CSV is a text file format in which the values of a table are separated by commas, with one record written per line.
What the table looks like
The same line in CSV
Store | Visitors
Store,Visitors
Main store | 120
Main store,120
Station store | 85
Station store,85
Fields across a row are separated by commas, and once a record is complete, a line break starts the next one. Because the content is plain text, it does not depend on any particular program and is widely used to pass data between spreadsheet software and databases. Cell colors and formulas, however, are not kept; only the values remain.
AnalogyIt is like a roster written one person per line with a mark between each item, so any program can read it by following the same rule.
How to rememberIt stands for Comma-Separated Values, so the name itself says values separated by commas.
Exam points
ExamWhen a table is output as CSV, the cell values in the same row are joined by commas, and a record (レコード) separator goes at the end of each row. The data are written out row by row as the table is laid out, not read down each column and rearranged.
Mix-upThe delimiter is a comma, and data separated by a slash or some other symbol is not called CSV.
A shapefile (シェープファイル) is a geographic data file format that stores the shapes of map features (points, lines, and polygons) together with their attributes.
Shapefile (one set of data)
.shp: shapes as points, lines, polygons
.shx: index for locating shapes
.dbf: attributes such as name and area
One dataset is made up of several files with different extensions that work as a set. For instance, a bus stop is recorded as a point, a road as a line, and a town block as a polygon, with attributes such as name and area attached in table form. Most geographic information system software can read and write the format, so it is widely used to exchange map data.
AnalogyIt is like putting a stack of shapes drawn on a map and a stack of note cards describing each shape into one envelope.
How to rememberAs the word shape suggests, it is a file that records shapes.
Exam points
Mix-upWhile CSV is a table format that only lists values, a shapefile (シェープファイル) can record the shapes themselves as points, lines, and polygons.
GIS data (GISデータ) is the general term for data tied to locations on a map and handled in a geographic information system (GIS).
Location and shape data
Link attributes
Overlay as layers on a map
Analyze distribution and distance
Read imbalances and distances from the overlaid result and use them in decisions.
4 / 4
GIS is a system that overlays location-based data on a map and analyzes distribution and distance. For instance, layering surrounding population and road data over store locations helps when deciding where to open a new store. A shapefile is one of the formats used to store such data.
AnalogyPicture laying several transparent sheets showing population and roads over a blank map and comparing them, except a computer does the work.
How to rememberGIS stands for Geographic Information System, and the word Geographic is the clue.
Shapefile vs. GIS Data
Aspect
Shapefile
GIS Data
Key differenceWhat it refers to
One file format for storing geographic data
All data tied to locations
Relationship
Serves as a container for GIS data
Can also be held in other forms, such as images, besides shapefiles
Survey data (調査データ) is data gathered by examining the current situation through questionnaires or observation, without changing anything about the subject.
You can ask people questions and collect their answers, or count and record what you see on site. Since no conditions are changed, it suits getting a picture of actual conditions and how opinions are spread. Even when two things move together, however, it is hard to confirm whether one causes the other.
AnalogyIt is like a notebook you carry around town, counting shops and asking passersby what they like, writing things down just as they are.
How to rememberThe English word survey originally meant to look over from above, so think of it as data that looks over things to capture how they are now.
Experimental data (実験データ) is data obtained by deliberately changing conditions and measuring the results.
Decide the condition to change
Keep other conditions the same
Change it and measure
Compare the results
Compare the results and judge whether the amount of fertilizer affected growth.
4 / 4
Because everything except the changed condition is kept the same, it is easier to confirm whether that condition caused the result. In business, preparing two designs for a Web page, splitting users between them, and comparing their responses is one such method.
AnalogyIt is the same idea as cooking two plates that differ only in the amount of salt and tasting them side by side.
How to rememberThe English word experiment comes from the idea of trying something out, so it is data you get by trying.
Exam points
Mix-upExperimental data (実験データ) comes from controlling conditions, so cause and effect are easier to confirm, while survey data (調査データ) leaves conditions alone and suits capturing the current situation.
Survey Data vs. Experimental Data
Aspect
Survey Data
Experimental Data
Key differenceHandling of conditions
Left unchanged and examined as they are
Deliberately changed and measured
Best suited for
Grasping actual conditions and how opinions are spread
Behavior log data (行動ログデータ) is data in which a system automatically records the actions and operations people perform on Web sites, apps, and so on.
Time
User
Action
9:02
Member 017
Opens the top page
9:03
Member 017
Moves to the umbrella product page
9:05
Member 017
Adds it to the cart
Records build up of which pages were viewed and in what order, what was clicked, and what was bought. Actual behavior is kept as it happened without asking the person, so it helps reveal trends in how users act and is used to improve services and recommend products.
AnalogyLike footprints left in snow, the path someone took can be traced afterward even if they never thought about it.
How to rememberThe word log comes from a ship's logbook, so think of this as a logbook of behavior.
Exam points
ExamThe technique of interpreting user behavior from records of how often a site's pages were viewed and which paths users took between pages within the site is access log analysis (アクセスログ分析).
Operation log data (稼働ログデータ) is data that automatically records the operating status of equipment such as machines, facilities, and servers.
For instance, it records how long a machine ran, when it stopped, when errors occurred, and changes in temperature or rotation speed. Looking through the accumulated records lets you notice early signs of failure and decide when to carry out inspections.
AnalogyLike a car's odometer or driving recorder, it is a diary in which a machine writes down how it has been working.
How to rememberThe Japanese word 稼働 means a machine running and doing its job, so read it as a machine's work diary.
Exam points
Mix-upIf what is recorded is human actions, it is behavior log data (行動ログデータ); if it is equipment operation, it is operation log data (稼働ログデータ).
Behavior Log Data vs. Operation Log Data
Aspect
Behavior Log Data
Operation Log Data
Key differenceWhat is recorded
Human actions and operations
Operation of machines and equipment
Main uses
Analyzing usage trends, recommending products
Detecting early signs of failure, timing inspections
Chartjunk (チャートジャンク) refers to decoration added to a chart that does nothing for understanding the data and instead gets in the way of reading it.
Cluttered chart
Clean chart
3D effects
Tilted in 3D, so the front slices look larger
Kept flat, so sizes can be compared directly
Background
Contains photos or heavy patterns
Plain and unobtrusive
Gridlines
Many thick, dark lines
Only as many as needed, thin and light
The term was coined by the statistician Edward Tufte, and 3D effects, flashy backgrounds, and meaningless pictures are typical examples. In a 3D pie chart the front slices look bigger, which easily leads to misreading the proportions. When making a chart, cut back on decoration so that the parts showing the data stand out.
AnalogyIt is like a plate piled so high with garnish that you cannot tell where the actual food is.
How to rememberJunk means rubbish in English, so read it as rubbish on a chart.
Key points
CSV is a plain-text file in which values are separated by commas and each line holds one record, and it is used to pass data between programs.
GIS data is data tied to locations, and overlaying it on a map lets you analyze distribution and distance.
Data can also be classified by source, and where and how it arises changes what it can tell you and which analysis suits it.
Chartjunk is decoration that gets in the way of reading, and a chart should be built so that the parts showing the data stand out.