Deciding Which and How Much Data to Illustrate

Chapter 12

第十二章

Deciding Which and How Much Data to Illustrate

決定哪些資料需要呈現

By Margie Henry

Let’s lay some groundwork for successful data presentation. If done thoughtfully, it will go a long way in helping you determine which aspects of your data to visualize and how. We’ll begin with a little brainstorming. You can do this in just a few thoughtful moments alone or working as a team. Your work here is two-fold: define your message and define your intended audience. You can flip this sequence around, but we’ll begin with defining your message.

讓我們為了有效的數據奠定一些基礎。如果(想要)精準完成(這件事情),將會(花費)很長一段時間以幫助您確定哪個方面的資料(需要被)可視化及如何(可視化)。 (讓)我們從一個簡單的腦力激盪開始吧。經過些許思辯(thoughtful moments),您可以獨自或在團隊中完成它。您的任務有兩個(方面):定義您的(期望)訊息以及定義您的預期觀眾。您可以認為決定其順序,但這裡我們將先從定義您的(期望)訊息開始。

Determine Your Message

決定您的(期望)訊息

Before tackling which data to present, take a few minutes to decide what you want to say. Close down your latest social media craze, step back from your computer, and consider the exact message you want to communicate. Ask yourself, “What do I know, what does it mean, and why do I believe it’s important?”

在暫定(tackling)您所想要呈現的資料之前,花幾分鐘決定您想表達(甚麼)。關掉(那些)社群軟體(譯者案:如Facebook,LINE等),暫時離開(step back)您的電腦,然後考慮您所想溝通的確切訊息。捫心自問“我所知悉的是甚麼,(分別)代表甚麼含義,以及為何(會被)我認為這是重要的?”

Consider a dataset containing observations on different types of caffeinated beverages and the effects of their consumption. Don’t stop at “caffeine affects the body.” You never want to present information that solicits a “well, duh” response. Dig deeper. Be more specific. What do your data say about how caffeine affects the body? Are the effects all good, all bad, or maybe an interesting combination of both? Do the effects change with a person’s age and/or sex? Are some caffeinated beverages better or worse for overall health? Your answer should be concise: short, sweet, and to the point. A statement such as “Coffee has an ability to reduce the risk of certain diseases and ailments when consumed in moderation because it contains key antioxidants.” goes a lot further than our original example. Even better, it establishes a pretty clear focus for our visuals and some common language to use with our audience.

(請)考慮一個資料集(例子)涵蓋了不同類型的(含)咖啡因飲品以及他們的銷售情況。切勿因“咖啡因影響身體(健康)。”(而停止思考)。您絕對不想在經過呈現資訊(後)得到 “咄~,”(原文為:well, duh。有敷衍的涵義)之類的反應。(再)深入探討。更具體(地說)。你的資料有說明咖啡因是如何影響人體的嗎? 這些影響全部都是好的,(還是)全部都是壞的,又或者是兩者有某些趣味的集合?影響會隨者(使用者)的年紀或性別有所不同嗎?對整體健康而言,含咖啡因飲品是較好還是較壞的影響? 您的回答必須是簡潔:簡短、貼心、扼要。例如“適量應用咖啡,可以降低罹患疾病的風險,因為它包含關鍵的抗氧化劑” 比我們先前的例子更進一步。更棒的是,它為我們的視覺化建立了明確的重點,並提供給觀眾共同語言(了解這件事)。

Right about now you should be having a flashback to English 101. That’s because determining your key message is just like writing a good thesis statement. If you can’t summarize your key message in a few concise sentences then you probably need a better grasp of the topic. Sound harsh? Maybe, but not as harsh as presenting information to a crowd of your yawning disinterested peers. Fight the urge to skip this step! If you’re the paper-and-pencil type, go ahead and write your message down! You can use it as a reference throughout your data visualization process.

現在您或許有種回到英文入門的感受。這是因為決定您的關鍵訊息就像是在些一篇好的論文一樣。如果您無法用幾句簡潔扼要的句子概述您的觀點時,(建議)您(還是)好好理解(所選的)主題。聽起來很殘忍吧?或許,但是總比您在報告時,一大群(公正的)同伴正在打哈欠還不殘忍吧。儘全力完成這個步驟吧!如果您是(習慣)用紙筆的話,(那請)繼續寫下您的想法(message)!您可以將其作為整個資料可視化(data visualization)處理的過程

Simply put, your chances of creating a compelling, well-organized visual argument are immeasurably greater if you begin with a clear and focused message.

簡而言之,您創造了引人注意的機會,若您一開始就有明確且聚焦的訊息,則組織良好的視覺化參數將無可限量。

Understand Your Audience

了解你的觀眾

You’ve determined your message. Let’s now consider the importance of understanding your audience. This knowledge will go just as far in helping you determine which and how much of your data to illustrate.

您(已經)決定了你的(期望)訊息。現在,讓我們來仔細思考了解您的觀眾是多麼重要。 到目前為止的這些知識將協助您確認您的資料中哪些是(需要被)圖示說明的。

Take another couple of minutes and ask yourself “what information is most valuable to my audience,” “what role will my visuals play in this dialogue,” and “what action(s) do I want to incite?” Would you spend time explaining algebra to a group of engineers? (The correct answer is no.) What would be the point? The better you know your audience, the better your chances of creating a successful visual presentation.

再花幾分鐘問自己 “甚麼樣的資訊是你的觀眾(覺得)最有價值的,” “在這次的對話中,您的圖示扮演甚麼樣的腳色,” 以及 “您所策畫的目的?” 您會花時間和一群工程師解釋線性代數嗎? (標準答案是'否')那甚麼將會是重點呢?您越了解您的觀眾,您越能創造一個成功的視覺簡報(visual presentation)。

Let’s imagine presenting data on “Environmental Conservation in the New Millennium” in the following scenarios: (1) on a small-scale blog visited mostly by lay environmentalists; (2) in a classroom of high school students; and (3) at a fundraising event for an environmental conservation organization. Would you create and explain your data the same way to each audience? Hopefully not. You should be able to make a few assumptions about what’s most relevant to present even if you’ve never met a single audience member.

讓我們想像一個關於 “新世紀環境保護”的簡報資料包括下列情境: (1) 一群主要是由業餘(lay)環保主義者(所組成)的人來訪問(您的)部落格時; (2) 高中職的課堂上; (3) 某個環保組織的募款活動上。您會以相同的方式針對每種不同的觀眾解釋您的資料嗎?希望(您)不是如此。你即使不認識任何一個觀眾(也)會做出一些相關的假設。

In our first scenario, we can assume visitors are already interested in conservation. They may have spent time doing actual research. A portion are return visitors who may rely on your specific perspective; they might see you as a content area expert. Your site is, most likely, not the only blog on which they rely, but one day it could be their favorite! At minimum, we can assume they’ve stumbled upon your blog intentionally, and not because of issues with autocomplete. In this instance, breadth and depth are key. You can take more time to explore, deconstruct and restructure the data. If the intention of your site is to incite further exploration, you can presents visuals that pose questions or make viewers question their own beliefs.

在我們的第一個場景中,我們可以假設瀏覽者對環保感興趣。 他們可以花了一些時間做了實際研究。一部分的人可能憑藉著(認同)你的特殊觀點而重新瀏覽(這個blog);他們也許會將您視為這個領域的專家。你的網站(指部落格)最後可能是,但並非唯一他們所認同的部落格。但是在某天(你的部落格)會成為他們的最愛! 至少,我們可以假設他們(花了一些心思)無意中發現您的部落客,但並非自動提醒的情況。在這種情況下,廣度(breadth)和深度(depth)是關鍵(因素)。你可以花更多時間去探討,解構和重構這些資料。如果你的網站目的是策劃更進一步的探討,你可以以視覺化呈現(一些資料)以發表(您的)見解或讓其質疑自己所相信的。

Our high school student scenario is a bit different. You can assume that your audience possesses very little familiarity with the topic. (Though, as always, some members will know more than others.) Attendance may be mandatory, not voluntary: keeping their interest will be key. You’ll want to present fascinating, high-level, attention-grabbing visuals, that address immediate and pressing issues. Approach your vocabulary carefully: explain less-common terminology, and include more visual indicators of good/bad, positive/negative. Your visual display is intended to clearly present the importance of conservation, leaving little room for doubt.

我們的高中同學場景會有一些不同。你可以假設你的觀眾微乎其微熟悉這個話題(雖然,一如既往,有一些同學會了解的比別人多。) 也許(他們都是)被迫出席的,並非志願(出席):(那麼)保持他們的興趣將會是關鍵點。您想呈現迷人的,高層次,吸睛的視覺,這是個是刻不容緩的議題。, (請)注意您的用字遣詞:解釋罕見的術語, 並且包括多個好/壞,正面/負面 的視覺指標。您的視覺化(資訊)是為了清楚地呈現環保的重要性,沒有質疑的餘地

At last, we have our fundraiser attendees. This audience needs to feel that environmental conservation is a cause worthy of their monetary support. It will likely be a mixed crowd: interested donors, their disinterested partners (who just came for free food and drinks), field experts, employees, and interns. You can assume they’ll expect a balance of sentiment, the need for urgency, and solid fact. We’ve assumed the crowd is mixed, so you’ll want to use language that is both familiar and easily understood while not appearing condescending. This audience expects to have their interest in the importance of conservation confirmed and your visuals should accommodate this. As with your student group, leave no obvious question unanswered.

最後,我們還有募款活動的與會者。這些觀眾需要一個讓他們覺得環保是值得他們資助的原因。這可能是一個混合的群體:感興趣的投資者,(與其同行)無利益關係的合作夥伴(僅為了免費的食物和飲料與會),(某)領域專家,員工和實習生。您可以假設他們將期待(達到某種)情感平衡,(是)需要急迫且確實的事實。我們假設這個族群是混合型的,所以你需要使用熟悉且淺顯易懂的語言,同時勿表現居高臨下(的感覺)(譯者案:指自命清高地用生澀的語言),這些觀眾希望他們所認為的環保重要性(課題)是被認同(or確認)的,您的視覺效果應該要確保這一點。作為您的學生族群,是無庸置疑的。

Presenting emotion-driven content doesn’t mean leaving out key facts if they don’t fit into your ideal storyline. Be extra careful when sharing cause-driven content, and do your best to ensure that your values don’t interfere with an accurate presentation of the data!

呈現感情導向(emotion-driven)的內容並非表示脫離了關鍵的事實,(即使)(這些設定)不適合你所屬意的故事情節。當您分享成因導向(cause-driven)的內容時,請務必小心,請確保當您使用精準的資料呈現時,您的價值觀(values)不受干擾。

Now that we’ve discussed the importance of determining a key message and understanding its audience, let’s delve into deciding which data to illustrate.

現在我們已經討論了決定關鍵訊息以及了解相對應的觀眾的重要性,讓我們更深入探討哪些資料該被呈現。

Deciding Which Data to Illustrate

決定哪些資料(是需要被)製圖

You can begin the process by expanding your key message into a narrative or story. Our goal is to present a sequence or set of facts which gradually leads your audience to the key message. The data you choose to illustrate should set the context, establish the main points of interest, and explain how these are interconnected. Be intentional in what you present, but do not censor data to further your argument. Your visual story should be based on what the data—and not only what you want to—say.

您可以開始進行這階段的工作,(藉由)將您的關鍵訊息擴展成敘述或故事(開始)。我們的目標是呈現一系列或遺族事實以逐漸引導你的觀眾(了解您的)關鍵訊息。你所選擇的呈現資料應(具)上下文關聯,建立(讓人感興趣的)主要的(觀)點,並解釋他們是如何互相關聯的。有意圖地呈現你(所想呈現)的(論點),但切勿過度解釋(你的論點)。你(所呈現的)視覺故事應該是基於資料而不僅僅是您想 甚麼而已

Take, for example, the following table presenting the I.Q. scores of children who were adopted at a young age and the socioeconomic status (based on income and occupation) of both their adoptive and birth parents. These data are taken from C. Capron and M. Duyme’s 1989 study, “Children’s IQs and SES of Biological and Adoptive Parents in a Balanced Cross-Fostering Study,” published in the European Bulletin of Cognitive Psychology.

舉例來說,下表描述了兒童時期被收養的孩童智商(I.Q)與養父母、生父母社會經濟地位(socioeconomic status)(譯者案:簡稱社經地位,英文縮寫SES)(指基於收入和職業的評量)(的關係)。這些參考自 C. Capron 和 M. Duyme’s 在1989的研究, “Children’s IQs and SES of Biological and Adoptive Parents in a Balanced Cross-Fostering Study,” 發表於 European Bulletin of Cognitive Psychology

I.Q.

養父母社會經濟地位(SES)

生父母社會經濟地位(SES)

136

99

121

133

125

131

103

115

116

117

94

103

99

125

111

93

101

94

125

91

98

99

91

124

100

116

113

119

92

91

98

83

99

68

76

115

86

116

Let’s discuss two possible narratives that you could create from this dataset: “Children’s Intelligence Limited by Adoptive Parents’ SES,” and “Adopted Children’s Intelligence Influenced by Both Biological And Adoptive Parents’ SES”.

我們來討論(藉由)這些資料集可以得到的兩個可能的敘述:“兒童的智力受限於養父母的社會經濟地位(SES)”和“受領養兒童的智力受生父母與養父母的社會經濟地位(SES)影響”。

Children’s Intelligence Limited by Adoptive Parents’ SES

兒童的智力受限於養父母的社會經濟地位(SES)

We can create a story supporting the first message by solely looking at the adoptive parents’ socioeconomic status: children of those adoptive families with a high SES had a mean I.Q. of nearly 112 whereas those adopted by a low SES family had a mean I.Q. of 99. But, this narrative would only include half of the relevant information: it leaves out entirely the SES of the child’s biological parents. Understandably, this could play just as big a role as the family’s socioeconomic status would likely impact the level and quality of prenatal care, and, in turn, the in utero development of the child.

我們可以創立一個故事以支持第一個論點--僅藉由觀看養父母的社會經濟地位(SES):這些被收養的孩子(因)養父母擁有較高的社經地位(SES)而平均智商接近於112,而養父母(若)較低的社經地位(SES)則(孩子的)平均智商為99。但是,這樣的陳訴只包括了一半相關的訊息(量):(這)完全忽略了孩子的生父母。可以理解的是,家庭的社會經濟地位對於產前胎教的水準與質量扮演(者)很大的角色,(進一步低說,)(影響到)在母體中的胎兒。

Adopted Children’s Intelligence Influenced by Both Biological And Adoptive Parents’ SES

受領養兒童的智力受生父母與養父母的社會經濟地位(SES)影響

So, more correctly, a child’s I.Q. is a function of both his or her biological and adoptive parents’ socioeconomic status. If both have a high SES, the child’s I.Q. will tend to be the highest. If one has a high SES and the other a low SES (it doesn’t matter which set of parents has which), the child will typically have an average I.Q. And finally, if both have a low SES, the child will tend to have a below-average I.Q.

所以,更確切地說,孩子的智商(IQ)是他們的生父母與養父母社經地位(SES)的函數。如果兩者(都具)較高的社經地位,則孩子的智商將高於平均值。如果其中一方較高而另外一方較低(無論那邊的父母(都一樣)),孩子通常是一個平均智商。最後,如果兩者都是較低的社經地位(SES),則孩子(的智商)將將低於平均值。

Our first example is a clear illustration of what happens when you create a story based on what you want to say, and not what the data say. Unfortunately, applications of data such as this are neither uncommon nor farfetched. We see this done on the news and during casual conversation. The omission of key facts and related variables creates a visual that is full of misinformation. It lacks credibility and presents obvious biases. The second instance presents far less outright bias, is a plausible story based on the data available, presents context, introduces all variables, and explains how the variables are connected. Although it will usually result in a less-sensationalized title, a full presentation of all relevant data is the only way to maintain a credible and airtight argument.

我們的第一例子非常清楚地呈現發生了甚麼事情--當你創立了一個故事(是)基於你想說的(論點),而不是數據想說的(論點)。 很遺憾的是,資料的運用(就像這樣)即不罕見也不(如此)牽強附會。我們可以在新聞和閒談中(看到這些現象)。關鍵事實與相關變化因素(variables)的遺漏創建了一個充滿錯誤訊息的視覺(資料)。它缺乏公性力(credibility)並存在明顯的偏見(biases)。第二個例子則遠遠少了明顯的偏見,是一個基於可用數據的合理故事,(所)呈現的內容介紹了所有的變化因素(variables),並解釋這些變化因素(variables)是如何被連結的。雖然它的結果往往是一個不太煽情的標題,(將)所有相關資料(做成)一個完整的報告是唯一一種方法(可以)維持可信度(credible)和嚴謹論點(airtight argument)

Deciding How Much Data to Illustrate

決定哪些資料(被)呈現

In previous sections we’ve gone over how to determine a key message, the importance of identifying the audience, and a process for isolating facts to illustrate. We can work on determining how much of our data we need to visualize.

在前面的章節中,中我們已經討論了如何確定關鍵訊息,重點在於分辨觀眾(的不同)以及(將完全)不同的事實轉換為圖示說明。我們可以(進行)決定多少資料需要轉換成視覺呈現的工作(了)。

If illustrating data is supposed to make information more digestible, then care should be taken not to present more than the audience expects, or more than they need to be able to understand your message. As you decide how much data to illustrate, keep in mind the idea that more is not always synonymous with better unless it’s meaningful and presented in support of your key message. In most instances, your visuals will be taken as part of a narrative, contents in a storehouse, or maybe a combination of both.

如果資料圖示應該讓資訊更易理解,那麼(就要)注意應採取不要過度解釋(而)超過觀眾的期望,或(你的訊息)超出他們所能理解(的範圍)。當你決定要(使用)多少的圖示(來呈現),(請)記住這個觀念: 並非總是 的同義詞,除非它是有意義的,並且支撐著你的關鍵訊息。在大多數的案例中,你(所想呈現)的視覺效果應採取一段敘述、智庫內容(所取代),或者結合兩者。

As previously discussed, a narrative is a simply a story presenting a sequence of facts which gradually lead your audience to the key message. When you think of the narrative, think of written reports, PowerPoint presentations, and individual articles in newspapers and magazines or online. You want to illustrate just enough data for your audience to easily identify and understand your perspective without becoming exhausted. Each illustration should have a specific purpose. Avoid including visuals simply because they impress. As a test, try removing one or more illustrations or rearranging the presentation order. Does your narrative still make sense? Each illustration should stand alone, without too much verbal or written explanation, but if it doesn’t add to the audience’s understanding, it’s probably not needed.

如前文(所)討論,敘事(narrative)是一種(將)序列化的事實呈現為簡單的故事並藉此逐漸引領你的聽眾(了解)你的關鍵訊息。當你考慮(想用)敘事(narrative)(呈現時),(請)想想(or參考)書面報告,PPT簡報,報章雜誌或者網路上的個別文章。你為想(達到)不需要精疲力盡(就)可以讓你的觀眾輕易辨識和了解你的觀點(而)使用充裕的圖檔(illustrate)。每一個圖檔(illustration)應該有一個特定的目的(purpose)。避免包含簡易的視覺因為(為了使得)他們留下深刻映像。作個實驗,嘗試刪除一個或者數個圖檔(illustrations)或重新編排您報告的順序。你的敘事(narrative)仍然有意義嗎? 每個圖檔(illustration)應該各自獨立,不需要太多口頭或書面的解釋,但是如果它沒有增加你的觀眾了解(你的論點),(那麼)它可能是不需要的(圖檔)。

For the audience members wanting more, you can always provide links or references to additional takes on your data along with detailed commentary to contextualize and further explain the information. If you’d like to impress a technically savvy audience, a graphical appendix could be even be shared as a GitHub repository or a gallery of code gists hosted on bl.ocks.org.

針對(那些)意猶未盡的觀眾,你可以隨時提供額外的連結或參考資料在你的資料上,連同詳細解說整個脈絡與進一步解釋資訊。如果你想給一個精通技術的觀眾留下深刻印象,(可以參考)一些被分享在GitHub的圖像附錄或參考位於bl.ocks.org上的代碼範例。

A storehouse, on the other hand, can be thought of as an information repository. Usually consisting of multiple narratives and stand-alone content, this is an example of when more can be better. Unlike those of a narrative, storehouse visitors are less susceptible to data fatigue. They respond well to large quantities of data because they expect to spend time building or enhancing their understanding of a topic. The storehouse doesn’t need to focus on presenting a single message. Its audience seeks new treatments of data, a diversity of perspectives, and various dissections of a topic or content area. In the storehouse scenario, the main criterion for deciding how much data to illustrate should be whether something will create redundancy. If your illustration fails to add something new to the mix or to expand on a topic, it can likely be omitted.

智庫(storehouse),在另一個方面,可以被當作是一個訊息的倉庫,通常由多個敘事(narratives)和獨立的內容(所組成),這是一個越多越好的例子。不像那些單一的敘事(narrative),訪問智庫(storehouse)不太容易感受到資料疲乏(data fatigue)。他們對於大量量體(quantities)的資料有較好的反應(respond)因為他們預計花費時間去建置或增強他們理解這個議題(topic)。智庫(storehouse)不需要在單一的訊息上聚焦。它的觀眾在尋覓資料的新解法(treatments),觀點的多樣性(diversity),以及一個主題或區域內容的各種剖析(dissections)。在智庫(storehouse)的場景中,決定需要圖像化(illustrate)的資料多寡的主要標準是創立這些東西是否(將造成)冗餘性(redundancy)。如果你的圖檔(illustration)無法新增一些新的(效果)以混合或解釋所談的話題,(那麼)它可能需要被忽略。

To exemplify, let’s imagine a cinephile and store manager. Both are browsing a blog filled with upcoming movie release dates, reviews, and critiques of directors. The cinephile spends hours on the site, soaking up each and every visual and reading through its content. The manager simply wants to know what popular movies he should order for the next holiday season. The manager probably wouldn’t want to spend hours trying to find his answer. For our cinephile, more is better; for the manager, less is more.

舉例說明,讓我們想像一個電影愛好者(cinephile)和一個店經理(store manager)。同時瀏覽一個充滿即將上映電影的上映日期,影評以及導演評論的部落格(blog)。電影愛好者(cinephile)花了數小時在這個網站上。浸淫每一個視覺內容並讀取其內容。(而)店經理只想知道那部是熱門電影好讓他在下個假期(前)訂購。店長大概不會想花數小時(來)試圖找到他想要的答案。(所以)對電影愛好者(cinephile)(而言),(資料)越多越好,對店長(manager)(而言),少即是多。

Editing and Revising

編輯與矯正

Here’s a frequent and frustrating occurrence: you did your brainstorming, made a bunch of visualizations, and edited down to the best subset to include in your project. You were careful not to overwhelm your audience and you made sure that your illustrations covered the most important key points without being redundant.

How maddening, then, to field questions in a presentation, or see comments in a story or blog post, calling for the very visualizations that you left on the cutting room floor! You second-guess your calls, resist the urge to argue with the person asking the question, grit your teeth and grumble.

It’s okay. If you do a good job and engage your audience, they will naturally be curious and want more information. They might want to see the same data presented in a different way, to dig down, or to zoom out. If these questions mirror the decisions you were making in your selection process, that’s good news! It means you are on the same wavelength as your audience, and that they are involved and interested in the story your data tell.

There are several ways to keep (but de-emphasize) the visualizations that did not make the cut in your main collection. For slideshows, it is common practice to have a collection of extra slides after the “thank you” or conclusion slide that contain information that might be interesting but that won’t fit within the time limit. “Yes, I do have that broken down by [industry sector/year/country/gender],” you say confidently as you flip to the prepared slide. Voila!

Another way to do this would be to publish interactive versions of your visualizations that allow the viewers to dive in and explore the information themselves. If you’re able to share the raw datasets, that’s even better! That way, those who wish to dig deeper and understand the data in new ways will have the option to do so. We’ll talk more about static and interactive graphics later in the Print vs. Web chapter.

If you’re looking for early feedback and you’re not exactly sure where to turn, you can check out HelpMeViz, a community site where you can post your works-in-progress and receive friendly suggestions on how to improve. Getting feedback from your audience and revising your visuals to better fit their needs is all a part of the process!

如果您正在搜尋較早的回饋且您無從下手時,您可以查閱HelpMeViz, 這個社群網站可以發表您的工作進度,並且可以獲得友善的改進建議。 收集您的觀眾的回饋並依據他們的需求改進,這將會是這個過程的所有環節。 </div>

Last updated