February 18, 2021
    7:00 PM EST - 8:30 PM EST
     Add to Calendar

    Zoom Meeting



    February 2021 Chapter Meeting Recording


    DAMA GA - Data Modeling.pdf


    Conceptual vs. Logical vs. Physical Data Models

    In our field there appears to be general agreement on the definition of each of these kinds of data models. However, upon closer examination, the definitions and distinctions are quite fuzzy. This presentation challenges the common understanding and naming of conceptual, logical, and physical data models. Then it offers some alternative ways to think about them which are more grounded and useful.

    What you'll get from this presentation:

    * A brief review of "Kinds of Data Models" by David Hay which eloquently expresses the common understanding (excerpts from his YouTube video at ).

    * Exploring what characterizes and differentiates conceptual, logical, and physical data models.

    * Insight into what is meant by a conceptual model using several examples.

    * Managing multiple conceptual models, keeping them up to date and in sync with your logical models.

    * The methods and guidelines which drive the logical data modeling process. Are all data models logical, even conceptual models?

    * What is a data modeling scheme in logical data modeling, why it is so important?

    * To help us understand the differences, we examine the steps taken in the logical data modeling process. What are the various constructs and when are they introduced?

    * When does a logical model become a physical model; where is the separation?

    * Is a relational data model logical or physical (with its entities, attributes, and tables satisfying 1NF)?

    * What is the difference between a physical model and an implementation model?

    It has become common to speak of levels of data models.

    Conceptual is commonly thought of as a high-level, enterprise-wide, abstract data model. The conceptual model is the one to present to executives for an initial understanding of the data model. It may also be a preliminary model to which detail is added as the modeling process progresses.

    Logical is something in between, adding detail to the conceptual model but free of physical implementation details which do not contribute to the logical understanding of the model. Sometimes a Relational or Entity-Relationship (ER) model is considered the logical model.

    Physical, at the bottom, is how the data is stored in a database implemented in some database management system (DBMS) or NoSQL tool, its physical realization.

    In preparation for this presentation, attendees are encouraged to view the 35 minute video from David Hay entitled “Kinds of Data Models -- and How to Name Them.” (Search on YouTube or go to: ). David is one of the pre-eminent thinkers and authors on data modeling. This video captures the essence of popular thinking regarding the characterization and distinctions between conceptual, logical, and physical data models. My presentation could be considered a rebuttal or contrary view to the one promulgated by David Hay.



    Dr. Everest is Professor Emeritus of MIS and Database in the Carlson School of Management at the University of Minnesota.  With early "retirement," he continued to teach Advanced Data Modeling until 2018.  His Ph.D. dissertation at the Univ of Pennsylvania Wharton School entitled "Managing Corporate Data Resources" became the text from McGraw-Hill, "Database Management:  Objectives, System Functions, and Administration" in 1986.

    Gordon has been teaching all about databases, data modeling, database management systems, data administration, and data warehousing since he joined the University in 1970.  He is a frequent speaker at professional organizations such as DAMA Enterprise Data World, Data Modeling Zone, and local DAMA chapters.  He received the DAMA-I Outstanding Academic Achievement Award in 2006 and again in 2011.