DATA: THE PERFECT RECIPE FOR SCHOOL IMPROVEMENT

“It is a capital mistake to theorise before one has data.”

Sherlock Holmes in “A Scandal in Bohemia” by Arthur Conan Doyle

Every #school, every stakeholder wants to improve. However when it comes to the ‘How’, confusion, repetition, ambiguity, frustration – all run supreme. Two of the ingredients which ensure the right pathway are #collaboration and #reflection. Collaboration amongst teachers, senior management, and parents helps to pool in collective knowledge, skillsets, and ideas to not only carve a pathway for overall success but helps in raising the potential of every student. Reflection, a companion to collaboration, helps the stakeholders think about the information in front of them and adjust their actions accordingly. On the improvement journey, reflection is necessary not only for staying on the improvement path but also for discovering the best path. Successful reflection depends on thought-provoking information and time for individual and team study. Put together, reflective collaboration is a powerful process that can steer any institution to the pinnacle of success. And data, present in the system will be the messiah in the process. The data in schools provide important clues about the current actions and students’ performance. But how does one embark on this reflective collaboration process?

Effective school improvement processes are cyclical and continuous, with no clear beginning or end. An early version of the plan-do-study-act cycle for school improvement originally was developed by Dr. Walter Shewhart (1939), and it provided a foundation for much of the work by corporate management expert W. Edwards Deming (see Rinehart, 1993). This cycle contains four major activities:

1. Plan: Develop a plan for improvement.

2. Do: Implement the plan.

3. Study: Evaluate the impact according to specific criteria.

4. Act: Adjust strategies to better meet criteria.

For schools to implement effective data cycles, into their everyday operations and their general approach to learning, they must have support at two levels:

At the Organizational level – where the Culture and Infrastructure supports and promotes the use of data.

At the Practical level – where the Practices that teachers, administrators, and other school staff engage in also supports and promotes the use of data.

The educational system regularly accumulates huge amounts of data; so the systematic processing of the same is of paramount importance. Improving student results through personalised support, reducing drop-outs especially in marginalised communities, customising or evaluating impact of educational programs, all can be provided valuable insights through a data-driven approach. The recent #PISA 2022 published results throw some interesting insights related to mathematics anxiety and also the effect of COVID-19 in #reading, #mathematics, and #science performance of students across 80 plus nations. Timandra Harkness in her book, Big Data: Does Size Matter? explains the Big Data characteristics by referring to ‘BIG’ as obvious big, relating to the concept of ‘volume’ associated with the definition of big data by other analysts. She however creates her own acronym for DATA and how it makes data new and distinctive:

D stands for Dimensions (related to variety described by some others)

A stands for Automatic (related to velocity described by some others)

T stands for Timely

A stands for Artificial Intelligence

While undertaking a data-driven approach, it is important to acknowledge some of the basic steps:

1. Data Collection: In context of schools, teachers are engaged in collecting data throughout the academic year. They collect classroom (student) data generally through various types of #assessment activities. After all good data can provide important information to #teachers and #students and go a long way toward improving #teaching and #learning. Assessment is a continuous process, embedded in the teacher’s daily instruction which help to select and implement plans to monitor and check on progress and change strategies for a desired behaviour.

2. Data Organisation: When data are initially obtained from questionnaires, interviews, experiments, administrative sources, or other methods, a statistical analyst encounters a list or array of values of variable(s). If the data are quantitative, these are lists of numbers; if the data are qualitative, these are lists of the words or phrases associated with each response. Organising data enables to quickly spot some of the characteristics of the dataset. It is important to plan in advance how best to organise data. Using a logical structure and ensuring that all collaborators understand the structure before starting to analyse and interpret the dataset is one of the initial crucial steps towards data handling.

3. Data Analysis: This is where the real fun happens. Data analysis is a pivotal step where raw data is scrutinised, interpreted, and transformed into meaningful insights. Depending on the core purpose, analysis should be done and data should be looked using a certain lens. This process also involves cleaning of the data to remove inconsistencies or errors, exploring data through visualization tools, and applying analytical methods to extract relevant information. Critical thinking is essential to ensure accurate interpretations and avoid biased conclusions. Ultimately, data analysis drives decision-making, informs strategies, and facilitates problem-solving in various fields, from business to healthcare and beyond.

4. Data Visualisation: Data visualisations are a vital component of data analysis, as they have the capability of summarising large amounts of data efficiently in a graphical format. The visual displays of information communicate complex data relationships and data-driven insights in a way that is easy to understand. Dashboards are effective data visualisation tools for tracking and visualising data from multiple data sources, providing visibility into the effects of specific behaviours by a team or an adjacent one on performance. To start thinking visually, consider the nature and purpose of your visualisation:

a. Is the information conceptual (based on idea) or data-driven (based on statistics)?

b. Am I declaring something or exploring something?

If the first question identifies what you have, the second elicits what you are doing: either communicating information (declarative) or trying to figure something out (exploratory).

5. Organising Data-Driven Dialogue: After all the visualisations, it is important that a dialogue is initiated between the stakeholders. A dialogue can be among any number of people, not just two. Some of the key components of a data-driven dialogue include:

a. Making predictions

b. Recreating the data visually

c. Analysing the data for trends and patterns

d. Making inferences and drawing conclusions

My colleague Munish Gupta and I recently conducted a workshop with 50+ brilliant teachers from Seth M.R. Jaipuria School, Gomtinagar-Lucknow; covering all of the above and more. With hands-on tasks to learn basic analysis of school-level data, the workshop was interactive, fun, and a lot of learning. The teachers, as part of their action plan, and next steps will be trying their hands on actual data-set and contribute towards meaningful interventions, living up to one of the school’s principles – Student First.

Do reach out to me if:

a. you would like to know more about how we are using data into all our work,

b. you would want to conduct a similar workshop with your teachers.

Quoting a viral statement from #BETT2023 made by Kanak Gupta, “In God we Trust; for everything else we need Data!”