some chapter 3 narrative #31
Conversation
OTTR Check ResultsSummary
🎉 All checks passed!Last Updated: 2026-01-05-17:17:55 |
|
Re-rendered previews from the latest commit:
* note not all html features will be properly displayed in the "quick preview" but it will give you a rough idea. Updated at 2026-01-05 with changes from the latest commit 60c6393 |
|
Note that I've added several issues about places I should add references and figures. (Issues #18 - #30 ) https://github.com/fhdsl/DataViz_Considerations/issues?q=is%3Aissue%20state%3Aopen%20label%3Achapter3 |
|
Note to self: need to add a note about "preattentive attributes" and which encodings are easier for audiences. Also add example of uno cards that have color and shape now! (or probably put this in chapter 4) |
carriewright11
left a comment
carriewright11
left a comment
There was a problem hiding this comment.
Looks good! I wonder if we should split this into two chapters - one about the elements of plots/encodings and one about types of data and deciding which plot to use- what do you think?
Adding visuals and examples for the encodings would be useful
| ```{r, include = FALSE} | ||
| ottrpal::set_knitr_image_path() | ||
| ``` | ||
| A data visualization represents data in a simplified way. The visualization is made up of the represented data as well as the means by which the data is represented: textual and graphical design elements. Each conventional plot type utilizes a combination of design elements in a predictable way or layout. The design elements are the building blocks for the data visualizations. Deciding which design elements and plot type to use relies on several considerations: The type(s) and amount of data that you're visualizing as well as the question you are trying to answer or message you want to communicate. |
There was a problem hiding this comment.
| A data visualization represents data in a simplified way. The visualization is made up of the represented data as well as the means by which the data is represented: textual and graphical design elements. Each conventional plot type utilizes a combination of design elements in a predictable way or layout. The design elements are the building blocks for the data visualizations. Deciding which design elements and plot type to use relies on several considerations: The type(s) and amount of data that you're visualizing as well as the question you are trying to answer or message you want to communicate. | |
| A data visualization represents data in a simplified way. The visualization is made up of the represented data as well as the means by which the data is represented: textual and graphical design elements. Each conventional plot type utilizes a combination of design elements in a predictable way or layout. The design elements are the building blocks for the data visualizations. | |
| Deciding which design elements and plot type to use depends on: | |
| - The type(s) and amount of data that you're visualizing | |
| - The question you are trying to answer or message you want to communicate. |
| ## Data encodings | ||
|
|
||
| ### Shape | ||
| Data encoding is the process of representing data using visual elements or graphical attributes. These include geometric marks and visual channels which are the building blocks of data visualization. Different graph types utilize different encodings to represent data: geometric marks are typically the primary means of representing data across graph types while the visual channels are usually used as secondary or supplementary methods (e.g., shape, color, and textures). However, color can be a primary means of data encoding for certain graphs like heatmaps or choropleths <ADD REF>. |
There was a problem hiding this comment.
| Data encoding is the process of representing data using visual elements or graphical attributes. These include geometric marks and visual channels which are the building blocks of data visualization. Different graph types utilize different encodings to represent data: geometric marks are typically the primary means of representing data across graph types while the visual channels are usually used as secondary or supplementary methods (e.g., shape, color, and textures). However, color can be a primary means of data encoding for certain graphs like heatmaps or choropleths <ADD REF>. | |
| Data encoding is the process of representing data using visual elements or graphical attributes. These include geometric marks and visual channels which are the building blocks of data visualization. Different graph types utilize different encodings to represent data. Geometric marks are typically the primary means of representing data across graph types while the visual channels are usually used as secondary or supplementary methods (e.g., shape, color, and textures). However, color can be a primary means of data encoding for certain graphs like heatmaps or choropleths <ADD REF>. |
|
|
||
| #### Area {-} | ||
|
|
||
| **Area** is often proportional to the amount of data its representing rather than the raw data values itself. Larger areas correspond to larger data values. Area is commonly used in density plots |
There was a problem hiding this comment.
| **Area** is often proportional to the amount of data its representing rather than the raw data values itself. Larger areas correspond to larger data values. Area is commonly used in density plots | |
| **Area** is often proportional to the amount of data its representing rather than the raw data values itself. Larger areas correspond to larger data values. Area is commonly used in density plots. |
|
|
||
| ### Visual channels {.tabset} | ||
|
|
||
| These visual building blocks are also referred to as "visual channels" and they control the way that the geometric building blocks will appear. You can change the shape, color, or size of a point, but it's still a point. Visual channels include |
There was a problem hiding this comment.
| These visual building blocks are also referred to as "visual channels" and they control the way that the geometric building blocks will appear. You can change the shape, color, or size of a point, but it's still a point. Visual channels include | |
| These visual building blocks are also referred to as "visual channels" and they control the way that the geometric building blocks will appear. You can change the shape, color, or size of a point, but it's still a point. Visual channels include: |
|
|
||
| ### Ranking and distribution plots {.tabset} | ||
|
|
||
| These plots are primarily |
There was a problem hiding this comment.
did this get finished?
There was a problem hiding this comment.
It did not - Thanks for flagging that! I think I was hoping to add a sentence here and another sentence for the other conventional plot subgroup on group composition.
I think I generally like that idea for splitting the material, but I wonder about keeping the types of data with the elements of plots and encodings and making the split in between the coordinate system and the types of questions/goals of visualizations. The types of data are a fairly often referred to thing for the elements of the plots since certain elements work better with categories than numerical data. Feels like the types of data are very related to how the data is encoded. Do you think so too? |
The changes in this PR focus on the narrative within Chapter 3. Chapter 5 (field specific plots) are outlined and a few of the specific ones I covered in the talk are added using that material.
Chapter 3 aims to answer the question "Which visualization should I use?" and makes the argument that the answer to that is dependent upon the type and amount of data and the visualization goal. Because of this, the chapter begins by discussing types of data. From there, it discusses ways to encode or display data and then briefly coordinate systems as well as broad goals of data visualization. It transitions to providing some standard graph types and for each providing the primary type of data, how much there is, how it's primarily encoded, and the goal of that visualization typically (all the info the chapter has covered). It focuses on ranking and distribution plots (violin plots, boxplot with jitter, ridgeline plots, raincloud plots) and then group composition plots (stacked bar charts, heatmaps, and sankey diagrams). For each of those plots a description, when to use, strengths, weaknesses, and alternatives are provided. Still need to add example plots as mentioned in Issue #27 . A brief call out to chapter 5 and field standard plots are included then galleries are mentioned and then a summary.
Within chapter 5 I've included info for the caterpillar plot and waterfall plot, though they could probably use a single sentence each at the beginning about how the field uses them/why. I've included spots for maybe the repeated measures plot, definitely the kaplan meier curve, definitely the volcano plot, maybe an enrichment plot, definitely an MA diagram, probably a PCA plot, maybe a circos plot, and probably a biomedical knowledge graph.