Introduction to the NISRA Data Portal

Introduction to the NISRA Data Portal

The NISRA Data Portal is an application where you can access data produced by NISRA.

The application will give users the opportunity to view and filter datasets, run meta-data searches, plot interactive charts and visualise data on interactive maps, save queries, create favourite tables and widgets, share results, access data in multiple formats (CSV, JSON-stat, PX and XLSX) following the Open Data recommendations and automate processes by reading data via API queries.

This manual will help you to navigate and use the application. A training video and a document containing FAQs are also available. If you have any issues with the application, please email info@nisra.gov.uk.

Accessing the Data Portal

When you click on the NISRA Data Portal https://data.nisra.gov.uk you will arrive at the homepage below.

The data can be accessed, and most functionality is available, directly from the home page, without the need to create an account. Some extra functions are available when you create an account and log in. Creating an account and logging in is discussed in more detail in the section on logging in to the system, along with the extra functionality this enables.

In the middle of the home page there is a box showing the last updated tables and when they were last updated. You can choose the date from which you wish to display your tables by using the date box at the top right. The API for these tables sits just below the box. On the left hand side of the page there is a browse tab where data can be found according to the themes and subthemes to which they belong. There is also a search box which can be used as a shortcut to find the datasets you require.

Quick links are available below the browse option, linking to the training materials and FAQs for the data portal, the Census 2021 Flexible Table Builder, the NISRA website, and the NISRA Visualisations. These links will be available on each page as you move through the application.

Search facilities

There are two main ways to find the data you need:

Use the Browse facility

The browse facility is positioned on the left hand side of the screen. This is where you will find a list of themes. Selecting a theme will open a further list of subthemes to help narrow your search, and within each subtheme is a list of specific topics. Beside each topic, in the blue box, is the number of tables associated with that specific topic. For example, when you open Health and Social Care in the browse menu and choose the subtheme Health Inequalities, the following list of topics, along with the number of tables, will be displayed. This number will update automatically as tables are added to the system.

Opening the Diet and Dental Health topic under Health Inequalities within the browse facility will display the available tables within that topic, as below. Each dataset has a unique code, a title, and below that, the variables included in the table. In the example below, the unique table code is CHILDBMILGD, the title is Childhood BMI, and the variables that will appear in the table are Local Government District, Gender and Grouped Academic Year. Note that tables can have the same title but will always have a unique code and the variables will also differ in some way.

There are a number of icons which may appear alongside the table code and these can be identified by hovering your mouse over them. The two icons linked to the table below denote that these are official statistics (blue icon) and that an interactive map (globe icon) is available for this table.

Use the Search facility

If you know the unique code or the name of the dataset you want, you can simply enter it in the search box (which appears at the top of every page).

So, for example, entering the unique code CHILDBMILGD in the search box will take you straight to the requested table, and the page below will be displayed.

The search facility can also be used to find the contents of a table even if it is not in the table title. So for example, entering ‘obese’ in the search box returns the the same two tables as they are the only ones currently on the system that contain the word ‘obese’ as a category within their statistics or classifications. The system will return all tables containing the term “obese” as they are added to the system. On the left hand side you can also order your selection alphabetically, or by date (newest first or oldest first). You can also filter your results by the subject or product categories you are interested in.

Manipulating/Customising the Tables

Once you click on the table you are interested in (here we are using CHILDBMILGD - Chidhood BMI), the following screen layout appears. In the main body of the table, the statistic(s) always appear in the first box. The categories contained within each variable can be seen in each of the relevant boxes, and the number of categories is highlighted in blue beside the variable name.

Within the table you can customise the data, plot a chart, or generate a map (if one is available). The size of the dataset is indicated by the number of cells, which is shown at the bottom right hand side of the page (1152 in total for this dataset).

On the left hand side of the screen there are a number of other options.

• The share table option allows you to share a link to the dataset through a number of media such as X, WhatsApp and Facebook
  
• The note box will contain any relevant notes associated with the table, or links to notes ascribed to it
  
• If you wish to download the full un-pivoted dataset use the full download option. There are a number of formats including CSV, JSON-stat, PX, and XLSX
  
• In the contact box you will see the contact details for the statistician/department responsible for the table
  

While you can use the full download function here, you can also view or download the full dataset by scrolling to the bottom of the table and clicking on view or download. When choosing this option, there is no need to highlight anything within the boxes. Note that if you do not highlight anything within a variable then all of the options within that variable are selected by default.

Where you do not require the full dataset and you are looking for specific data, for example in relation to your local area, you have the option to filter the data. You can do this by highlighting the particular categories you want to use. To pick two that are not adjacent, hold down the Ctrl button and use your mouse to highlight the ones you want. To highlight a selection of categories adjacent to each other, click on the first category you want, hold down the Shift key, click on the last category, and all those in between will be highlighted.

If you require all categories within a variable you can scroll and highlight them, or simply click on the box to the left of the variable name and all options within that variable will be selected. Again, you can view, download your selection, or use the pivot option, as required. These functions are illustrated in the example below.

When viewing your chosen output, if there are too many columns for the screen you are using, a ‘+’ symbol will appear at the top left of the statistic boxes in your results table. This will enable you to view the data within the space available to you. This may be the case if you are using a mobile device to download the tables where the view will be more limited than if using a larger screen.

Example 1 – Proportion of P1 children who are obese by local government district

In this example, we are going to view Proportion of P1 children who are obese, by each local government district in the grouped academic years 2017/18-2019/20.

To do this, use table CHILDBMILGD (Childhood BMI) to look at levels of obesity in P1 children by local government district.

• First, choose your statistic from the first box, in this example the first statistic, click on ‘Proportion of P1 children who are obese (%)’, this will then become highlighted.
  
• From the next box choose your required time period(s). As we are looking at the most recent results, highlight 2017/18-2019/20.
  
• In the local government district box, you can highlight the districts you are interested in or, in this case, highlight everything either by right clicking on your mouse and scrolling down, or checking the box to the left of the variable name. If you don’t highlight anything then all categories in the box will be included by default.
  
• Since we are not looking at breakdown by gender, select the aggregated measure ‘All Persons’.
  

If there is only one option in a box (e.g., some tables only have one statistic) then this option will be used by default. As discussed above, if there is more than one category and none is highlighted, then all will be included in the output. This may be important when considering the size of your output, especially in larger datasets. The Data Cells Selected will give an indication of the cell count for your current choice (here it is 12 out of a possible 1152).

Using the view function will produce the following results table. Within this table you can click on the value button (or any of the other header buttons) to sort the results alphabetically, or from low to high or high to low, whatever is relevant.

You can use the download facility to download your data selection to different software packages such as CSV, JSON-stat, PX, and XLSX.

The pivot function allows you to pivot by a particular variable, changing the format of the table and making it easier to read the output in the table. This is illustrated in Example 2 below.

Once you have chosen your pivot option you can use the download option and choose the XLSX option from the menu. Open the XLSX with any spreadsheet application and the pivoted sheet will be available at the bottom.

The advanced pivot option below is guidance to help you if you wish to pivot further using excel.

Example 2 – Proportion of P1 children who are obese by local government district and gender

Using the reset button in the table allows you to clear your previous statistics and run a new table. Clear your previous table and this time we will look at levels of obesity in P1 by local government district and gender.

Highlight the data you are interested in, as before, except this time highlight females and males in the gender box as we are going to look at these separately. Again the data can be viewed or downloaded. Using the view option will produce the output below.

In this case, using the pivot option will re-format the results and allow you to view them in a way that makes the comparison easier. The results below have been pivoted by gender.

To view your pivoted output in Excel, click download and choose the XLSX option from the menu. Open the spreadsheet and from there you can choose the pivoted sheet.

APIs and Widgets

Upon selecting a table or selections within a table, an API is automatically generated.

A widget function also exists which can be used if you wish to publish your selection to your own blog, website or dashboard, or for your own personal use. To utilise the widget:

• First select the specific data (or whole table) you are interested in, then select view.
• Scroll down and click on the widget tab – circled in the screen shot below.

The following options are available within the widget and there is also an information (i) icon beside each option.

• Auto-update - when set to true, your selection will automatically update if the relevant data is updated by NISRA in the portal, so you will not need to do this manually.
• Fluid time – when set to true, maintains the same time pattern when new time points get published. This means that if you select the last 5 years of data then as the data is updated on the portal it will always take the last 5 years. However, if you select ‘all’ time periods the system will be set to the exact number of (current) years of data (e.g., 10), meaning that if the data is updated it will only save 10 years data and the oldest will be dropped. If you always want all time periods when you update, then set fluid time to false.
• Copyright - when set to true, the copyright appears.
• Link - when set to true, the source will be hyperlinked back to the data source.
• Title - when set to true, allows you to add a title to the widget – the default is the current table title.
• Pagination - when set to true the outputs are collated into pages, rather than one full table – this is important if you are working with a large selection.
• Buttons – when set to true, allows you to add CSV and print options to the widget.
• Search – allows you to add a search box to the widget.
• Responsive - when set to true, automatically adapts the widget for different screen sizes.
• Hide columns – gives the option to hide columns in your output when all variables are the same, e.g., data based on output from a single year. You will be given options to either keep or eliminate these columns from your output (see screen shot above).

The Custom JSON contains the API for your selection.

The snippet code in the widget can also be embedded into blogs, dashboards, reports etc. Once you are happy with your selection, you can either copy the code to clipboard or download. Copying to clipboard can be used to paste into your blog, website or dashboard. Selecting download creates a html of your selection which can be saved to your desktop and will automatically update if the table updates on the portal, as long as the relevant options in your widget have been selected.

Charting the data

The chart function enables you to chart the data, with a range of options available as shown below. Once you have chosen the chart option, use the chart type box to choose the one you want.

Example 1 – Column chart showing proportion of P1 children who are obese by local government district

In this example a simple Column chart is used to illustrate levels of obesity among P1 children in each local government district.

• Use the chart type option to choose a Column chart.
  
• This will open the x-axis label box. Name the x-axis - e.g. ‘Local Government District’
  
• Highlight all districts, including NI (to give a reference number).
  
• Click on series at the bottom left of the table (highlighted above). Note that the ‘Local Government District’ box has now disappeared from the main body of the table.
  
• This will now open the y-axis label box (below). Name the y-axis - e.g. ‘Percent’.
  
• Here you can also name series 1 - e.g. ‘P1 children who are obese’.
  
• To populate the chart with the appropriate statistics, highlight the relevant statistic (Proportion of P1 children who are obese), grouped academic year (2017/18-2019/20), gender (all persons).
  

The options button lets you choose how to display your chart, with different options available for different chart types. For example, When creating a column, bar or line chart, the following display options will be available using the options function above, although not all may be applicable.

• Stacked - when set to true will stack each series to show comparisons between data
• Stacked percentage - when set to true will stack each series as a percentage
• Sort by value - when set to true will sort the variables by value (note that this can only be done on a chart with one series)
• Auto-scale - if set to true the starting value of the y-axis is automatically calculated; if set to false, it automatically starts at zero
• X-axis max no. of steps - permits the user to set the maximum number of ticks and gridlines to show on the x-axis
• Legend - allows users to position the legend relative to the chart

• Finally, use the view option to generate the chart.
  
• Using your mouse to hover over the chart will highlight the numbers within each sector.
  

You can also download the chart as a simple PNG or by using the interactive widget facility.

As described earlier, once you have downloaded your chart, the snippet code in the widget can be embedded into your blogs, dashboards, reports etc., and will automatically update your chart if the relevant data is updated by NISRA on the portal (if auto-update is set to true), so you do not need to do this manually.

Example 2 – Bar chart showing proportion of P1 children who are obese by local government district and gender

In this example, we will use a Bar chart to look at the proportion of P1 children who are obese by local government district and this time broken down by gender. The following steps are required to set up the chart.

• Firstly use the reset button before creating another chart.
  
• Use the chart type option to choose a Bar chart.
  
• This will open the y-axis Label box. Name the y-axis - e.g. ‘Local Government District’
  
• Highlight all districts, including the aggregate measure (to give a reference number).
  

• Because we are including the gender breakdown, 2 series will need to be added to the chart (for females and males).
  
• Click on series at the bottom left of the table (highlighted above). This will now open the x-axis label box (below). Name the x-axis, e.g. P1 children who are obese (%).
  
• In this box you can also name series 1 (Females).
  
• To populate the chart with the appropriate statistics, highlight the relevant statistic, grouped academic year (2017/18-2019/20), gender (females).
  

• Click on the add series button (above) to add series 2.
  
• The following box will be displayed, containing the x-axis label and an option to name series 2 (Males).
  
• Again, highlight the relevant statistic, grouped academic year (2017/18-2019/20), gender (males).
  

• Use the options function to choose how to display your chart.
  
• Click on view to generate the chart.
  

This shows comparisons across the different local government districts by gender. You can see the figures by hovering your mouse over the bars.

You can also look at the results for each gender by clicking on the legends below. So, if you click on the ‘Females’ legend at the bottom of the table the results for females will be temporarily removed and only the results for males will display in the graph. Click on it again again to go back.

As before, you can download the chart as a PNG or by using the widget facility.

Mapping the data

For those datasets with geographical data a map may also be available. This will allow you to visualise and compare statistics across the different geographical areas. Maps are available for datasets containing geo-data such as local government district, assembly area, health and social care trust. The map icon beside the table code lets you know if a map is available and the map function within the table allows you to access and manipulate information in the map.

You can change what is shown on the map by using the options to the left. The map also shows the proportion of P1 children who are obese across different local government districts in the grouped years 2017/18-2019/20. You can highlight the numbers by hovering your mouse over a particular area on the map. In this example you can explore by different statistics - there are three others here, as well as by different time periods. You can also change the map to look at the different patterns for males or females.

The map can be downloaded as a PNG and as with charts and tables can also be downloaded using the widget facility. As before, the interactive snippet code can be embedded into your blogs, dashboards, reports etc., and will automatically update your map if the data is updated by NISRA on the portal.

The widget for the map has two extra features:

• Borders – when set to true, allows you to set boundary borders to the map.
  
• Colour Scale – allows you to choose a colour scale for the map.
  

Logging in to the system

As discussed in Chapter 1, logging in to the system enables additional functionality - a saved queries function, and a favourite tables function. The saved queries option allows you to save queries for returning tables, charts or maps, which can be easily rerun without having to rebuild them. The favourite tables option lets you save tables to a favourites list making them much easier to find and access.

You can log in to the system using any of the methods below, except the Management login which is for NISRA staff only.

If you choose to create an account using your email address, you can use the sign up option above and you will be asked to create a password. This password must meet the minimum requirements of:

• 8 characters
  
• 1 lowercase letter
  
• 1 uppercase letter
  
• 1 number
  
• 1 special character
  

Once completed, you will receive an email to verify your account. Once verified you are now free to sign in to the system. You can manage your account and delete it at any stage if you wish.

You can also sign in using your existing Facebook, X or Google account details.

Once logged in the saved queries option is now visible on the top banner beside data, while the favourite tables box can be found below the browse option on the left of the page.

When you are logged in and have opened a table you can add tables to your favourite tables box by clicking on the add to favourites button highlighted below. Once activated, the same button can also be used to remove the table from your favourites.

To activate the save query function, first select the data you require, and use the view option to run the results table. The save query function now becomes available to you – just below the view button.

Using this option will allow you to save your current selection to the saved queries box on your dashboard. From here you can revisit and quickly re-run your output without having to rebuild your selection. You can also download the associated csv file for your data, or print the output. When saving your query you will be given the option to save with fixed or fluid time. Fluid time allows you to maintain the same time pattern when new time points get published, for example always displaying the last five time points even if new ones are added to the data. See further explanation on fluid time in the section on widgets above.

The save query function is also available when you run a chart or a map.

Help and support

Along with this manual, there is a video demonstrating use of the tables, charts and maps in the Data Portal, and a separate document containing FAQs.

If you experience any issues with the application, please email info@nisra.gov.uk.