue. Usually when browsing, I only see the brand’s country and the materials used. I rarely see this “Material Sustainability” and “History Issues”. So when buying, I would think, just, oh, after buying it, maybe it’s just, after using it up, it’s just trash. But I didn’t think that, like, oh, I didn’t think that maybe the facial mask could contain some unknown plant extracts and chemical preservatives. I should say I knew, but I didn’t know that it might, hmm, pose potential risks, things like that. |
| Me: 那妳覺得這個，妳剛知道這個會影響妳的，妳的想法嗎？想法嗎？就是因為妳之前 (註: TXT 用 最近) 有買過這個產品，那妳剛知道好像它裡面可能有什麼化學的…妳對妳關於那個這個產品的那個品牌，妳的想法會改嗎？ | Me: So you think this, knowing this just now, will it affect your, your thoughts? Your thoughts? Because you have bought this product before (Note: TXT used “recently”), and you just found out it might contain some chemicals… Regarding this product’s brand, will your thoughts change? |
| Interviewee: 嗯，會有一點點，會有一點點。可能會去找可能更方便的吧，就是不需要用到面膜，可能只要，嗯，要怎麼說…就是有，有看到就是不需要敷面膜，就是很像就是塗那個…塗，塗那種蘆薈的那種面膜，應該可能之後會買這種。 | Interviewee: Hmm, a little bit, a little bit. Maybe I’ll look for something more convenient, like something that doesn’t require using a sheet mask, maybe just, hmm, how should I say… like, I’ve seen ones where you don’t need to apply a sheet mask, it’s like applying… applying that kind of aloe mask, maybe I’ll buy that kind later. |
| Me: 嗯嗯嗯。好，那下面有一個部分可以按，妳覺得這個，這個是什麼意思？ | Me: Mm-hmm mm-hmm. Okay, there’s a section below you can press. What do you think this, this means? |
| Interviewee: 積蓄 (註: TXT 誤聽為 金序) 嗎？嗯？假如說我不買這個東西啊，我可以存到多少錢嗎？ | Interviewee: Savings (Note: TXT misheard as “Gold Order”)? Hmm? Suppose if I don’t buy this item, can I save some amount of money? |
| Me: 好。那妳要先，呃，思考。 | Me: Okay. Then you need to first, uh, think. |
| Interviewee: 在思考…妳要吃什麼？我沒有錢吃。 …妳要給我錢，好嗎？ | Interviewee: Thinking… What do you want to eat? I have no money to eat. …You should give me money, okay? |
| Interviewee: 哦…哦！所以積蓄是那個碳排放量 (註: TXT 誤聽為 彈放量)，減少碳排放量？不會是錢？喔？然後說我不買那個面膜我可以…我可以省多少錢？喔，好貴。 | Interviewee: Oh… Oh! So savings means the carbon emissions (Note: TXT misheard as “bullet release amount”), reducing carbon emissions? Not money? Oh? And it says if I don’t buy that mask I can… I can save how much money? Oh, so expensive. |
| Me: 所以中文來說，可能這個詞 (註: TXT 誤聽為 車) 不是最適合的。就是剛剛那個詞 (註: TXT 誤聽為 車)，因為妳剛剛是想… | Me: So in Chinese, maybe this word (Note: TXT misheard as “car”) isn’t the most suitable. Just that word (Note: TXT misheard as “car”) earlier, because you were thinking… |
| Interviewee: 賺錢？ (註: TXT 誤聽為 罐牛腱/罐牛肉) | Interviewee: Making money? (Note: TXT misheard as “Canned beef tendon/Canned beef”) |
| Me: 對。 | Me: Yes. |
| Interviewee: 對對對，以為是不買這個東西，或者是可能，可能就是你買這個東西可以省多少錢，或是你不買這個東西可能會省多少錢。這個積蓄的話，我覺得…要怎麼看 (註: TXT 誤聽為 乾)？ | Interviewee: Right, right, right. I thought it was about not buying this thing, or maybe, maybe buying this thing could save how much money, or not buying this thing might save how much money. This “Savings”, I feel… how should I look at it (Note: TXT misheard as “dry”)? |
| Me: 好。那第三個部分，投資？ | Me: Okay. Then the third section, Investment? |
| Interviewee: 投資嗎？投資喔？感覺很像是我要投資這一家的品牌？哪裡啊？投資這是…投資代號？美國、美國、韓國。感覺就是，對啊，就是投資這個品牌。 | Interviewee: Investment? Investment, huh? Feels like I’m supposed to invest in this company’s brand? Where? Investment this is… investment code? USA, USA, Korea. Feels like, yeah, investing in this brand. |
| Me: 那妳會嗎？ | Me: Would you? |
| Interviewee: 會。 | Interviewee: Yes. |
| Me: 好，那妳可以幫我回第一個部分嗎？第一個，買東西。然後下面下面下面…對，有兩個按鈕。如果妳不想要按也可以。 | Me: Okay, can you help me go back to the first section? The first one, buying things. Then below, below, below… Right, there are two buttons. If you don’t want to press them, that’s okay too. |
| Interviewee: 欸，等我看一下。哇哇哇，這個有點跑掉了。嗯…哇喔。我想一下…「對關心環保 (註: TXT 誤聽為 光環保) 的消費者來說，選擇可持續性的品牌可能會更好。以下是可替代的品牌，並附上簡單的比較表」。哦…這個，這個…那個，它可以再隔一個空位嗎？對，有點看不太懂。 | Interviewee: Hey, let me take a look. Wow wow wow, this one is a bit off. Hmm… Wow oh. Let me think… “For consumers concerned about environmental protection (Note: TXT misheard as”only environmental protection”), choosing sustainable brands might be better. Below are alternative brands with a simple comparison table attached.” Oh… This, this… that, can it have another space? Right, it’s a bit hard to understand. |
| Me: 好。好。那妳覺得…妳有另外一個按鈕 (註: TXT 誤聽為 案子)，妳想要問 (註: TXT 誤聽為 玩) AI嗎？ | Me: Okay. Okay. So do you think… you have another button (Note: TXT misheard as “case”), do you want to ask (Note: TXT misheard as “play”) the AI? |
| Interviewee: AI嗎？隨便都可以。嗯…我想想看喔…「產品會浪費掉多少的碳排放量」？(註: TXT 誤聽為 網品櫃/癱瘓量) …按到了。 | Interviewee: AI? Anything is fine. Hmm… Let me think… “How much carbon emission will the product waste?” (Note: TXT misheard as “Net product counter/paralysis amount”) …Pressed it. |
| Me: 欸，等一下，按到了。 | Me: Hey, wait a minute, you pressed it. |
| Interviewee: 我本來是要按說「會浪費掉多少的碳排放量」… | Interviewee: I originally wanted to press “How much carbon emission will be wasted”… |
| Me: 哇，好清楚喔。 | Me: Wow, so clear. |
| Me: 好。那要說什麼？妳覺得最…重點是什麼？ | Me: Okay. What should I say? What do you think is the most… main point? |
| Interviewee: 重點是什麼？蠻多的捏。就是光是從生產過程就已經開始產生碳排放，然後包裝本來就是也會產生，運輸過程也不用說就是本來就會產生的，只是我沒有想到說就是，最後的使用跟處理，使用過程中的水電消耗以及產品的包裝最終處理回收或掩埋，也會影響到碳排放。 | Interviewee: What’s the main point? There’s quite a lot. Just from the production process, carbon emissions already start being generated. Then packaging itself also generates them. The transportation process goes without saying, it inherently generates them. It’s just that I didn’t think about, like, the final use and disposal, the water and electricity consumption during use, and the final disposal of the product’s packaging – recycling or landfill – would also affect carbon emissions. |
| Me: 嗯。好。那這樣就可以了。妳幫我拍照這個畫面 (註: TXT 誤聽為 玩意)，還有那個剛剛那個Momo…Momo，對。好，Momo，等一下…這個部分就可以了。好的。 | Me: Hmm. Okay. That’s fine then. Help me take a photo of this screen (Note: TXT misheard as “thingy”), and also that Momo from just now… Momo, right. Okay, Momo, wait a moment… this part is fine. Okay. |
| Interviewee: 這上面有… | Interviewee: On top here there’s… |
| Me: …那個號碼，接收號碼，這個，那個幫我寫。 | Me: …that number, the receiving number, this one, help me write that one down. |
Example Transcript 2024-12-23 - Taichung (NCHU) - 50R6E
Source Files: * 2024-12-23 - Taichung (NCHU) - 50R6E.json *
2024-12-23 - Taichung (NCHU) - 50R6E.txt
Ziran ID Code: 50R6E
| Traditional Chinese Transcript | English Translation |
|---|---|
| Interviewer: 所以，呃，基本上這個是我的論文的軟體，那大概幫你，呃，知道你買的產品是多環保。那，呃，這樣的話，我可以錄影你的聲音嗎？ | Interviewer: So, uh, basically this is the software for my thesis, it roughly helps you, uh, know how eco-friendly the products you buy are. Then, uh, in that case, can I record your audio? |
| Interviewee: 嗯嗯。 | Interviewee: Mhm. |
| Interviewer: 可以嗎？ | Interviewer: Is that okay? |
| Interviewee: 哦，我說可以可以。 | Interviewee: Oh, I said yes, yes. |
| Interviewer: 好，謝謝。呵呵。好，那你剛剛開Momo嗎？之前有用過Momo嗎？ | Interviewer: Okay, thank you. Hehe. Okay, did you just open Momo? Have you used Momo before? |
| Interviewee: 有。 | Interviewee: Yes. |
| Interviewer: 那你用Momo基本上你買什麼樣的東西？ | Interviewer: What kind of things do you basically buy using Momo? |
| Interviewee: 家電啊、保養品都有，食物也有買過。 | Interviewee: Home appliances, skincare products, I have bought food too. |
| Interviewer: 那你有什麼特別喜歡的品牌嗎？ | Interviewer: Do you have any particularly favorite brands? |
| Interviewee: 好像沒有。要看什麼產品，不一定。 | Interviewee: Seems like no. Depends on the product, not necessarily. |
| Interviewer: 那Momo上，你可以幫我找一個產品，你想要買的嗎？ | Interviewer: Then on Momo, can you help me find a product you want to buy? |
| Interviewee: (Searches) 網路有點慢…我就隨便找一個…我通常都會去看什麼特價的地方…從特價的地方看。哦，看得懂，感覺不錯耶，杯子。 | Interviewee: (Searches) The internet is a bit slow… I’ll just randomly find one… I usually go look at the special offer sections… look from the special offer section. Oh, I understand this, looks pretty good, cups. |
| Interviewer: 好。那你剛剛有找到什麼？ | Interviewer: Okay. What did you find just now? |
| Interviewee: Kitty牛奶。 | Interviewee: Kitty milk. |
| Interviewer: 嗯。那剛剛有什麼新的東西嗎？ | Interviewer: Um. Was there anything new just now? |
| Interviewee: 這個框框。 | Interviewee: This frame/box. |
| Interviewer: 嗯嗯。那這是什麼？這是… | Interviewer: Mhm. What is this? This is… |
| Interviewee: 就是肉。 | Interviewee: It’s meat. |
| Interviewer: 那剛剛那個AI幫手有跟你講什麼？ | Interviewer: What did the AI assistant tell you just now? |
| Interviewee: AI幫手，這個…買東西…他說買東西也是一種投資，我可以幫你分析花的錢是好的還是不好的…公司，然後他有介紹這間公司，然後講到ESG的評比分數。 | Interviewee: AI assistant, this… buying things… it says buying things is also a type of investment, I can help you analyze if the money spent is good or bad… company, then it introduced this company, and mentioned the ESG rating score. |
| Interviewer: 那你覺得這個內容有什麼有意思的嗎？還是你覺得可以使用嗎？ | Interviewer: Do you think there’s anything interesting in this content? Or do you think it’s usable? |
| Interviewee: 應該說平常買東西不會特別看ESG評比分數。可是就是覺得字有點…這一段，第一個框框，就是這一段字有點多，就一下子會比較難抓到重點嘛，對我來說。 | Interviewee: I should say, normally when buying things I don’t specifically look at ESG rating scores. But I just feel the words are a bit… this section, the first frame, this section has a few too many words, it makes it a bit hard to grasp the main point quickly, for me. |
| Interviewer: 那你之前有買過這個產品嗎？還是這個品牌嗎？ | Interviewer: Have you bought this product before? Or this brand? |
| Interviewee: 我有去這個品牌吃過飯，但我沒有買過他們的產品，可是我有去那邊吃過飯。 | Interviewee: I have eaten at this brand’s place, but I haven’t bought their products, but I have eaten there. |
| Interviewer: 好。那跟AI幫手，呃，它要說什麼呢？它還要講我做這個AI幫手。它還要講這個品牌的名稱、材料、可持續性、製造工程、集團。 | Interviewer: Okay. What about the AI assistant, uh, what does it want to say? It also wants to talk about me making this AI assistant. It also wants to talk about this brand’s name, materials, sustainability, manufacturing process, group. |
| Interviewee: 嗯。 | Interviewee: Um. |
| Interviewer: 那有什麼你之前不知道的嗎？ | Interviewer: Is there anything you didn’t know before? |
| Interviewee: 應該說，我有點好奇這個可持續性評比還有勞工評比是怎麼來的。嗯哼。對。然後這個我知道，因為我知道他們還沒有上市櫃。他們還沒有上市櫃？我怎麼記得他們有上過？嗯，我記得他們好像有股票代號，還是我記錯了？好好。喔，我蠻好奇他分數怎麼評的。 | Interviewee: I should say, I’m a bit curious about how this sustainability rating and labor rating came about. Uh-huh. Right. And this I know, because I know they haven’t gone public yet. They haven’t gone public? How come I remember they have? Um, I remember they seem to have a stock code, or am I remembering wrong? Okay okay. Oh, I’m quite curious how they rate the score. |
| Interviewer: 好。那上面還有會按鈕，你想要按的嗎？（環保省錢, etc.） | Interviewer: Okay. There are also buttons above, do you want to press any? (Eco-friendly & Save Money, etc.) |
| Interviewee: (Clicks ‘環保省錢’) 王品集團。我想要環保省錢。嗯。從哪可以增加我的儲蓄？嗯。它是按這個會有反應嗎？嗯，有有有。還是要等一下？ | Interviewee: (Clicks ‘環保省錢’) Wang Steak Group. I want to be eco-friendly and save money. Um. From where can I increase my savings? Um. Does clicking this trigger a response? Um, yes yes yes. Or do I need to wait? |
| Interviewer: 你要計時一下。 | Interviewer: You need to time it a bit. |
| Interviewee: 哦。未來，特別是妳想要考研究所？為什麼他會說特別是我想考研究所？哦，有點好奇。自訂預算，OK。減少目標？凱子？好的。我以為他會跟這個產品有關。所以跟這個牌子沒有什麼關係。哦。 | Interviewee: Oh. Future, especially if you want to apply for graduate school? Why would it say especially if I want to apply for graduate school? Oh, a bit curious. Custom budget, OK. Reduce target? Spendthrift? Okay. I thought it would be related to this product. So it has nothing to do with this brand. Oh. |
| Interviewer: 那有什麼其他的事情想問他嗎？ | Interviewer: Is there anything else you want to ask it? |
| Interviewee: (Doesn’t ask) | Interviewee: (Doesn’t ask) |
| Interviewer: 不然看一下環保總碳排。應該可以省錢一下。還是？ | Interviewer: Otherwise, take a look at the total carbon footprint. Should be able to save some money. Or? |
| Interviewee: (Clicks ‘替代品’) | Interviewee: (Clicks ‘替代品’) |
| Interviewer: 當然可以。那剛剛你覺得他給你回覆有什麼有趣的嗎？什麼有意思的嗎？ | Interviewer: Of course. Did you find anything interesting in its response just now? Anything meaningful? |
| Interviewee: 有意思的嗎？好像，我覺得好像還好。但是我知道一些新的品牌。可是我不太想要植物性的肉類比，類比。 | Interviewee: Meaningful? Seems, I think it seems okay. But I learned about some new brands. However, I don’t really want plant-based meat analogues, analogues. |
| Interviewer: 好，你就知道一些新品。 | Interviewer: Okay, so you know some new products. |
| Interviewee: 對。 | Interviewee: Right. |
| Interviewer: 那可以幫我回到Momo那邊？ | Interviewer: Can you help me go back to the Momo page? |
| Interviewee: (Goes back) | Interviewee: (Goes back) |
| Interviewer: 上面，最上面還有這些部分。（綠色 & 紫色 tab） | Interviewer: Above, at the very top, there are still these sections. (Green & purple tabs) |
| Interviewee: (Clicks green tab) | Interviewee: (Clicks green tab) |
| Interviewer: 你覺得這個什麼？ | Interviewer: What do you think this is? |
| Interviewee: 就是那個應該講在講碳足跡還是什麼的嗎？可是我不知道他的，就是這個條件是跟什麼東西去做相比。就我蠻問號的。對。 | Interviewee: It’s that… it should be talking about carbon footprint or something? But I don’t know its… like, what this condition is being compared to. I’m quite puzzled. Right. |
| Interviewer: 好。那最後一個部分。（Clicks purple tab）投資。哦。 | Interviewer: Okay. Then the last section. (Clicks purple tab) Investment. Oh. |
| Interviewee: 這是什麼？就是講這間公司有它的股票代號。 | Interviewee: What is this? It just says this company has its stock code. |
| Interviewer: 好。OK。那差不多這樣子。可以幫我拍照你覺得最重要？就是剛剛這個內容的重點是什麼？可以幫我拍那個… 等一下。你你你幫我填那個，呃，問卷那邊會需要照片。 | Interviewer: Okay. OK. Then that’s about it. Can you help me take a photo of what you think is most important? Like, what was the main point of the content just now? Can you help me take a picture of that… Wait a minute. You you you help me fill out that, uh, the questionnaire part will need the photo. |
| Interviewee: 哦，那我就拍，選一個拍就好嗎？ | Interviewee: Oh, then I’ll just take a picture, just pick one to take? |
| Interviewer: 對。 | Interviewer: Right. |
| Interviewee: 那我拍這個。 | Interviewee: Then I’ll take this one. |
| Interviewer: 好。那上面還有一個號碼，還有一個，不是，剛剛，剛剛那個Momo，Momo那邊。Momo。對對對。最上面。那你可這樣出來，這個號碼嗎？ | Interviewer: Okay. There’s a number at the top, also one, no, just now, that Momo one, over at Momo. Momo. Right right right. At the very top. Can you get it out like this, this number? |
| Interviewee: 50R6E。 | Interviewee: 50R6E. |
| Interviewer: 好。OK。好。那這樣就可以了。等下應該會，這邊。那，啊，不是不是。剛剛那個，對。那最下面有一個問卷可以寫這個號碼，剛剛那個號碼。對。然後幫我剪，對。這樣都是另一個。好，謝謝。 | Interviewer: Okay. OK. Okay. Then that’s fine. Later it should, here. Then, ah, no no. That one just now, right. Then at the very bottom there’s a questionnaire where you can write this number, the number from just now. Right. Then help me cut, right. That’s all another one. Okay, thank you. |
Example Transcript 2025-02-12 - Pingtung (NPUST) - CO2N2
Source Files: * 2025-02-12 - Pingtung (NPUST) - CO2N2.json *
2025-02-12 - Pingtung (NPUST) - CO2N2.txt
Ziran ID Code: CO2N2
| Traditional Chinese Transcript | English Translation |
|---|---|
| Interviewer: 你同意我錄音我們的對話嗎？ | Interviewer: Do you agree to let me record our conversation? |
| Interviewee: 可以同意。 | Interviewee: Yes, I agree. |
| Interviewer: 好。那基本上我是臺南應用科大設計系的碩班的學生。這個是我的論文的 app，那我要給你介紹這個 app，好，問你一些問題，可以嗎？ | Interviewer: Okay. Basically, I am a master’s student from the Department of Design at Tainan University of Technology. This is the app for my thesis. I want to introduce this app to you, okay, and ask you some questions, is that okay? |
| Interviewee: 可以。 | Interviewee: Yes. |
| Interviewer: 好。呃，你先幫我開那個 momo。你之前有用過 momo 嗎？ | Interviewer: Okay. Uh, first help me open that momo. Have you used momo before? |
| Interviewee: 沒有。 | Interviewee: No. |
| Interviewer: 啊。那其他的可以買東西的平台嗎？例如是 Shopee？ | Interviewer: Ah. What about other platforms for buying things? For example, Shopee? |
| Interviewee: Shopee 有。 | Interviewee: Shopee, yes. |
| Interviewer: 好。那 Shopee 的周圍還有其他嗎？還是 Shopee 而已？ | Interviewer: Okay. Besides Shopee, are there others? Or just Shopee? |
| Interviewee: 嗯，Shopee 而已。 | Interviewee: Um, just Shopee. |
| Interviewer: 好。那你 Shopee 商品上會買什麼？什麼樣的東西？ | Interviewer: Okay. What do you buy on Shopee? What kind of things? |
| Interviewee: 嗯…貓咪寵物的用品。 | Interviewee: Um… cat pet supplies. |
| Interviewer: 好。那我們這次使用 momo 一下。那你目前有什麼你想要買的嗎？ | Interviewer: Okay. Let’s use momo this time. Is there anything you currently want to buy? |
| Interviewee: 嗯…好，貓砂。 | Interviewee: Um… okay, cat litter. |
| Interviewer: 可以啊。我什麼都可以打嗎？ | Interviewer: Yes, you can. Can I type anything? |
| Interviewee: 你可以直接打。對。 | Interviewee: You can type directly. Yes. |
| Interviewer: (Types) 好。 | Interviewer: (Types) Okay. |
| Interviewee: (Browses) | Interviewee: (Browses) |
| Interviewer: 你有找到嗎？嗯。貓砂…哪一個？ | Interviewer: Did you find it? Um. Cat litter… which one? |
| Interviewee: 這裡？你有想要買這種的嗎？ | Interviewee: Here? Do you want to buy this type? |
| Interviewer: 這個。好。 | Interviewer: This one. Okay. |
| Interviewer: 那是哪一個？ | Interviewer: Which one is that? |
| Interviewee: 我要點嗎？ | Interviewee: Should I click? |
| Interviewer: 嗯。 | Interviewer: Um. |
| Interviewee: 好。(Clicks) | Interviewee: Okay. (Clicks) |
| Interviewer: 好。你可以講出來這是什麼？ | Interviewer: Okay. Can you say what this is? |
| Interviewee: 這個是混合貓砂。 | Interviewee: This is mixed cat litter. |
| Interviewer: 那這個品牌你有買過嗎？ | Interviewer: Have you bought this brand before? |
| Interviewee: 嗯…沒有。很像我買的，但是沒有買過。 | Interviewee: Um… no. It looks like the one I buy, but I haven’t bought this one. |
| Interviewer: 好。那這個網頁上面，你覺得哪一個部分最重要的？對你來說。 | Interviewer: Okay. On this webpage, which part do you think is most important? For you. |
| Interviewee: 最重要？圖片。 | Interviewee: Most important? Pictures. |
| Interviewer: 圖片。為什麼？ | Interviewer: Pictures. Why? |
| Interviewee: 因為我可以最直接看到我要買的商品。 | Interviewee: Because I can most directly see the product I want to buy. |
| Interviewer: 好。那還有其他的嗎？ | Interviewer: Okay. Are there others? |
| Interviewee: 嗯…你要參考一下？價格。 | Interviewee: Um… need to refer to it? Price. |
| Interviewer: 好。那第三個你覺得重要的是什麼？ | Interviewer: Okay. What’s the third thing you think is important? |
| Interviewee: 成分。 | Interviewee: Ingredients. |
| Interviewer: 成分、材料。好。那這裡這個成分它說什麼？ | Interviewer: Ingredients, materials. Okay. What do the ingredients here say? |
| Interviewee: 這個成…成分是豆渣、澱粉、膨潤土、咖啡渣、礦砂等等。 | Interviewee: This ingre… ingredients are soybean residue, starch, bentonite, coffee grounds, mineral sand, etc. |
| Interviewer: 那你覺得這個可以嗎？ | Interviewer: Do you think this is okay? |
| Interviewee: 嗯…跟我平常用的不太一樣。 | Interviewee: Um… it’s a bit different from what I usually use. |
| Interviewer: 哦？為什麼不一樣？ | Interviewer: Oh? Why is it different? |
| Interviewee: 嗯…我的成分比較簡單，這個加了很多不同的東西。 | Interviewee: Um… mine has simpler ingredients, this one adds many different things. |
| Interviewer: 啊。那你可以看下面一點嗎？還有什麼其他的有趣的嗎？有興趣的嗎？ | Interviewer: Ah. Can you look down a bit? Is there anything else interesting? Anything you’re interested in? |
| Interviewee: 沒有。 | Interviewee: No. |
| Interviewer: 好。那你覺得剛剛那個黃色的，它是這個網頁的一部分嗎？ | Interviewer: Okay. That yellow part just now, do you think it’s part of this webpage? |
| Interviewee: 看起來不是。 | Interviewee: Doesn’t look like it. |
| Interviewer: 好。為什麼？ | Interviewer: Okay. Why? |
| Interviewee: 因為這裡是白色的，這個黃色，然後有特別的框框。 | Interviewee: Because here it’s white, this is yellow, and it has a special frame. |
| Interviewer: 好。那最上面，你可以看一下有三個不一樣的部分？ | Interviewer: Okay. At the very top, can you take a look, there are three different sections? |
| Interviewee: (Looks) | Interviewee: (Looks) |
| Interviewer: 可以幫我按第二個？ | Interviewer: Can you help me press the second one? |
| Interviewee: (Clicks) | Interviewee: (Clicks) |
| Interviewer: 你覺得這是什麼？ | Interviewer: What do you think this is? |
| Interviewee: 它的材料跟環保的關係。 | Interviewee: Its materials and the relationship with environmental protection. |
| Interviewer: 啊。其他的品牌你有買過嗎？你有聽說過嗎？ | Interviewer: Ah. Have you bought the other brands? Have you heard of them? |
| Interviewee: 沒有。其實我才養貓耶，兩三個月。 | Interviewee: No. Actually, I just got a cat, only two or three months ago. |
| Interviewer: 啊。那你對這些品牌有興趣嗎？ | Interviewer: Ah. Are you interested in these brands? |
| Interviewee: 這個…因為看不懂…感覺不一樣。 | Interviewee: This one… because I don’t understand… feels different. |
| Interviewer: 你講出來是哪一個？ | Interviewer: Which one are you pointing out? |
| Interviewee: OK，應該是 EcoCat。 | Interviewee: OK, should be EcoCat. |
| Interviewer: 好。那最上面的第三個部分？ | Interviewer: Okay. Then the third section at the top? |
| Interviewee: 投資？ | Interviewee: Investment? |
| Interviewer: 你覺得這是什麼？ | Interviewer: What do you think this is? |
| Interviewee: 哦，這是有關這個商品的公司。 | Interviewee: Oh, this is about the company of this product. |
| Interviewer: 嗯。它說什麼？ | Interviewer: Um. What does it say? |
| Interviewee: 它說它有代號，還有國家跟公司的名稱。 | Interviewee: It says it has a code, and the country and company name. |
| Interviewer: 那這個會影響你的想法嗎？ | Interviewer: Would this affect your thoughts? |
| Interviewee: 可能會。 | Interviewee: Maybe. |
| Interviewer: 為什麼？ | Interviewer: Why? |
| Interviewee: 因為有些人不買中國的東西，所以有些人會因為這個公司有不好的事情，他就會影響他要不要買。 | Interviewee: Because some people don’t buy things from China, so some people, because this company has had bad incidents, it will affect whether they buy or not. |
| Interviewer: 哦，好。那… OK。所以這個商品是從中國來的？ | Interviewer: Oh, okay. So… OK. So this product is from China? |
| Interviewee: 對。 | Interviewee: Yes. |
| Interviewer: 好。那…好。那你可以幫我回第一個部分那邊嗎？這個？那最上面有一個號碼，可以幫我寫這個號碼在卡片上？ | Interviewer: Okay. Then… okay. Can you help me go back to the first section there? This one? Then at the very top, there’s a number, can you help me write this number on the card? |
| Interviewee: 這裡啊？就是這個號碼？ | Interviewee: Here? Just this number? |
| Interviewer: 那你可以幫我講出來嗎？ | Interviewer: Can you help me say it out loud? |
| Interviewee: C-O-Z-N-2。 | Interviewee: C-O-Z-N-2. |
| Interviewer: 好。這個都是匿名的，所以這樣我可以知道。好。那你可以幫我取下面一點嗎？好。所以下面有…有兩個案子？你可以幫我按一下按鈕？對。 | Interviewer: Okay. This is all anonymous, so this way I can know. Okay. Can you help me go down a bit? Okay. So below there are… there are two cases? Can you help me press the button? Yes. |
| Interviewee: (Clicks button) 然後這裡還有什麼？ | Interviewee: (Clicks button) Then what else is here? |
| Interviewer: 嗯。 | Interviewer: Um. |
| Interviewee: 啊…我點進來之後，它就幫我顯示出我的商品，嗯，品牌，然後它就給我其他建議的品牌。不同的。 | Interviewee: Ah… after I clicked in, it showed me my product, um, the brand, and then it gave me other recommended brands. Different ones. |
| Interviewer: 哦，那這些你有看過嗎？之前有看過嗎？ | Interviewer: Oh, have you seen these before? Have you seen them previously? |
| Interviewee: 沒有。就不同的成分，也不同的牌子。 | Interviewee: No. Just different ingredients, and different brands. |
| Interviewer: 那你還有什麼其他的問題，你想要問他嗎？這個 AI？ | Interviewer: Do you have any other questions you want to ask it? This AI? |
| Interviewee: 嗯…可以點這個嗎？ | Interviewee: Um… can I click this? |
| Interviewer: 你可以點，還是也可以直接直接打字。 | Interviewer: You can click, or you can also directly type. |
| Interviewee: 啊…我怎麼減少亂買？(Clicks button, scrolls) | Interviewee: Ah… How do I reduce impulse buying? (Clicks button, scrolls) |
| Interviewer: 你要去上面一點吧？哦，有有。有。哦有了。 | Interviewer: You need to go up a bit? Oh, yes yes. Yes. Oh, got it. |
| Interviewee: (Scrolls up) 點兩次了。嗯。 | Interviewee: (Scrolls up) Clicked twice. Um. |
| Interviewer: 他就會給你一些建議。 | Interviewer: It will give you some suggestions. |
| Interviewee: 制定預算、列清單、三思而後行。 | Interviewee: Set a budget, make a list, think twice before acting. |
| Interviewer: 那你覺得這裡面的重點是什麼？ | Interviewer: What do you think is the main point in here? |
| Interviewee: 嗯…他會針對我的問題回答。 | Interviewee: Um… it answers my question specifically. |
| Interviewer: 嗯。然後你說他的回答的重點嗎？ | Interviewer: Um. And you’re saying the main point of its answer? |
| Interviewee: 對。嗯…要想好自己到底要的是什麼。 | Interviewee: Yes. Um… need to think clearly about what one really wants. |
| Interviewer: 好。OK。你還有其他的問題嗎？ | Interviewer: Okay. OK. Do you have any other questions? |
| Interviewee: 沒有。 | Interviewee: No. |
| Interviewer: 好。那這樣這樣可以了。好，謝謝你。謝謝。好嗎？ | Interviewer: Okay. Then this is fine. Okay, thank you. Thank you. Okay? |
| Interviewee: (Nods) | Interviewee: (Nods) |
Appendix 2: College Students’ Prototype Self-Testing Survey Data
Due to the large amount data captured by the survey (over 100 people self-tested the Green Filter app and 32 people filled out the survey after in-person testing was completed), I’ve chosen to exclude it from the PDF version of my thesis. The data is available upon request - and has also been published on Hugging Face as a public dataset here: https://huggingface.co/datasets/krishaamer/taiwanese-college-students
Appendix 3: Expert Interviews’ Transcripts
Due to the length, I’ve decided to exclude full raw interview transcripts. The data is available upon request. Below is one example interview transcript.
Example Expert Interview 2024-05-19 - Chen-Ying Huang
MA Thesis: “Green Filter” (綠濾) (2024‑05‑19 11:07 GMT+8) – Transcript
Attendees: Chen‑Ying Huang, Fireflies.ai Notetaker, Kris Haamer
This editable transcript was computer‑generated and might contain errors. People can also change the text after it was created.
Kris Haamer: Yeah, so let me repeat the question so I can get your answer into this recording.
Chen‑Ying Huang: Okay.
Kris Haamer: So we’re discussing the flow of the user and how it affects the decisions,…
Chen‑Ying Huang: right So basically I’m thinking that people are lazy.
Kris Haamer: right?
Chen‑Ying Huang: I look at myself. I’m really lazy when shopping and so you should if it’s easier to get the information that I don’t have to click on the button to receive the information. It’s more likely that I will even pay attention to that and…
Kris Haamer: Right, right.
Chen‑Ying Huang: also that words are very difficult to read through so I don’t really have incentive to read a lot but if it’s just a picture showing that the product somehow is green — like saving whatever picture you can put on the iceberg or the polar bear or whatever — that will immediately let me realize that this product is green and maybe there could be a scale like how green it is from zero to a hundred and then I can immediately know how green this product is.
Kris Haamer: even…
Chen‑Ying Huang: Maybe there would be an effect on me. So I’m thinking more along this line.
Kris Haamer: right, right, visually understandable or really quickly understandable images instead of this kind of report.
Chen‑Ying Huang: right. Yeah.
Kris Haamer: Yeah. I haven’t tested that. How this research started was from my own interest — I’m kind of a nerd. I’ve always loved Star Trek; they have medical professionals with a scanner that can scan your whole body just by waving it and then understand what’s wrong with you. So every time I bought some product I felt like I want this kind of scanner just to give me quick information about whether this is a good product, and I thought I’d try to build something like that.
Chen‑Ying Huang: Okay.
Kris Haamer: But then I realized if I actually want to make a working prototype it’s really difficult on a mobile phone, even though that would be the obvious choice, because the Apple ecosystem is closed, so you can’t really do a lot. On the other hand, the Google Chrome browser is very open, so I can take any website like Momo and basically rewrite the page using an extension.
Chen‑Ying Huang: Okay.
Kris Haamer: That’s how I ended up focusing on online shopping, Momo in particular. And then getting it to technically work‑—AI is slow, that’s the main challenge. If I pre‑bake these pages and test them, I could do pre‑made images or pre‑made reports. But here it’s actually working in real time: you can go to any page on Momo and it will do this analysis using actual data. That’s why there’s just a button instead of a pre‑rendered overlay. And then a bit further down I came up with this kind of scoring…
Kris Haamer: Because my intuition was that maybe the social aspects of a product — what other people who bought it think — could also be relevant. But so far, among the 20 people who actually saw this, they didn’t honestly understand what these numbers were or why they were there; none of them understood it. Then if you go further down there’s another button that takes you to a new page and you get to the actual app part, which is much bigger and that I have more control over than a Momo page. Even though Chrome is quite open there are still lots of limitations on, for example, how to track the user. So if you come to my page I can track where you scroll, where your cursor goes, and I can use that data.
00:05:00
Chen‑Ying Huang: Wait, so you’re not randomly sampling people from Momo, right? You’re asking subjects to participate in your experiment. So in that sense…
Kris Haamer: Yes, yes.
Chen‑Ying Huang: Wouldn’t it be possible… You want a natural environment where they’re shopping? Otherwise, why wouldn’t it be possible for them to read some information about the product here — the information you provide — before they really look at the specification of the product itself?
Kris Haamer: Mmm. Yes. Because I’m from the design department, our approach to experiments isn’t very scientific. Design thinking is kind of quasi‑research: you do workshops, get some people, try to use this Double Diamond format to get insights, but it’s not scientific at all. It mostly gives you keywords or intuitions. I tried to do better and ran a survey — my Chinese isn’t good enough for in‑depth interviews — so I focused on a survey. It’s 63 questions so I get a lot of data. Then I use k‑means (or actually k‑prototypes) to cluster it and create personas based on similarity. Maybe I’ll just share that page…
Chen‑Ying Huang: Mm‑hmm, yeah.
Kris Haamer: It’s easier to talk about if you can see it.
Chen‑Ying Huang: So from the survey data…
Kris Haamer: Right. So these are the three personas that came out of the data…
Chen‑Ying Huang: Okay, okay.
Kris Haamer: There are different ways to do it, but basically one persona really cares about the environment, one is frugal, one is more parent‑like, etc. There’s overlap: e.g. product origin is relevant for most respondents, materials are more important for the “parently” persona, price for the frugal persona. The social score didn’t resonate — people didn’t understand or didn’t care. So after all that I still need to test what I built. My professor suggested one‑on‑one user tests with the live extension and see how far people get…
00:10:00
Kris Haamer: If I go back to the original idea of the research: maybe the product page is an entry point, and I could create a funnel and get people to invest in that company in some way. Investing in a broad sense: either buying the product, or becoming a shareholder…
Chen‑Ying Huang: No…
Kris Haamer: But I’m realizing this is way too big for a master’s degree — shopping, saving, investing are huge fields. Maybe I was naïve — no finance background, thinking purely in terms of UI. I thought I could build this funnel…
Chen‑Ying Huang: It’s still good to have a dream.
Kris Haamer: But now I see that I can’t even get people from the Momo page into my app — most drop at the first step. So maybe the big takeaway is that it’s really hard to get people from point zero to that far‑away place.
Kris Haamer: Coming back to the app: there’s shopping, saving, investing. I tried to build a friendly ChatGPT‑like assistant here…
Chen‑Ying Huang: A fun question is really interesting: “Where can I buy those products in Tainan?” I think that’s useful.
Kris Haamer: Those questions are based on the survey. But most testers didn’t know what to ask.
Kris Haamer: I tried to localize: the more precise and relevant you are, the higher the conversion. And this page (not ready yet) will generate e.g. a factory map so you can see where the product comes from, maybe even pictures from the factory. Particularly for food: milk, meat—you could see distance and animal conditions. Maybe livestreams. But that data is hard to get, so for now it’s a mock‑up.
00:15:00
Kris Haamer: Because this is a journey, I added sections where you can set a financial goal. In testing, students gave very standard answers (get a master’s, get a job), so it wasn’t so useful. I thought the AI could use a specific goal to give advice. Then I had a stock portfolio section with real data. I wanted ESG data live, but that costs USD 4 000/month — impossible for consumers. Only some companies publish free ESG data, so I can only do a mock‑up with last‑year numbers. Rabbit holes everywhere…
Kris Haamer: I mentioned degrowth, doughnut economics, etc., and wanted an economist’s view. That’s why I wanted this call.
Chen‑Ying Huang: From your email I thought I could maybe help a little with your first goal — influencing shopping behaviour. The other goals are too big. My intuition is: make the information relevant and simple. Then you have a better chance. Sometimes introducing an obviously worse decoy option works: if you show a product that’s clearly worse environmentally, people choose the target one…
Kris Haamer: That would be a product comparison, maybe a sidebar with alternatives and different scores.
Chen‑Ying Huang: If the company spends part of its profit on green initiatives, that’s costless to the consumer but may sway them, assuming price is competitive. But detailed information like your report targets experts, not rushing consumers.
Kris Haamer: Right. I need to make it obvious, simpler.
Chen‑Ying Huang: Key still: spec, price, reputation/durability. Green pages must have figured out something — have you looked at sites selling only green products?
00:20:00
Kris Haamer: Yes, there are green shopping malls, but you already have to be a green consumer to visit them. That’s why I focus on mainstream sites like Momo/Shopee. Layout on Shopee is harder to inject an AI box. Even Momo’s sidebar catches people’s eyes; maybe I should target that. On Shopee the layout is too cramped.
Chen‑Ying Huang: On those green pages, do they show which product is greener? If all products are “green,” there’s no comparison.
Kris Haamer: They don’t really do comparisons; they just curate. There are rating sites (like Fashion Transparency Index) but if I focused only on fashion I could be more precise. Testing then becomes: “Look for a brand of clothes you like.” For now I let users pick any product — but many pick luxury items they can’t afford, so how useful is that?
Chen‑Ying Huang: It’s not a purchasing experiment then.
Kris Haamer: Exactly. They look at aspirational items. Then why read a sustainability report if you won’t buy it?
Kris Haamer: The way I test now isn’t working. I want to redesign before semester ends in June. My idea: since the survey worked online, ask people to download the extension themselves. They fill out the survey first (so I know the persona) and then use the extension.
Chen‑Ying Huang: So subjects do the survey, then the extension. One green version, one control version. Randomly assign. Compare behaviour. If there’s a difference, it’s the app. Step by step.
Kris Haamer: Right. I realise that means losing my big funnel vision, but maybe it’s fine.
00:25:00
Chen‑Ying Huang: Show me the first click? Can we go back to Momo… Where’s your button placed?
Kris Haamer: Right now it loads here… People miss it, so I plan to move it next to the price. Most people look at the price. About half click without prompting; with prompting, nearly all do.
Chen‑Ying Huang: That’s okay. Next to the price is good.
Kris Haamer: Internet is slow here in Macau…
Chen‑Ying Huang: It’s impressive you put this into action. If you can increase the click‑through rate, that’s key. Maybe shorten the info so people actually read it.
Kris Haamer: The report changes every time; it’s dynamic. Eye‑trackers could show what people read, but that’s overkill. Maybe mouse tracking is enough. I can only track on my own page, not on Momo due to Chrome restrictions.
00:30:00
Chen‑Ying Huang: Maybe simplify each step so if a user really wants to proceed, they have to click, and you just track clicks. Two extensions: green vs. control.
Kris Haamer: Yes. I’m already over my head; adding eye‑tracking is too much.
Chen‑Ying Huang: A quick mouse tracker is fine. And talk to a marketing expert: I have a colleague at NTU (international business) who does field experiments like this. She could help with funnel optimisation. And another colleague who does climate-change econometrics.
Kris Haamer: That would be amazing.
00:35:00
Chen‑Ying Huang: I’ll email introduce you. Meanwhile, focus on click‑through rate. That’s where your data comes from.
Kris Haamer: Yes, my teachers need a grounded story — economics theory, why focus on this angle.
Chen‑Ying Huang: Motivation, yes, but I’d worry more about the funnel metrics.
Kris Haamer: Changing one sentence in Chinese on the Momo page could be key — some testers said it sounds awkward now. I tried to stay neutral and not preach sustainability, so non‑green users might still click and learn.
Chen‑Ying Huang: GPT can help rewrite. People care about money; phrase it that way.
00:40:00
Kris Haamer: I’ve had 50 versions of that tooltip already…
Chen‑Ying Huang: Marketing colleague can help. Shorten content so there’s a chance users read it. Probably more influence.
Kris Haamer: The report ID lets me track which version they saw. But on Momo I can’t track anything else.
Chen‑Ying Huang: Simplify the page; I wouldn’t read that many words.
Kris Haamer: Right.
00:45:00
Kris Haamer: This has been super helpful. Thank you.
Chen‑Ying Huang: I’ll email the two colleagues and copy you. The econometrician does climate change, the marketing professor does funnel experiments.
Kris Haamer: Perfect. I’m wrapping this into a presentation for my thesis defence — too many things, but I need to ground it.
Chen‑Ying Huang: I’d be happy if my student came up with an app like this.
Kris Haamer: Thanks so much.
Chen‑Ying Huang: I’ll email later. Bye!
Kris Haamer: Thank you. Bye‑bye.
Meeting ended after 00:48:18 👋
Appendix 4: College Students’ Survey Data
Due to the large amounts of data captured by the initial survey (over 1600 people started the survey, after filtering, leading to over 800 Gen-Z current college students finished it fully), I’ve chosen to exclude it from the PDF version of my thesis. The data is available upon request - and has also been published on Hugging Face as a public dataset here: https://huggingface.co/datasets/krishaamer/taiwanese-college-students
Appendix 5: Research Tech Stack
For anyone who would like to repeat this research, here’s a table with the software that this research project made use of.
| Tool/Service | Category |
|---|---|
| Matplotlib | Data Visualization |
| NumPy | Scientific Computing |
| jupyter | Development Environment |
| seaborn | Data Visualization |
| Python | Programming Language |
| VSCode | Code Editor |
| GitHub | Version Control |
| fireflies.ai | AI Meeting Assistant |
| Google Meet | Communication |
| databricks | Data Platform |
| HuggingFace | Machine Learning |
| TypeScript | Programming Language |
| Tally | Form Builder |
| Zotero | Reference Manager |
| Quarto | Publishing/Documentation |
| Tool/Service | Category |
|---|---|
| Apache Parquet | Data Storage Format |
| Haystack | Search Framework |
| MongoDB | NoSQL Database |
| langchain | AI/ML Development |
| GitHub Copilot | AI Coding Assistant |
| Figma | Design Tool |
| Sketch | Design Tool |
| OpenAI ChatGPT | AI Assistant |
| Anthropic Claude | AI Assistant |
| Google Gemini | AI Assistant |
| Google Colab | Development Environment |
| Google Sheets | Spreadsheet |
| META Llama | AI Model |
| Mistral | AI Model |
| OpenAI API | API Service |
| Postman | API Testing Tool |
| Tool/Service | Category |
|---|---|
| Next.js | Web Framework |
| fullstory | Analytics |
| Polygon.io | Financial Data API |
| Hotjar | Behavior Analytics |
| Lottie | Animation Library |
| Docusaurus | Documentation |
| Google Analytics | Web Analytics |
| Google Maps | Maps/Location Service |
| vis.gl | Data Visualization Framework |
| Pinecone | Vector Database |
| Vercel | Deployment Platform |
| Vercel AI SDK | AI SDK |
| Radix UI | UI Library |
| highcharts | Data Visualization |
| React | Web Framework |
| Tailwind | CSS Framework |
| Markdown | Markup Language |
Appendix 6: Further Reading
Some recommended books on the topics covered in this thesis include, but are not limited to:
- R. Buckminster Fuller “Operating Manual for Spaceship Earth”
- Victor Papanek “Design for the Real World”
- Jonathan Chapman “Emotionally Durable Design”
- Carlo Vezzoli “Product-Service System Design for Sustainability”
- Ezio Manzini “Design, When Everybody Designs”.