Reading between the lines: Dispelling the fears surrounding qualitative feedback
Posted on
14 January 2008
by
Dan O'Shea
Why do so few businesses refuse to place their faith in text-based analysis when relaying staff feedback? Simple - they don’t trust the results: what if anonymity is compromised? What if the results are imbalanced? How do you get round the subjectivity of the analyst? These are all, of course, highly legitimate concerns but can be swiftly allayed by recognising the rigorous methodology underpinning the qualitative process.
1: Respecting the text
The first rule for the analyst charged with converting raw data into qualitative feedback is to remain unswervingly loyal to the source text. He/she must never read between the lines by making inferences or unravelling the ‘true’ meaning of remarks: for the analyst, there is nothing beyond the text. But loyalty, of course, does not mean slavishly relaying every single recorded comment and insight. Rather, the analyst needs to meticulously study the raw data and begin synthesising the material.
2: Synthesising the text
Experience shows that some reviewers often repeat their opinions across different parts of the competency framework so the analyst needs to slot the remark into the most appropriate cluster and then cut the repetition elsewhere. The text is thus contracted and more tightly arranged without losing any key details. Similarly, if multiple reviewers express the same viewpoint under a particular behavioural indicator, the analyst is required to tie these remarks together and convey it as succinctly as possible. But the most critical element of synthesising the raw data is establishing the balance of opinion. Take this example: if one reviewer is highly critical of the reviewee and supports this stance both specifically and evidentially yet four other reviewers are far more complimentary without substantiating their praise, the analysis needs to be constructed very carefully: the negative comments represent a minority view and should therefore not dominate the analysis. They should of course be included but must be acknowledged as anomalous.
3: Structuring the text
It is highly recommended that the qualitative analysis mirrors the competency framework in terms of its flow and structure as the resultant report will be far more accessible, familiar and digestible to the recipient. This method also means the analyst is much less likely to inadvertently omit important information.
4: Anonymising the text
Once the content and structure have been determined, the analyst needs to ensure anonymity is safeguarded at all times. Quotations, for instance, must be handled with extreme caution: does the raw data contain highly idiosyncratic vocabulary which might betray identity? If so, paraphrasing is paramount: in this way, the message is delivered by the analyst on behalf of the reviewer without losing any of the resonance or impact of the original. This is not to say that quotes can’t be used – they are sometimes extremely useful. The key criterion is this: the analyst must respect the sensitivity of the raw data at all times.
5: Translating the text
Contemporary translation theorists (Venuti, Bassnett, Lefevere) concur that the translated text is inextricably bound to the original: it does not exist without it and it is this notion of interdependence which must be borne in mind every time the analyst begins the process of converting raw data into qualitative feedback.