Westward Ho: New Frontiers

1
Westward Ho New Frontiers

• USG Academic Data Mart Project Update
• Georgia Summit (September 8-10, 2004)
• Savannah, GA
• Presented by Charles Gilbreath (GSU) and Debbie
  Head (KSU)

2
What A Data Warehouse Is Not!

• Transactional systems are designed to respond
  rapidly to individual events such as registering
  for a course, paying fees, etc.
• Transactional structure is highly normalized
  (broken into many small pieces)
• Transactional systems are not designed for
  queries

3
What a Data Warehouse Is!

• A data warehouse is a set of tables that are
  designed to respond quickly to queries.
• They are denormalized (data may be repeated
  within a table)
• They are designed to store history.
• They are designed to bring pieces together from
  different transactional systems, such as Student
  Information, Human Resources, Facilities, etc.
• They may contain multiple data marts that store
  related data

4
USG Academic Data Mart (ADM) Defined

• The USG Academic data mart is designed to
  incorporate the institutional data from the
  legacy systems of SIRS, CIR, FARS, RUR, Graduate
  Salary Survey, High School Feedback, and Learning
  Support/Core Curriculum.
• The data collected in the Academic Data Mart can
  be used by institutions for both local and
  official reporting needs.

5
ADM in simple terms.

• Final objective of Enterprise-wide data warehouse
  will be a hybrid of old reporting systems (SIRS,
  CIR ,etc) with new data structures that will
  consider institutional needs
• Transactional systems (Banner) will feed the data
  directly to the warehouse.
• Data elements are arranged in tables in a
  database managed by OIIT. Data structures reflect
  institutional needs.

6
ADM expectations

• The data fields were selected initially based on
  the data fed to SIRS and CIR.
• It will expand beyond those elements when it
  proves its functionality
• Still working on how to load longitudinal data
• Canned reports, SER for example, will be
  available
• Sharing of reports generated by others in our
  group so you wont have to recreate the wheel
  each time.

7
Why Does IRP Care about the ADM?

• For institutions with limited resources, people
  and equipment, they can access their own data to
  do internal analyses as desired.
• Brings the USG more in line with the current
  technology in terms of housing and using data
• Takes the data jail concept and lets us
  actually get some data out
• Hopefully brings some consistency and
  understanding about what goes into reporting
• Reporting should become easier.
• Data warehouse tables should match production
  tables

8
Data Warehouse Structure

• ERD Entity Relationship Diagrams show the main
  table (FACT table) and how other tables
  (Dimensions) connect to the main table. It is a
  detailed scheme of the many elements within each
  component
• Find these at this link

http//www.usg.edu/usgweb/sitcap/usg123_aca/index.
phtml?idpmd/pmd_br
9
What data are accessible?

• There are 5 different data components of the ADM
  organized into data marts that are collections
  of associated data
• Class Session (Class schedule/catalog)
• Student Profile (Demographics)
• Course Enrollment (Registrations)
• Student Term Enrollment (Student Record)
• Student Test Results (Test scores)

10
Class Session

• Class Session data are extracted from Banner
• Includes course catalog data such as Course
  number, section, times and days offered, credit
  hour value of the course
• Does not include credit hours generated or number
  of students enrolled

11
Student Profile

• This data mart will provide much of SIRS data.
• Includes many of the SIRS data fields
• You can access and report and clean the data
  prior to releasing it to OIIT.
• Editing reports should let us scrub it better
  before it goes into the official warehouse

12
Student Course Enrollment

• Will load some of CIR enrollment data
• Will be the source for Credit Hour Production
  Reports by the USG.
• Will link to the Class Session Component so that
  individual student enrollment information can be
  accessed

13
Student Term Enrollment

• Contains data on each student enrolled in one or
  more courses in an academic term
• Contains cumulative data for each student
• Is linked to demographic, geographic, etc. data
  for each student.

14
Student Test Results

• Contains information on detail level of test
  results as recorded in Banner
• Allows selection on individual test types (ACT,
  SATV, SATM, etc.)
• Allows selection by student characteristics
  (ethnicity, sex, etc.)

15
Getting Data Back Out

• Business Objects pre-selected sets of data
• What makes sense in terms of types of information
  we (IRP) need to know?
• For example A predefined First-time Full-time
  Freshmen grouping so average SAT, gpas, ages,
  ethnicity, gender, CPC, LSP could be gathered
  just about that group?
• What else?

16
Process for Meeting IRPs needs

• Identify data needs and generate list of desired
  reports
• Timetable for us and OIIT
• What is review process for requests of reports?
  (Does IRP recommend a standing data warehouse
  committee?)

17
Standing Data Warehouse Committee

• Identify data needs and pass on to report
  developers (Some developers may be OIIT and some
  may be IRP members)
• Facilitate sharing reports
• Develop a process for recommending changes to
  data warehouse structure
• Members reflect data warehouse user community

18
Finding Information About What is in the ADM!
http//www.usg.edu/usgweb/sitcap/
19
ERD Web site
20
Using Discoverer

• Reporting tool provided by OIIT
• Administered at system level
• Allows us to see our own institutional data
• Can build our own ad hoc reports
• If a report that would be beneficial to all,
  submit it to committee for review and approval to
  be put in the master list of available reports

21
Round the campfire

• Questions, comments, suggestions
• Meet the trail bosses of the ADM
• Lori Jarrard
• Glenn Fernandez