Tutorial:Regional yearbook map production
From Statistics Explained
An external contractor (Abaco) will produce all maps for the Eurostat regional yearbook and it is important that the Excel worksheets that are used for producing the maps are being made in a consistent way for all chapters in the publication. The intention of this article is to formulate detailed rules to both the external contractor making the data extractions and Excel worksheets (Informa) and for the contact persons in the statistical units checking the maps when they have been produced. The same rules generally apply for all statistical maps in Eurostat publications, but these rules are specifically intended for producing the maps for the Eurostat regional yearbook. At the end of this article, you have a short checklist for all the elements that should be included in the Excel files.
Here is the link to Eurostat regional yearbook 2012 on the Eurostat website for reference. The page lists links to one PDF in high resolution per chapter and one Excel file per chapter with all tables, figures and maps.
- 1 To choose suitable statistical indicators
- 2 File structure and numbering of the maps
- 3 Map template
- 4 Data coverage and protocol order
- 5 Data not available
- 6 The NUTS classification
- 7 NUTS level displayed on map
- 8 Other spatial units
- 9 Main title and subtitle
- 10 Footnotes
- 11 Eurostat data codes
- 12 Three examples of map titles with footnotes and Eurostat data codes
- 13 The EU-27 value
- 14 The statistical classes
- 15 Colour shades for the statistical classes
- 16 Checklist for the Excel worksheets to be used for producing the maps
To choose suitable statistical indicators
The maps published in the regional yearbook are usually "choropleth" maps, please see: Wikipedia article on Choropleth map, which is a kind of thematic map where the data is sorted into a number of statistical classes, represented by colour shades on the map.
This kind of map is most suitable for statistical indicators where the total numbers/levels of one indicator are divided by another indicator, like population or area (good examples: "Population density", "GDP per inhabitant" or the "Employment rate"), the map should ideally not show totals/levels.
File structure and numbering of the maps
All maps belonging to the same chapter are defined in the same Excel file with one worksheet per map. The first three columns of each Excel worksheet contain; 1) the NUTS codes, 2) the region names and 3) the statistical value.
The other columns (to the right of the actual data) contain other specifications, namely the map legend; the title, subtitle, the footnotes (specifying the exceptions to the title) and the Eurostat data code. All contextual information needs to be provided in three languages: German, English and French.
The map legend should also contain the EU-27 value (if available) and the statistical data classes to the map. Please compare with the data in this example file: Examples of map definitions to check if everything has been included. Here you find examples of data for all NUTS levels; national, NUTS 1 regions, NUTS 2 regions, NUTS 3 regions.
The maps should be numbered in a sequence that is logical for the text structure of the chapter/article and you should avoid changing the numbering of the maps at a later stage in the publication process. Please give numbers to the maps according to the chapter number with a sequential order to the numbering of the maps you would like to include, for example; Map 1.1, 1.2, 1.3, 1.4 and 1.5 for the first chapter in the publication.
The map numbering is not included in the map legend of the map, because sometimes the same map is reused in another publication or in a presentation and then the numbering is not relevant or even false. It's still possible to identify the maps because the map number should be included in the file name of the output file of the map, generally a PDF file.
The same map template is used for all standard maps produced for the Eurostat regional yearbook.
Data coverage and protocol order
The coverage of each maps is the following: the 27 Member States of the European Union, the 4 EFTA countries (Iceland, Liechtenstein, Norway and Switzerland) and the 5 accession or candidate countries (Montenegro, Croatia, the former Yugoslav Republic of Macedonia, Serbia and Turkey). All codes have to be present even if Eurostat does not have data for all these countries - not having data is also an important piece of information to our users.
Iceland (IS) is in fact both an EFTA country and a candidate country (since the 27th of July 2010), but for the sorting, Iceland still stays in the group of EFTA countries.
Croatia (HR) is now an accession country and should become new Member State as of 1 July 2013, but for the first half of the year 2013 Croatia is still sorted among the candidate countries.
Serbia (RS) is a new candidate country since 1 March 2012, but statistical regions have not yet been decided for Serbia, so for the time being only national data can be displayed on maps.
Please remember that the protocol order should be respected not only for the sorting of the countries in the actual data sets, but also for the order of the countries mentioned in the footnotes.
Do not use the Extra-Regio codes (for example FRZ, FRZZ, FRZZZ for France), since it's not possible to display these values on the maps.
For the most up-to-date list of country names in different languages and correct protocol order, please see: Tutorial:Country codes and protocol order.
Data not available
It is important that the NUTS codes for all countries and/or regions are present in the Excel woorksheets, even if data is missing, because of the category called "Data not available". This is one of the data classes in the map and is represented by a dark grey colour. Territories outside the coverage of the standard map template are represented by a light grey colour on the map and for these countries only the country borders are shown, not the regional borders.
"Data not available" should be represented by a colon ":" in the third column of the Excel worksheet called "Value", please avoid having empty cells in the value field, because that is not accepted by the map production tool and will give error messages when trying to produce the map.
The NUTS classification
The EU Member States are divided into statistical regions according to the NUTS Regulation, which is a legal act. The abbreviation "NUTS" stands for "Nomenclature of territorial units for statistics". Regions in the EFTA and candidate countries are called "Statistical regions" and follow the same general rules as the NUTS classification, but for the statistical regions there is no legal base, it's based on so-called "Gentlemen's agreements".
The three different NUTS levels are based on the size of the population living in the regions. In densely populated areas, the regions are smaller and in more sparsely populated areas larger. Existing national administrative structures in each country are taken into account when deciding on the creation or amendment of NUTS regions. The population thresholds for deciding on the NUTS level of regions are the following:
|NUTS 1||3 million inhabitants||7 million inhabitants|
|NUTS 2||800 000 inhabitants||3 million inhabitants|
|NUTS 3||150 000 inhabitants||800 000 inhabitants|
This means that for some small Member States (both regarding population and/or area), the same territory is "repeated" on one or more NUTS levels. For example, in the following six Member States the whole country consists of one single region on NUTS level 1 and 2: Estonia (EE), Cyprus (CY), Latvia (LV), Lithuania (LT), Luxembourg (LU) and Malta (MT). This is also the case for two EFTA countries; Iceland (IS) and Liechtenstein (LI) and two candidate countries; Montenegro (ME) and the former Yugoslav Republic of Macedonia (MK).
The European Commission decides in agreement with the European Council and the Parliament on revisions of the NUTS classification based on proposals from Eurostat within regular intervals. A new version of the NUTS Regulation, "NUTS 2010", entered into force on 1 January 2012 and that is the current NUTS version. The Eurostat regional yearbook 2013 will only use NUTS 2010 and will not accept any old NUTS codes according to NUTS 2006.
The codes for the Statistical regions in Croatia (HR) was also revised in 2012 and the new codes for Croatia will be used in the 2013 edition of the Regional yearbook.
For more information about the NUTS classification, please consult: the NUTS dedicated section on the Eurostat website
NUTS level displayed on map
In the document called "Examples of map definitions in Excel", you will find examples of map definitions for all NUTS levels; national level, NUTS 1 regions, NUTS 2 regions and NUTS 3 regions.
The most common NUTS level used for the maps in the Eurostat regional yearbook is NUTS 2 regions, but we would also like to introduce more maps for NUTS 3 regions, for all subjects for which data is available, e.g. Population, GDP, Labour market and Tourism.
In some cases, regional data for a given country is missing in the database and then the general rule is to take data from the NUTS level immediately above. If, for example, data for NUTS 2 regions is missing for a country, then you can show data for NUTS 1 regions instead. In the Excel worksheets, you just insert the codes for the NUTS 1 regions instead of NUTS 2 regions and the mapping software will "understand" and show the correct borders on the map. You must however not forget to mention all exceptions to the general NUTS level given in the main title in the footnotes to that map (see examples below).
It is also important only to give reference to a meaningful NUTS levels in the footnote. For instance, if regional data for NUTS 2 regions is missing for Denmark, please give reference to "national level" instead of "NUTS 1 regions" in the footnote, since the country of Denmark consists of one single NUTS 1 region; this is what we mean by meaningful.
Other spatial units
There are also other spatial units, other than the NUTS regions, in the maps for the Eurostat regional yearbook, like for example Urban Audit cities, airports, or sea ports. The general rules for these maps are the same as for the regional maps; the first column contains the geographical code, the second the label of the geographical entity and the third column contains the statistical value. The rules for writing the textual information are also the same as for any other statistical map.
The maps for the chapter called "Coastal regions" only displays the NUTS 3 regions along the coastline according to the methodological definition of "Focus on Coastal regions", please see examples of these maps in in the 2012 edition of the publication.
Main title and subtitle
The titles of the maps should be written as consistently as possible throughout the publication. A map title should be self-explanatory and unambiguous for the users. From the title alone you should be informed about which statistical subject the map is about, which reference year it concerns, which unit is being used, and so on. Please remember that the maps are often taken out of their context and used for example in presentations, so all explanations needed should in principle be included in the map itself and in the footnotes.
The title consists of a main title and a subtitle. If totals are being used (e.g. for age groups or gender) this does not have to be mentioned explicitly in the title. Please avoid using abbreviations, if they not are very well known and easily understandable to a wider public.
The title consists of the following elements and in this order:
Main title (bold):
Indicator or variable name of the statistical values shown on the map (please use the same title as in the database and/or Main table, if you don't have a strong reason for not doing so)
Then the different statistical breakdowns are listed, for instance by different age groups, by sex, by NACE-activities, by NUTS 2 regions and so on (always with the preposition "by" before the breakdown and the NUTS level should be mentioned as the last element before the year).
Reference year or years concerned
Footnote number (¹) if the map needs a footnote (always at the very end of the main title). The footnote number is written in superscript and in brackets.
Subtitle (normal, in brackets):
The unit or units of the statistical indicator, please use % for percentage, but spell out "percentage points" if that is the case. The subtitle usually starts with a lower case letter (unless national spelling rules prevent it, which is for example the case with German nouns).
The footnotes are normally specifying exceptions to the title and should only be used if needed. If you have a footnote to the map, you add (¹) at the end of the main title. A map can only have one footnote, called (¹). If you have many exceptions to the title, you list all the exceptions next to each other in the same sentence, with a comma between the country or region names and the exception and with a semicolon between the different elements of the footnote (see examples below).
The exceptions to the title in the footnotes can be of different types; exceptions regarding the reference year and the NUTS level are the most common, but there can be other exceptions as well, regarding for example estimated values, confidential data or definitions of the statistical content.
When specific countries or regions are mentioned in the footnote; please spell out the country names instead of using the country codes and list them in protocol order. Please note that the names of the countries are written in the language of the publication (German, English and French), but the region names are written as in the NUTS Regulation, which means in the official language of each country and with the NUTS code in brackets immediately after the region name (see examples below).
Eurostat data codes
The Eurostat data codes are being used in all Eurostat publications and they are placed directly below all tables, figures and maps and hyperlinked in the electronic version. The purpose of the data code is to give users quick and easy access to the most recent data available for a given statistical indicator, without having to spend time on searching the database. The practical use of the Eurostat data code is of course more evident in the electronic version of the publication (PDF-version and Statistics Explained) where the actual code is hyperlinked to the data set, but the data codes are also included in the printed version, as information for identifying the correct dataset in the Eurostat database.
There are two different kinds of online data codes, one leading to a Main table (previously called "Predefined table") and the other one leading directly to the Eurostat database (open dataset). Here are examples of both types of Eurostat data codes:
Source: Eurostat (online data code: tgs00007) Main table
Source: Eurostat (online data code: reg_e2gdp) Open dataset
If a Main table exist for the indicator, please use that as your first choice, but if no Main table has been defined, you have to link to one or several dataset codes instead.
In the Excel file and in the PDF version of the map the data code is static; the layout company (Jouve) will insert the actual hyperlink at a later stage in the publication process.
The hyperlinks that will be inserted at a later stage will be so called generic or stable hyperlinks, which actually lead to a search query for the dataset code instead of leading directly to the dataset. The advantage of the stable links is that it is possible to redirect the link from an old dataset to a new one, if the dataset code changes. For more clarifications regarding stable hyperlinks, please see: Tutorial:Stable Eurostat URLs
If any data set code in your chapter is changed before the final print version of the regional yearbook is produced, please give notice to unit E4 as soon as possible, because the code in the printed version should then also be corrected.
Three examples of map titles with footnotes and Eurostat data codes
Map 1.1: Population density, by NUTS 3 regions, 2010 (¹)
(inhabitants per km²)
(¹) Population density is calculated as the ratio between (annual average) population and the surface (land) area; land area is a country's total area, excluding the area under inland water; Denmark, Germany, France, Cyprus, Italy, Hungary, Poland, Portugal, Slovakia, Liechtenstein and Montenegro, total area has been used instead of land area; Belgium, Sachsen (DED), Illes Balears (ES53), Canarias (ES7), France, the United Kingdom and Norway, 2009.
Source: Eurostat (online data code: demo_r_d3dens)
Map 4.1: Participation rates of four year-olds in pre-primary and primary education (ISCED levels 0 and 1), by NUTS 2 regions, 2010 (¹)
(% of four year-olds)
(¹) EU-27, Estonia, Cyprus, Latvia, Lithuania, Luxembourg, Malta, Iceland, Liechtenstein, Switzerland, the former Yugoslav Republic of Macedonia and Turkey, 2009; Greece, 2008; Vlaams Gewest (BE2), 2007; Région wallonne (BE3), 2001; Germany and the United Kingdom, by NUTS 1 regions; Greece, the Netherlands, Switzerland and Croatia, national level.
Source: Eurostat (online data code: educ_regind)
Map 4.3: Early leavers from education and training, by NUTS 1 regions, 2010 (¹)
(% of 18-24 year-olds)
(¹) Proportion of those aged 18-24 years having attained at most a lower secondary education and not being involved in further education or training; France, Sweden and Switzerland, provisional; Hamburg (DE6), 2008; Brandenburg (DE4), 2007; Mecklenburg-Vorpommern (DE8), 2005; Finland, national level.
Source: Eurostat (online data code: edat_lfse_16)
The EU-27 value
Previously the value for EU-27 was sometimes mentioned in the map legend and sometimes not. Now we would like to include the EU-27 value in as many maps as possible, because it gives an indication of the relation of each region in comparision to the EU-27 average. The EU-27 is displayed just above the statistical classes in the map legend (below the actual map).
The statistical classes
For defining the statistical classes to each map, you have to analyse the distribution of the data and decide which method for grouping the classes you are going to use. The maps in the regional yearbook usually have 3-5 statistical classes and there are different methods for sorting the data. The method you apply depends on the message you want to convey, different sorting methods give very different results on the map!
One recommended method is to use natural breaks in the distribution of the data series and group the data according to them; another recommended method is to make one class around the EU-27 average and to make two classes below and two classes above the EU-27 average (5 classes in total).
Please remember to define the statistical classes you want to have for each map in the Excel worksheets.
Colour shades for the statistical classes
The regional yearbook will follow the Eurostat publication theme colours for the colouring of the statistical classes on the map. The theme colours are described in detail in the Eurostat publications graphical style guide
The Eurostat publication themes are the following (please consult the Eurostat styleguide for more details regarding the colours):
|Publications theme||Statistical subject||Theme colour|
|Theme 1||General and regional statistics||lilac blue|
|Theme 2||Economy and finance||lilac|
|Theme 3||Population and social conditions||yellow|
|Theme 4||Industry, trade and services||blue|
|Theme 5||Agriculture and fisheries||green|
|Theme 6||International trade||red|
|Theme 7||Transport||(grey) turquoise|
|Theme 8||Environment and energy||turquoise|
|Theme 9||Science and technology||orange|
For Theme 7, Transport, the main theme colour is grey, but since grey already is needed for the category "Data not available" on the maps, this colour cannot be used as a theme colour, so for Transport we will use the second theme colour, which is turquoise.
The map production contractor will automatically apply the correct colour scheme to the maps for each chapter; this doesn't need to be mentioned in the map definitions in Excel. You should however remember to check if the correct colours have been applied when you get the map back for checking and approval.
Checklist for the Excel worksheets to be used for producing the maps
1st column: NUTS codes in protocol order and for a specific NUTS level, the NUTS codes should be in upper case, formatted as text and with no leading space before or after the code.
Please note that the code DEC0 for the German NUTS 2 region "Saarland" will automatically be converted to "Dec-00" or to a numeric value, due to specific Excel parameterization, if the cells are not formatted as text.
Different NUTS levels can be used for individual countries if data is missing, but there should be no territorial overlap and no missing codes and the NUTS level exceptions have to be mentioned in the footnote to the map.
2nd column: the label of the Region name in the official language of the country as spelled in the NUTS Regulation (please compare with "Examples of map definitions in Excel" if you need to fill in the region names).
3rd column: the statistical Value of the indicator for each NUTS region, formatted as number and with two decimal places. Missing data is represented by a colon ":" instead of a value.
Textual information (map legend):
Map number: the number of the chapter and sequence within the chapter (e.g. Map 12.3)
Main title (in bold): name of statistical indicator, different statistical breakdowns, "by NUTS X regions", reference year(s) + footnote number (¹) if needed
Subtitle (new line, normal style and in brackets) name of statistical unit
Statistical classes (3 - 5 classes)
Footnote, (if needed). The footnote starts with (¹) and lists any exceptions to the title. List the countries and or regions in protocol order and use comma between the countries/regions and the exception and semicolon between the different elements of the footnote. Normally the footnote is in one single sentence.
Eurostat data code e.g. Source: Eurostat (online data code: tgs00007)
All textual information in the map should be provided in German, English and French.
Please use this Excel file as an example for producing Excel worksheets to be used for producing maps: Examples of map definitions