Skip to content

Standard Shape

This page describes how to build and configure a standard shape data cube. For more information about the data shapes supported by csvcubed, see Shaping your data. The instructions below assume a basic understanding of writing a qube-config.json file.

The standard shape extends the common structure by requiring that each row has a measures column and a units column; these columns define the measure and unit (of measure) for each row.

The standard shape is most appropriate where you have a sparse data cube, i.e. there are a large number of possible combinations of dimension values, but very few of them have observed values recorded. If your data cube is dense, then consider using the pivoted shape. See Converting to standard shape for instructions on how to convert the shape of your data in Python and R.

Single Measure

In our example, the single measure observed is Number of Arthur's Bakes and the corresponding unit is Count.

Year Location Value Status Measure Unit
2022 London 35 Provisional Number of Arthur's Bakes Count
2021 Cardiff 26 Final Number of Arthur's Bakes Count
2020 Edinburgh 90 Final Number of Arthur's Bakes Count
2021 Belfast 0 Final Number of Arthur's Bakes Count

The simplest qube-config.json we can define for this data set is:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
{
    "$schema": "https://purl.org/csv-cubed/qube-config/v1",
    "title": "Arthur's Bakes stores in UK cities from 2020 to 2022",
    "description": "The number of Arthur's Bakes stores in cities across the UK between 2020 and 2022.",
    "creator": "https://www.gov.uk/government/organisations/hm-revenue-customs",
    "publisher": "https://www.gov.uk/government/organisations/hm-revenue-customs",
    "columns": {
        "Year": {
            "type": "dimension"
        },
        "Location": {
            "type": "dimension"
        },
        "Value": {
            "type": "observations"
        },
        "Status": {
            "type": "attribute"
        },
        "Measure": {
            "type": "measures"
        },
        "Unit": {
            "type": "units"
        }
    }
}

It is possible to use the configuration by convention approach to generate a valid standard shape cube without defining a qube-config.json at all. Just ensure that your columns use the conventional column names appropriate to their type.

Multiple Measures

One of the benefits of the standard shape is that it is relatively straightforward to add new measure types and unit types; all that you have to do is add additional rows to your data set with the appropriate units and measures present.

We can extend our example data set so that it now includes revenue values for the given year by adding rows to the table:

Year Location Value Status Measure Unit
2022 London 35 Provisional Number of Arthur's Bakes Count
2022 London 25 Provisional Revenue GBP Sterling, Millions
2021 Cardiff 26 Final Number of Arthur's Bakes Count
2021 Cardiff 18 Final Revenue GBP Sterling, Millions

Note that extending the data set to include multiple measures does not require any changes to the qube-config.json column definitions.

The same data could be represented in the equivalent multi-measure pivoted shape as follows:

Year Location Number of Arthur's Bakes Number of Stores Status Revenue Revenue Units Revenue Status
2022 London 35 Provisional 25 GBP (Sterling) Provisional
2021 Cardiff 26 Final 18 GBP (Sterling) Final