Introduction

This is a brief introduction into some of the core features of VAST. In doing this, I will simplify a few features and ignore others completely.

Types of elements (selection)

Arrows stand for IF-THEN relationships

If X, THEN Y

Six types of relationships:

• naming (n)
• conceptual implication (i)
• causation (c)
• prediction (p)
• transformation (t)
• reasoning (r)

Naming concepts

• The I in the box symbolizes a concept
  • A concept is an idea about a condition that may or may not apply to certain objects (e.g., people)
  • The labels of concepts (i.e., I) are abstract and basically arbitrary
    
• The n stands for a naming relationship
• A name is a word (or sentence, or paragraph) that is used to describe those objects to which a concept applies. Names (or “labels”) are set in quotation marks "" in VAST displays.

Naming concepts

Synonyms

A concept may have any number of names. Here, all three names are synonyms regarding concept I.

Naming concepts

Homonyms

This expresses the idea that the same name is used for conditions that may not be the same. This name might thus be a homonym.

Naming concepts

Visualizing a jingle-jangle situation

Data

• Data (left box, folder-shaped) is a special kind of concept, which explicitly refers to something observable.
  • Same as a manifest variable in Structural Equation Modeling
• This display is supposed to capture the idea that “Intelligence is what the intelligence test measures”

Data

Symbol

Note: In the original VAST paper, the Data concept has a different shape (but that shape is not available in the Graphviz system used to create the graphs in this presentation):

Data

• The r denotes reasoning relationships
• This display is supposed to capture the idea that more than one test may be used as evidence for people‘s intelligence

Data

• The p stands for a prediction
• The c stands for causal effects

Data

This display is supposed to capture the idea that …

• The same thing has a causal effect on the outcomes of both tests
• Therefore, the outcomes of the two tests correlate with one another

Splitting up the display elements 1

Readability may often be promoted by setting naming relationships aside.

Splitting up the display elements 2

Strength of relationship

• Default interpretation of a directional arrow X → Y: this relationship is considered relevant and positive (i.e., the more X the more Y)
• Relationship strength can be expressed with a number between -1 (perfect negative relationship), 0 (no relationship), and 1 (perfect positive relationship)
  • If no number is explicitly specified: default relationship strength is “> 0”
• The strength of a relationship can also be expressed in a vague verbal way, e.g., “strong”, “weak”, “negative”, …
• The relationship strength can be applied to any relationship type:
  • type p: “wearing glasses” makes it 70 percent likely for a person to also be “smart”
  • type r: It is 90 percent reasonable to assume someone “is in love with you” when that person “giggles a lot while talking to you”
  • type c: being “obese” makes it 50 percent likely for someone to develop “Diabetes Type II” as a consequence

Fuzziness as a feature

• The goal are precise theories - but in reality, they often are (at least partially) vague and fuzzy
• VAST allows to make this fuzziness explicit by allowing vague quantifiers of relationship strength, e.g. “weakly positive”, “strong”, “>0.5”
  • Better than assigning an arbitrary, pseudo-precise number
  • It’s about formalizing the current state/precision of a verbal theory - if it is imprecise, let’s show it!
• Show gaps, underspecifications, and inconsistencies in the theory

Notes on the different types of relationships

• Relationships between constructs can be …
  • on a theoretical level. Then it’s typically a causal path (c)
  • on empirical level: A relationship has been found! This is a prediction path (p)
  • Research cycle: One author’s exploratory p-path is the next author’s theoretical c-path
• Consider what you want to formalize:
  • A certain author’s theoretical point of view at a certain point in time?
  • Or your personal, current point of view?
  • The empirical support for a statement?

Exercise: How would you interpret this?

Exercise: How would you interpret this?

Contact

• @nicebread@scicomm.xyz
• ed.uml.ysp@tdorbneohcs.xilef
• https://www.nicebread.de
• https://github.com/nicebread