“You should look at this section if you have only a limited idea of what data visualisations and dashboards are and why they can be useful in educational assessment.”
The goal of most assessment is to collect data on student skills and knowledge that can help inform improvements in learning and teaching, overall leading to better student learning outcomes. Simply conducting assessment and reporting on results, however, is insufficient. Instead, the results need to be understood by educational stakeholders – parents, teachers, school principals, educational administrators and policy makers. Based on this understanding, stakeholders can then take actions to improve student learning.
Understanding educational assessment data can be very difficult, however. Many people lack the ability to read a chart or table and interpret what it means, Often, assessment data is reported in large documents that contain so much information that stakeholders are overwhelmed, and this means that they find it difficult to get a clear picture.
Data visualisations and dashboards are increasingly used to help overcome this challenge. They help to make the assessment data – and many other types of data – easier to understand and interpret, and therefore make it more likely that stakeholders will take action on the results that they are given.
Data visualisations are simply images that display data. They can include simple outputs such as charts, but also include maps, diagrams, infographics, geo-spatial analysis and time-series. Data visualisations are a form of visual art that is designed to make understanding data easier, more interesting and to help tell a story. They are often interactive, allowing users to manipulate the data to see different outputs. Some examples of data visualisations on different topics of interest to UNICEF can be seen at https://data.unicef.org/resources/resource-type/interactive-data-visualizations/
Dashboards are digital summaries of data that allow trends to be tracked over time and that highlights key findings. They have been used in the business sector for a long time but are increasingly being used in areas such as education and health. Dashboards enable the monitoring of key characteristics and provide relevant information to stakeholders which comes from a whole range of sources but is all shown in one place. An exemplar education dashboard is from California in the United States, with data for Monterey Country shown as an example https://www.caschooldashboard.org/reports/27102720000000/2018#college-career-card
To find out more about how data visualisations and dashboards can be used to explain assessment data, go to #Intermediate.
“You should look at this section if you already what data visualisations and dashboards are and would like to know how they can be useful in educational assessment . “
If educational assessment is to be optimised, it is essential that educational stakeholders are able to understand the data that they generate and also use this understanding to make changes to educational policy and practices. Data visualisations and dashboards are two ways that can support better understanding and interpretation, helping stimulate positive changes as a result.
There are many different ways of visualising data. A range of data dimensions can be shown in data visualisations including indicators (such as performance in grade 7 mathematics), time, regions or districts and sub-groups (such as girls and boys).This example from the OECD shows a number of different options in displaying the same data – in this case the performance of boys and girls in reading across countries that participated in PISA 2015
https://data.oecd.org/pisa/reading-performance-pisa.htm#indicator-chart. In this case a range of charts, diagrams, tables and maps can be seen and users can highlight certain countries.
Dashboards allow for a range of data from different sources – including assessment data – to be brought together into one place. An example is the UNICEF’s Education Pathways dashboard https://data.unicef.org/resources/education-pathway-analysis-dashboard/ which combines synthesises data from a range of different sources to compare student trajectories through education systems for a wide range of countries. To enhance usability, often elements such as graph axes and map legends can be customised, and individual elements can be clicked on to gain more information, as is the case in this example.
Dashboards can provide information about local issues as well as international ones. Dashboards can be used to provide individual students with results, or to summarise information about individual schools. For example, the MySchool website in Australia provides a range of detailed information about individual schools, including student enrolment, teacher numbers, finance, attendance and student results on the national Australian assessment NAPLAN, as shown in this example https://www.myschool.edu.au/school/45383/naplan Student dashboards are also increasingly common, as this example from a group of schools in the United Kingdom illustrates: https://support.arbor-education.com/hc/en-us/articles/205239921-Student-Portal-Dashboard
In addition to simply showing data from assessments and other statistical sources, dashboards can also bring together and summarise data on a range of different elements. For example, the World Bank is currently developing a Global Education Policy Dashboard, which will illustrate a range of factors that contribute to learning outcomes in primary education around the world. This will include data on student performance and school attendance as well as information on teaching, school management, infrastructure and education policies. Its goal is to provide policy makers with a holistic picture of the elements that drive high quality education.
To find out more about how data visualisations and dashboards can help in explaining and understanding assessment data, go to #Advanced.
” You should look at this section if you are already familiar with a range of different forms of assessment and how they can be used. “
Data visualisations and dashboards are increasingly used to help education stakeholders gain a better understanding of assessment data. This is in response to the fact that many stakeholders find that traditional ways of summarising assessment data – such as in lengthy documents – leads to confusion and results in a lack of action on the basis of results.
Many data visualisations and dashboards have been created by governments and multilateral organisations. Increasingly, however, individuals are creating their own data visualisations and dashboards, as there is increasingly software available to facilitate this. An example is a ‘Storytelling with Data’ challenge in which contributors were asked to visualise data about education among minority groups, with some incredible results: http://www.storytellingwithdata.com/blog/2018/2/9/education-visualized
It is important to understand what type of visualisations are relevant for use for what purposes. Some visualisations and the types of analysis they can illustrate include:
- Bar or column charts – For relative comparisons between subdivisions
- Thematic maps – For spatial (geographic) analysis
- Time series – For trends over time
- Column & mark charts – For time trends across subdivisions
- Multiple-indicator time series – For trends over time across indicators
- Column & line chart – For trends over time across 2 indicators with different scale
- Scatter plot – For the relationship between 2 indicators
- Bubble chart – For the relationships between 3 indicators
- Stacked column – For sums two or more indicators, for part-sum analysis
- Time-animated bubble chart – For trends over time, for 3 indicators
- Clustered bar / column chart, or column & line, or column & mark – For analysis of trends across two groups of subdivisions
- Dot plot – For analysis of trends across two groups of subdivisions
It is also important to remember that – where relevant – it is important to show error bars.
There are a large number of data visualisation tools available, some of which are free to use. They include Carto, FusionCharts, Google Motion Charts, Microsoft Power BI, Plot.Ly, QlikView, Sisense and Tableau.