29 Apr 2014

Excel and relational databases

#13  Excel and relational databases

NB: This post has some files (embedded in 1 zip), see ‘Downloads’ at bottom of page.

In my previous post:
I showed how you can calculate the medal-totals grouped by team, and I also said that this calculation could be much simpler in SQL, the language used in relational databases, like e.g. MS Access. In this post I´ll show you how you can do this in MS Excel, so in a spreadsheet-program. This is useful for those who don´t have an MS Office-version which includes MS Access.

The general idea is to split the application in 2 layers (2 Excel-files): the database-layer and the client-layer, like in a client/server-application.
The database has data about skaters, races and teams. Each type of data is put in it´s own table and between the tables exist relations as you can see in fig.1, e.g.: a team consists of 1 or more skaters and a skater participates in 1 or more races. The tables are linked by their primary keys (PK) and foreign keys (FK), columns which end with ID, e.g. TeamID (PK of table Team and FK of table Skater).

fig.1: Tables in database

The data in these tables you can see in fig.2 and file Skating_DB_XL.xlsx, which is an Excel-workbook with 3 worksheets, one for each table.

NB: I also made this database in MS Access for illustration-purposes, see fig.3 and file Skating_DB_AC.accdb 

fig.2: Excel-file with database

fig.3 Access-file with database

To be able to use these 3 tables in this Excel-file (´database-layer’) in another Excel-file (the ‘client-layer’), you must create a ‘Range’ for each table, see fig.4 and these sites how to do it:

fig.4: Named Ranges for tables in database

In the Excel with the client-layer, you now must create a connection to the Excel with the database-layer, which is stored in a dqy-file (Excel ODBC Query), see these sites how to do this:

With the connection to the database created, you now can use Excel Query to create your queries to the database,  using the Query Wizard (see fig. 5-6) or hand-written SQL (see fig.7 etc.)

fig.5: Excel Query

fig.6: Excel Query with result

Some examples of database queries in SQL:

* Example 1

File: Skating_XL_Query2b.xlsx, Skating_XL_Query2.dqy

Query: ranking for all skaters in top 3 and pivot-table

T_Skater.Gender, T_Skater.SkaterName,
T_Team.TeamID, T_Team.Trainer, T_Team.TeamName,
 T_Race.Race, T_Race.Ranking
FROM `C:\Temp\Skating_DB_XL.xlsx`.T_Race T_Race,
`C:\Temp\Skating_DB_XL.xlsx`.T_Skater T_Skater,
`C:\Temp\Skating_DB_XL.xlsx`.T_Team T_Team
WHERE T_Race.SkaterID = T_Skater.SkaterID
AND T_Skater.TeamID = T_Team.TeamID
AND T_Race.Ranking <=3

fig.7: Result query

fig.8: pivot table based on result query (fig.7)

*Example 2

Files: Skating_XL_QueryXL_6b.xlsx, Skating_XL_Query6.dqy

Query: medal-total per team


 Sum(IIF(T_Race.Ranking=1,1,0)) AS 'Total_Gold',
Sum(IIF(T_Race.Ranking=2,1,0)) AS 'Total_Siver',
Sum(IIF(T_Race.Ranking=3,1,0)) AS 'Total_Bronze', Sum(IIF(T_Race.Ranking=1,1,0))+Sum(IIF(T_Race.Ranking=2,1,0))+Sum(IIF(T_Race.Ranking=3,1,0)) AS 'Total',
FROM `C:\Temp\Skating_DB_XL.xlsx`.T_Race T_Race,
`C:\Temp\Skating_DB_XL.xlsx`.T_Skater T_Skater,
`C:\Temp\Skating_DB_XL.xlsx`.T_Team T_Team
WHERE T_Race.SkaterID = T_Skater.SkaterID
AND T_Skater.TeamID = T_Team.TeamID
AND ((T_Race.Ranking<=3))
GROUP BY T_Team.TeamName
ORDER BY Sum(IIF(T_Race.Ranking=1,1,0)) DESC, Sum(IIF(T_Race.Ranking=2,1,0)) DESC,  Sum(IIF(T_Race.Ranking=3,1,0)) DESC, T_Team.TeamName

fig.9: Result query

Note 7-5-2014:
I wrote a new post with more examples on the Skating-database in this post, using PowerPivot and PowerView, see:



No comments: