Data as Text
The student's story
In order to support a student we have to be able to construct, and continually
reconstruct, the student's story. The student's story includes his/her
- current context
- likely future
Think of all the time and effort that goes into case
meetings, meetings with parents, support team meetings, conversations with
colleagues, phone calls to support professionals....
The core activity in all these meetings is constructing the student's story in
order to make sense of
- the student's need for support
- the current opportunities for providing support
- the resources available
- the likely best course of action given the constraints
Without text, stories cannot be constructed, reconstructed and shared.
- "If we cease sharing our stories, our knowledge becomes lost."
Perhaps this Algonquin Indian saying is also a statement about what happens when
students change classes and change schools.
Data, information, knowledge and action
Different parties need different 'data' inputs
- Policy makers need data
- Managers need information
- Workers need knowledge in order to construct their actions
Text is the Planner's main difference
There are more and more student databases being developed and used. So what,
if anything, is special about the Planner?
- Almost everything in the Planner is free text.
Unfortunately this is counter intuitive for most people. Most people think of data
as quantified measures, and codified representations, that can be directly
collated and analysed. Often critical data is reduced in value by being
abstracted into some code or measure
Reducing data to codes and measures too soon, often
- misrepresents the reality - mainly through sweeping generalisations and
- filters out vital elements
- reduces the capacity for drill-down
- reduces the links to context and related data
- reduces the capacity for making links unique to the student
Key questions about data
There are several question related to data that determined its value
- Does the data exist?
- What is the quality of the data?
- Is it accurate, timely,
- Is it "delightful to the users now and in the future" (a
definition of "quality")
- Do the potential users see the data?
- Do the potential users respond to the data?
- Do the users connect the data with related data?
- Do the users make sense of the data?
- Do the users act on the data?
[Nb. The "Planner" provides a shared work space that addresses each of these
The 'elephant in the room'
But the need for data varies from role to role:
- managers generally want figures in order to be able to distribute
resources and to report
- support workers also need figures but the also need text (snippets of the students' stories)
Is it possible to have both?
Text and tagging
The answer is a resounding "Yes!"
Consistent with the current leading edge of knowledge management, the
strategy is to tag text. Tagging text enables
- original data to be retained making it available in a more original form
- data sets to be collated and analysed on the basis of the tags, then
- explored together with the related text for deeper insights
- managers to have figures while workers have text
- a small number of fields to manage a wide range of data
- rapid development of a simple yet powerful database without know all the
- adjustment and refinement over time through flexibility and rigour
- real-time research using live data
Text as input and in-process data
The purpose of acquiring data is primarily in order to make sense of what is
- to put things in context - thus converting data into information, in
- to know what to do next
There are three kinds of data related to supporting students
- input data: what there is known about the student
- in-process data: data about
- what is being done
- what commitments exist
- who is contributing what
- patterns and trends
- progress so far
- new insights
- output data: what the immediate results are
- outcome data: what the long term implications are
The majority of data currently gathered is 'output data', e.g., testing
results, attendance and retention, parent satisfaction, assessment reports...
But such data often has surprisingly limited value in making sense of what is
happening and what to do next. Two students with the same literacy ratings may
require very different support.
Thus to make sense of what is happening and what needs to be provided requires more than output data - in order
to make sense of what is happening requires substantial 'input data' about a
whole range of factors: family, social, emotional, health, historical,
material ... Unlike most output data, the key input data is often unique to the
student. This is especial true for the majority of those students with
high and multiple needs for support. that is, the greater and more complex the
student's need for support the more likely it is that key data will only be
available as text.
Unlike measures and codes, text has the capacity to represent much of this
idiosyncratic data quite well. In this sense text can be very valuable as inputs
to planning and managing effective support.
Most processes involved in education are iterative requiring in-process data.
for the above reasons text is more suited capturing in-process data including
events, changed circumstances, trends and responses. Tagging such text enables
trends to be readily identified and reported.
Text as data - the advantages of text
There are multitude of reasons for considering the use of text in relation to
data based on
Principle: the value of data is that it may be used inform
conversations and decisions about what to do next
- Text is data
- Text is more likely to make sense and thus more "visible"
- Text can provide richer representations of experiences,
hopes and intentions
- Text is more flexible
- Disparate data can be more easily brought together
- Example: Assessment data from different system-wide (NAPLAN, SARIS, PIPS...)
and local testing and related observations an be entered and collated
- Note: When there is an online system-wide "Planner",
system-wide testing data will be uploaded by the system thus attaching the
data to the student ready for collation and use; and reducing the cost of
data entry, loss and oversight of data
- There is less need for standardisation
- A text-based system is more adaptable and resilient in times of change,
- A text based system is more stable over time
- Text can edited to improve its quality
- The detail is scalable to refine the focus and meet a need
- Text items can be tagged with single (or multiple tags) for
- easy reading
- analysis leading to the identification of trends and patterns at
- accountability: who is doing what
- statistical summaries at various levels - individual, class, year
group, section, school...
- Text can be used for tagging text, e.g., the hash tags in Twitter - the
Planner had "clever coding' using slash codes well before Twitter was
- Text enables trends, patterns, concerns, queries, recommendations... to
be summarised and captured as text and acted upon
- Text can accurately capture Observations and associated Responses
- Text can accurately capture Actions and their Outcomes
- Text is essential to the construction of any story, including the