All Glossary Terms

Question 1

The data wrangling step in which errors in the raw data are corrected.

Answer 1

Cleaning

Question 2

Documentation of characteristics of the wrangled data such as names and definitions of the fields, units of measure used in the fields, the source(s) of the raw data, relationship(s) of the wrangled data with other data, and other attributes.

Answer 2

Data dictionary

Question 3

The process of cleaning, transforming, and managing data so it is more reliable and can be more easily accessed and used for analysis.

Answer 3

Data wrangling

Question 4

A tag or marker that separates structured data into various fields

Answer 4

Delimiters

Question 5

The data wrangling step in which the analyst becomes familiar with the data in order to conceptualize how it might be used and potentially discovers issues that will need to be addressed later in the data wrangling process.

Answer 5

Discovery

Question 6

A field that that takes a value of 0 or 1 to indicate the absence or presence of some categorical effect.

Answer 6

Dummy variable

Question 7

The data wrangling step in which the raw data are augmented by incorporating values from other data sets and/or applying transformations to portions of the existing data to ensure that all data that will be required for the ensuing analyses will be included in the resulting data set.

Answer 7

Enriching

Question 8

A characteristic of the observations in a data set.

Answer 8

Field

Question 8

A data file in which structured data are arrayed as a rectangle, with each row representing an observation or record, and each column representing a unique variable or field.

Answer 8

Flat file

Question 8

Instances for which there is an appropriate reason for the value of a field to be missing.

Answer 8

Illegitimately missing data

Question 8

Instances for which there is an appropriate reason for the value of a field to be missing.

Answer 8

Legitimately missing data

Question 9

Systematic replacement of missing values with values that seem reasonable.

Answer 9

Imputation

Question 10

Instances for which the tendency for a record to be missing a value of some field is related to the value of some other fields(s) in the record.

Answer 10

Missing at random

Question 11

Instances for which the tendency for a record to be missing a value of some field is entirely random.

Answer 11

Missing completely at random

Question 12

Instances for which the tendency for a record to be missing a value of some field is related to the missing value.

Answer 12

Missing not at random

Question 13

are data that are stored in a manner that allows mathematical operations to be performed on them. Data of this type generally represent a count or measurement.

Answer 13

Numeric data

Question 14

Combining multiple data sets that each have different data for individual records, when each record occurs no more than once in each data set.

Answer 14

One-to-one merger

Question 15

Combining multiple data sets that each have different data for individual records, when at least one record occurs more than once in at least one of the data sets

Answer 15

One-to-many merger

Question 16

The data wrangling step in which a file containing the wrangled data and documentation of the file’s contents are made available to its intended users in a format they can use.

Answer 16

Publishing

Question 17

Data that has not been processed or prepared for analysis.

Answer 17

Raw data

Question 18

A grouping of characteristics for a particular observation in a data set.

Answer 18

Record

Question 19

Data that does not have the same level of organization as structured data, but that allow for isolation of some elements of the raw data when they are imported.

Answer 19

Semi-structured data

Question 20

Data organized so that the values for each variable are stored in a single field.

Answer 20

Stacked data

Question 21

Data sets that are arrayed in a predetermined pattern that make them easy to manage and search.

Answer 21

Structured data

Question 22

The data wrangling step in which the raw data file is arranged so it can be more readily analyzed in the intended manner.

Answer 22

Structuring

Question 23

The extraction of fields and records that will be useful for ensuing analyses from data.

Answer 23

Subsetting

Question 24

Data that are words, phrases, sentences, and paragraphs.

Answer 24

Text data

Question 25

Data organized so that the values for one variable correspond to separate fields, each of which contains observations corresponding to these respective fields.

Answer 25

Unstacked data

Question 26

Databases that are not arranged in a predetermined pattern, and therefore in their raw form cannot be stored in a manner similar to a flat file.

Answer 26

Unstructured data

Question 27

The data wrangling step in which the accuracy and reliability of the wrangled data are confirmed so they are ready for the ensuing analyses.

Answer 27

Validating