Units of Analysis The Basics Chuck Humphrey ACCOLEDS/DLI Training December, 2001
Units of Analysis The Basics Chuck Humphrey ACCOLEDS/DLI Training December, 2001.
Mar 27, 2015
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Units of Analysis
The Basics
Chuck HumphreyACCOLEDS/DLI Training
December, 2001
Outline
An illustration
Definitions
Elements of the unit of analysis
Complexity
Data structure
An Illustration
A group of students in an econometrics class were sent to the Data Library to find some data for an assignment.
An Illustration
A typical request was like this one.
“I want to look at crime rates and a person’s level of education.”
An Illustration
crime rates are usually associated with spatial units or a time series
a person’s education is an attribute of individuals
This request raises problems.
An Illustration
does the student want crime rates and the percentage of the population with certain education levels for specific cities? This would be data aggregated over geography.
What are we looking for?
An Illustration
does the student want the crime rate for one city over time, such as the number of homicides in Edmonton over the past 40 years. This would be data aggregated over time.
What are we looking for?
An Illustration
does the student want the education level of criminals? This would be a special subpopulation of individuals convicted of crimes and consist of a microdata file of criminals.
What are we looking for?
An Illustration
does the student want the education level of victims of crimes? This would be a special subpopulation of individuals who were victimized and consist of a microdata file of victims.
What are we looking for?
An Illustration
Looking at crime rates and level of education can differ depending upon the unit of analysis.
•individuals
•geographic areas
•changes over time
An Illustration
After walking the student through these steps, he chose to build a model predicting income on the basis of highest educational attainment and a few other variables from the Census individual-level public use microdata file.
He completely abandoned his interest in crime!
An Illustration
Unfortunately, the student’s initial request not only failed to specify a clear unit of analysis, it included a mix of different units, which suggests that the concept was not understood.
The Point of the Illustration
The unit of analysis is fundamental to the data reference interview. Early identification of the unit of analysis will help focus a search on statistics, aggregate data, or microdata.
The Point of the Illustration
Furthermore, the unit of analysis is fundamental to secondary data analysis. It may be that knowledge of the unit of analysis is even more crucial in secondary analysis than in primary analysis, where the unit is implicit in the sample design, if not otherwise explicit.
The Point of the Illustration
Finally, the unit of analysis is a fundamental characteristic of statistical data structures, which are the formal ways in which data are organized for processing.
Definitions
The unit of analysis is the basic entity or object
about which generalizations are to be made based on an analysis, and
for which data have been collected
Definitions
How does the unit of analysis relate to the unit of observation?
The unit of observation is the entity in primary research that is observed and about which information is systematically collected.
Definitions
The unit of observation and the unit of analysis are the same when the generalizations being made from a statistical analysis are attributed to the unit of observation.
Unit of Observation– in original data collections, the unit of
observation is determined by the method by which observations are selected
Unit of Analysis– the unit of analysis is determined by an
interest in exploring or explaining a specific phenomenon
Definitions
Identifying a Unit of Analysis
As hinted in the earlier illustration, the unit of analysis is shaped by three attributes:
– Social Phenomena
– Time
– Space
Research Outputs
Let’s begin by looking at a finished product to display these attributes.
We’ll use a table from the Health Indicators Database about suicide.
Social Characteristics
Geography and Time held constant
Ordered by Time
Geography and Age held constant
Geography Emphasized
Time and Age held constant
Social Phenomena
observations of a single social entity, such as a person or an institution
observations of multiple entities with a defined relationship, such as family, employer-employee
Social Phenomena
transactional observations that are the result of actions among entities, such as labour strikes or international conflicts, including wars
Time
observations made at one point in time; commonly referred to as a cross-sectional study
Time
observations made at multiple points in time the data may be organized by
time; commonly referred to as a time series
time may structure some form of repeated measures of content or subjects
Space
observations made within a specific spatial area
observations made within a hierarchy of spatial areas
Complexity
Complexity occurs when multiple types of entities are introduced within the same study.
Examples
parent child teacherperson activity timeperson car trips
Complexity
This complexity can arise within one of the attributes just discussed.
– a study of parents, children, and teachers, which are all social units
or between attributes– a study of people, their daily
activities, and the length of time of each activity
Complexity
Complexity is often represented in an hierarchy when the units can be grouped or nested within one another. For example, children may be grouped with their parents.
Complexity
Children grouped (nested) with Parents.
Parent 1 Parent 2
Child 1 Child 2 Child 3
Complexity
Parents and their children may be grouped into families and families grouped into households.
Household 1
Family A
Person i
Person ii
Household 2
Family A
Person i
Person ii
Complexity
Complexity may also be represented by combinations of entities among units. Those entities that are associated with one another are combined and those that aren’t associated, aren’t combined.
Complexity
These combinations are often described as having been crossed. For example, activities may be crossed with people.
Complexity
Activities crossed with people.Activity 1
Activity 2
Activity 4Activity 3
Activity 5 Activity 6
X
=
Person B
Person A
Person A Activity 3 Activity 6
Person B Activity 1 Activity 5
Complexity
Up to this point, complexity has been described conceptually. We’ve mentioned how multiple units of analysis and the ways in which they are related can create complexity.
Complexity
Complexity also manifests itself structurally through the ways in which data are organized to represent the nesting or crossing of multiple units of analysis.
Thinking about Units of Analysis
Conceptually– What is the content? This is what
we’ve been reviewing up to this point.Structurally
– How is it organized? This takes us to a discussion about data structure.
Let’s review basic data structure.The unit of analysis defines the underlying structure of a data file.
Statistical Data Structure
This structure consists of a series of rows with each row containing the data of one member of the unit of the unit of analysis.This simple structure is known as the flat, rectangular data matrix.
Statistical Data Structure
Case 1
Case 2
Case 3
*
Case n
*
*
Case n-1
Statistical Data Structure
All of the information collected for each member of the unit of analysis is organized in a fixed location in the file called fields or variables.
Statistical Data Structure
Case 1
Case 2
Case 3
*
Case n
*
*
Field 1*Field2
Field 3* Field k-1
Field k
Case n-1
Statistical Data Structure
Case 1
Case 2
Case 3
*
Case n
*
*
Field 1*Field2
Field 3* Field k-1
Field k
Case n-1
Statistical Data Structure
This structure looks like the grid of a spreadsheet. However, there is one very important difference between a statistical data structure and a spreadsheet.
Statistical Data Structure
The spread sheet is organized around individual cells, while the statistical data structure is organized around the rows.
Statistical Data Structure
Spreadsheet
Statistical Data Structure
Cell B2
Cell E3
Cell C5
Cell F7
Spreadsheet
Statistical Data Structure
Statistical Data Structure
Row 1
Row 3
Row k-1
Statistical Data Structure
The next slide presents the way that this simple statistical data structure appears in SPSS.
Statistical Data Structure
Row 1
Row 1
Row 8
Row 1
Row 8
Row 15
Row 1
Row 8
Row 15
Field 8
00001 169895714620691266912141307220251100002 212294362410300523012070302230352100003 61737841020370633712140603220251100004 151962542420280422806979797441062000005 169587521220230312352100302240312100006 173783282420380633864979797140755000007 88434954710300523032070302240352100008 76062182420300523006979797110157000009 581476302410260422636979797331062000010 1234850712204407344949797972212570
Person: GSS 10 Main
00001 169895714620691266912141307220251100002 212294362410300523012070302230352100003 61737841020370633712140603220251100004 151962542420280422806979797441062000005 169587521220230312352100302240312100006 173783282420380633864979797140755000007 88434954710300523032070302240352100008 76062182420300523006979797110157000009 581476302410260422636979797331062000010 1234850712204407344949797972212570
RECID
WG
HTFNL
PROV
DVSEXDVAG
ECAP
Adding Complexity to Data
Structurally– hierarchical : order & different
record layouts for different units of analysis
– relational : 1 to n relations– compound records : combination
of units represented on each record
Complex Data Structure
Household 1
Person 1
Person 2
Household 2
Household 3
Person 1
Person 2
Person 3
Hierarchical Data Structure
4600000000000 000000 000000004600100000000 000000 000000004600100105024RM 024000 5010 820900004600100205024RM 024000 5010 820900004600100305024RM 024000 5010 820900004600100405027T 024000 5010 820904104600100505027T 024000 5010 820904104600100605027T 024000 5010 820904104600100705031RM 031000 5011 821000004600100805031RM 031000 5011 82100000
Geography: 1991 Census N9101 Population 15 years and over by age groups (17) and marital status (6a), showing labour force activity (8) and sex (3)
4600000000000 000000 000000004600100000000 000000 000000004600100105024RM 024000 5010 820900004600100205024RM 024000 5010 820900004600100305024RM 024000 5010 820900004600100405027T 024000 5010 820904104600100505027T 024000 5010 820904104600100605027T 024000 5010 820904104600100705031RM 031000 5011 821000004600100805031RM 031000 5011 82100000
PROV
FED
EA CD CSDCSD T
ype
CCSCM
A/CA
4600000000000 000000 000000004600100000000 000000 000000004600100105024RM 024000 5010 820900004600100205024RM 024000 5010 820900004600100305024RM 024000 5010 820900004600100405027T 024000 5010 820904104600100505027T 024000 5010 820904104600100605027T 024000 5010 820904104600100705031RM 031000 5011 821000004600100805031RM 031000 5011 82100000
PROV
FED
EA CD CSDCSD T
ype
CCSCM
A/CA
Complex Data Structure
Relational Data Structure
R1
R2
R3
R4
R5
R1 C1
R1 C2
R1 C3
R1 C4
R3 C1
R3 C2
R4 C1
R5 C1
R5 C2One to Many
0000111169122244421472240699799799779979000021113011219442077219069979979977997900003111371229344214729306997997997799790000511123522094421072090699799799779979000061133862280441047280019999973601999900007111303120344207720306997997997799790000831330021854421079970723099799780459000083233002235331097997072639979972028900010113449219344209219301287997293509490001032344923202220879970736399799720439
Person: GSS 10 Union
0000111169122244421472240699799799779979000021113011219442077219069979979977997900003111371229344214729306997997997799790000511123522094421072090699799799779979000061133862280441047280019999973601999900007111303120344207720306997997997799790000831330021854421079970723099799780459000083233002235331097997072639979972028900010113449219344209219301287997293509490001032344923202220879970736399799720439
RECIDUNIO
NTYP
UNIONRNK
00001111691222444214000021113011219442070000311137122934421400005111235220944210000061133862280441040000711130312034420700008313300218544210000083233002235331090001011344921934420900010323449232022208
00001 1698957146206900002 2122943624103000003 617378410203700004 1519625424202800005 1695875212202300006 1737832824203800007 884349547103000008 760621824203000009 5814763024102600010 12348507122044
GSS 10 Main GSS 10 Union
00001 1698957146206900002 2122943624103000003 617378410203700004 1519625424202800005 1695875212202300006 1737832824203800007 884349547103000008 760621824203000009 5814763024102600010 12348507122044
00001111691222444214000021113011219442070000311137122934421400005111235220944210000061133862280441040000711130312034420700008313300218544210000083233002235331090001011344921934420900010323449232022208
GSS 10 Main GSS 10 Union
Complex Data Structure
Compound Data Structure
R1 x T1 x A1
R1 x T2 x A4
R1 x T3 x A7
R1 x T4 x A3
R1 x T4 x A1
R2 x T1 x A2
R2 x T2 x A9
000041144504000800024010000000012518733000041144308000900006011222220012518733000041141709000930003031222220012518733000041141709301100009031222220012518733000041141211001330015011222220012518733000041149113301630018011222220012518733000041141216301800009011222220012518733000041143018002000012031222220012518733000041147920002015001541222220012518733000041143720152130007531222220012518733
GSS 2 Episode
000041144504000800024010000000012518733000041144308000900006011222220012518733000041141709000930003031222220012518733000041141709301100009031222220012518733000041141211001330015011222220012518733000041149113301630018011222220012518733000041141216301800009011222220012518733000041143018002000012031222220012518733000041147920002015001541222220012518733000041143720152130007531222220012518733
SEQNUM
DDAYNO
_EPIS
O
ACT_CO
DE
000041144504000800024010000000012518733000041144308000900006011222220012518733000041141709000930003031222220012518733000041141709301100009031222220012518733000041141211001330015011222220012518733000041149113301630018011222220012518733000041141216301800009011222220012518733000041143018002000012031222220012518733000041147920002015001541222220012518733000041143720152130007531222220012518733
SEQNUM
DDAYNO
_EPIS
O
ACT_CO
DE
Related Documents
![ACCOLEDS 2011 website screenshots€¦ · ACCOLEDS 2011 website screenshots [A separate text file has been sent with the HTML code for these pages.]](https://static.cupdf.com/doc/110x72/5fdc20f732ae9f747f799e62/accoleds-2011-website-screenshots-accoleds-2011-website-screenshots-a-separate.jpg)
