How to extract anecdotes for sensemaking
From Regional Knowledge Resource Kit
Return to Stage 3 - anecdote circles
Return to "Developing a Regional Knowledge Strategy" Quick Links Page
Contents |
Background
The purpose of this guide is to provide practical information and instructions to assist in extracting and organising anecdotes from the transcribed outputs of anecdote circles.
The anecdote extraction/organisation process involves:
- reviewing transcribed material & identifying anecdotes from the volume of information contained in that material
- numbering those anecdotes to be used in the sensemaking workshop
- recording indexed anecdotes on an Excel spreadsheet (referenced below under 'Materials' section)
Facilitation level
Anecdote extraction requires a basic level of skill and can be done by anyone who can identify an anecdote. Here are two articles that describe how to identify stories and anecdotes:
Objectives
The aim of anecdote extraction, as the name suggests is to extract the anecdotes from the transcriptions so they can be used in Stage 4 - sensemaking & action planning.
Materials
You will need the following materials for anecdote extraction:
- the word processor file with the transcription document open
- a spreadsheet, or you may wish to use the following Excel template
Excel template to record extracted anecdotes 
Time
Plan to set aside the same time it took to collect the stories as time to extract the anecdotes. For example, if you ran two anecdote circles and they both ran for 90 minutes, then set aside 3 hours for anecdote extraction.
Directions
Identification of Anecdotes
As a reminder, anecdotes are -
- Naturally occurring stories
- Typically told about an event (either actual or fictional)
- Told from someone’s personal perspective
The identification of anecdotes is a relatively subjective process. Humans are natural storytellers, and are generally able to recognise a “story” with little difficulty. However, with a large volume of transcribed oral material, it can sometimes be difficult to differentiate “stories” from other transcribed text such as statements, opinions or general comments.
The following examples may help to differentiate between a personal experience that is not an anecdote and one that is.
'When a personal experience is Not an anecdote'
The following text is not an anecdote. It is a statement and opinion:
- "So, I don't think I've ever had a problem finding information. What you do is you find a mob of it. And then you've got a mob of information, you've got to go, "Well which of this", you know, even on something such as this, you're going to find a lot of information. Which is the best? Which is the most up-to-date? Which has been most useful to somebody? So it's kind of like, do you read all of that before you find what was the most up-to-date."
'An example of an anecdote'
Here’s an example of an anecdote. It's the retelling of a personal experience.
- "The best way I found to find things out in this organisation is talking to other staff rather than trying to find it out themselves. I got so frustrated at times, including that little search the other day because I didn't know, I wasn't completely sure what I was doing, and there's always that nagging voice in your mind that says, "Don't ask that question. People will know that you're stupid if you ask that question. If you don't ask it, they won't know. They might think you are, but they won't know. But if you ask the question, they'll definitely know"; that sort of nagging little doubt. So I agree that I found I've got better results every time I've gone and asked someone individually"
In selecting the anecdotes to extract from your anecdote circles it might be ideal to work in pairs to:
- Read through each transcript together
- Identify & mark/highlight potential anecdotes as they appear in the material
- If you are using the Excel template Excel template to record extracted anecdotes to collect anecdotes you will notice there is an 'impact rating column' in the spreadsheet. This column can be used to give a ranking to your anecdotes based on the impact of the anecdote on the reader. Ranking can be made on a scale of 1 - 5 or you can employ your own ranking e.g. high/medium/low. This rating could be helpful where you have large numbers of anecdotes collected. Remember that a maximum of 40 - 50 anecdotes would be used in the sensemaking workshop. You can use reduce the numbers of anecdotes used by choosing those anecdotes with higher impact rankings.
Where there is uncertainty as to whether a section of narrative constitutes an anecdote the pair should discuss the section and agree its inclusion/exclusion
What happens to the anecdotes you've extracted?
Once you have extracted the anecdotes they can be recorded in the Excel spreadsheet template file. The Anecdotes can then we reviewed and reduced in number if necessary, or prioritised into those with major impact or lesser significance of impact.
As a rule of thumb include around 40-50 anecdotes in your sensemaking workshop.
In preparation for your Sensemaking Workshop:
- print out each of your selected anecdotes on a single A4 page.
- Use a large font type such as Ariel 36 or Ariel 40.
The aim is to produce the anecdote on one A4 page with as large a font as possible to make the anecdote easy to read when displayed on a wall in the Sensemaking workshop.
Tips
Inaudible / missing words
Some transcribed anecdotes may contain “inaudible” words or phrases, indicating that the speaker’s words were unclear or obscured by other speakers or extraneous noise. In these cases -
- Where the context means that the missing words or phrases are obvious, the transcription may be edited with the missing words added to complete the anecdote. You must have consensus among your group of editors before making any changes.
- Where the missing words or phrases are clearly incidental (ie. where it is obvious they do not add to or detract from the anecdote), the transcription may be edited to remove the “inaudible” reference.
- Where the missing words or phrases are not obvious or incidental to the anecdote, but the anecdote still makes sense, it should be extracted and recorded with the “inaudible” reference left intact.
- Where the missing words or phrases render the anecdote meaningless, it should not be extracted or recorded.
Interrupted or incomplete anecdotes
In some cases, anecdotes might not flow contiguously in the transcription – that is, there might be cases where an anecdote is interrupted by other comments or other speakers. In these cases –
- Reading through the transcription in detail may identify further sections of text where an anecdote is continued or completed – these sections should be combined, and any “interrupting” or extraneous text should be removed, to form one complete anecdote for indexing and recording
- Where an anecdote is incomplete and no further reference is made to it –
- provided an anecdote can still be discerned (ie. the section of narrative still constitutes a coherent “story” in itself within the context of a broader incomplete anecdote) this may be extracted and indexed;
- where the anecdote is clearly incomplete and this renders it meaningless, it should not extracted or indexed.
Narrative “fillers”, hesitations and pauses
When oral material is transcribed, it is full of the normal “extraneous” features of speech – it contains hesitations, pauses, and a number of narrative “fillers” (eg. “and, yeah….so I….like, I said…”).
Where such features are embedded in an anecdote, the anecdote should be extracted and indexed with these features intact.
However you should instruct your transcriber to ignore "ums" and "ahs".
Return to Stage 3 - anecdote circles
Return to "Developing a Regional Knowledge Strategy" Quick Links Page
