SurveyVitals has long helped our clients gain a better understanding of patient comments with semantic keyword analysis. With the release of our new Report Builder, we’re taking comment analysis to the next level with sentiment reporting.
A recent review of SurveyVitals data showed that 36% of patients leave at least one comment when responding to our digital surveys. In a March 2019 patient experience study conducted by NEJM Catalyst Insights Council, 97% of clinicians agreed that listening to a patient’s voice helps improve care. Patient comments give context to Likert-scale scores, helping to bring the data to life.
Comments are vital to understanding your patients’ needs, but reading through hundreds–or even thousands–of them can be time-consuming. SurveyVitals is excited to introduce our new sentiment analysis tool to help you gain deeper, faster insights.
How does it work?
Our sentiment analysis uses natural language processing (NLP) and machine learning algorithm to provide tonal or sentiment insights for text comments. The sentiment process, provided by Amazon Web Services, measures how positive, neutral, mixed, or negative a full comment is. Unlike semantic keyword analysis, sentiment measures the overall tone of the comment–for example, it looks for negation keywords such as ‘not’ in conjunction with emotional keywords for a better understanding (e.g., ‘The doctor made me not feel so afraid’).
Identifying the most critical feedback has never been easier. The ability to sort by positive, negative, mixed, and neutral sentiment scores allows for quick identification of the best and worst comments. Sorting by these scores or searching keywords may also be beneficial in reviewing comments for trends and prevailing concerns.
Sentiment analysis saves you time, drives awareness, and helps you to better understand your data and implement change.
Have questions or feedback about sentiment analysis? Reach out via the blue chat icon below or contact a member of your support team today.