Researching With Our Head In The Clouds? Using Tag Clouds To Analyze Free Text Patient Feedback
|
If you are the presenter of this abstract (or if you cite this abstract in a talk or on a poster), please show the QR code in your slide or poster (QR code contains this URL). |
Abstract
Background
Patient feedback in the form of free text responses are a potentially useful source of information about the patient experience of primary care. The free text responses may provide insights not picked up in structured questionnaires. Undertaking a manual review of the free text is usually tedious and time consuming. A rapid textual analysis tool may offer potential in quickly evaluating free text responses coming from patients.
Tag clouds, or word clouds, are compact visual displays of written information where size, colour and orientation of words represent underlying importance related to word frequency from the text source. They provide the researcher with a method of rapidly interrogating
large volumes of text, providing a concise visual representation to be scrutinised for messages or dominant discourses.
Objective:
Our aims were:
To look at the feasibility of using tag cloud technology to rapidly analyze large amounts of free text from patient feedback.
To determine whether the generated tag clouds provided insight on the key areas of concern voiced by patients in their free text feedback.
Methods
Free text responses from patients were collected in questionnaires from the patients seen in 13 GP practices in South West England. Out of 3462 questionnaires returned, 1415 contained answers to the free text feedback question. Tag cloud software was used to create tag clouds using the free text responses. Initially, a tag cloud was created from all the free text feedback from responding patients. Practices were then grouped according to their performance in the National GP Patient Survey. Tag clouds for the responses from the patients in each performance group (Upper and lower quartile in the GPPS) were generated, and compared to the initial tag cloud and to each other.
Results
The tag cloud of all patient responses showed that the five most occurring words were "doctor" "surgery" "appointment" "GP" and "practice". The remaining 45 words had mostly positive connotations, e.g. “helpful†and “excellent†or were neutral ("phone", "nurses"). A few words such as "problem" and “difficult†had negative connotations. When tag clouds from different categories of practices were compared, this pattern was repeated. The relative frequency of the words “difficult†and "problem" was slightly increased in the practices that scored in the
bottom quartile of the National GPPS Survey.
Conclusions:
Tag clouds may be used to quickly identify key concepts in large amounts of text. When used judiciously, they may highlight key areas of concern in patient feedback. They may also be used to support structured assessments of patient satisfaction.
Patient feedback in the form of free text responses are a potentially useful source of information about the patient experience of primary care. The free text responses may provide insights not picked up in structured questionnaires. Undertaking a manual review of the free text is usually tedious and time consuming. A rapid textual analysis tool may offer potential in quickly evaluating free text responses coming from patients.
Tag clouds, or word clouds, are compact visual displays of written information where size, colour and orientation of words represent underlying importance related to word frequency from the text source. They provide the researcher with a method of rapidly interrogating
large volumes of text, providing a concise visual representation to be scrutinised for messages or dominant discourses.
Objective:
Our aims were:
To look at the feasibility of using tag cloud technology to rapidly analyze large amounts of free text from patient feedback.
To determine whether the generated tag clouds provided insight on the key areas of concern voiced by patients in their free text feedback.
Methods
Free text responses from patients were collected in questionnaires from the patients seen in 13 GP practices in South West England. Out of 3462 questionnaires returned, 1415 contained answers to the free text feedback question. Tag cloud software was used to create tag clouds using the free text responses. Initially, a tag cloud was created from all the free text feedback from responding patients. Practices were then grouped according to their performance in the National GP Patient Survey. Tag clouds for the responses from the patients in each performance group (Upper and lower quartile in the GPPS) were generated, and compared to the initial tag cloud and to each other.
Results
The tag cloud of all patient responses showed that the five most occurring words were "doctor" "surgery" "appointment" "GP" and "practice". The remaining 45 words had mostly positive connotations, e.g. “helpful†and “excellent†or were neutral ("phone", "nurses"). A few words such as "problem" and “difficult†had negative connotations. When tag clouds from different categories of practices were compared, this pattern was repeated. The relative frequency of the words “difficult†and "problem" was slightly increased in the practices that scored in the
bottom quartile of the National GPPS Survey.
Conclusions:
Tag clouds may be used to quickly identify key concepts in large amounts of text. When used judiciously, they may highlight key areas of concern in patient feedback. They may also be used to support structured assessments of patient satisfaction.
Medicine 2.0® is happy to support and promote other conferences and workshops in this area. Contact us to produce, disseminate and promote your conference or workshop under this label and in this event series. In addition, we are always looking for hosts of future World Congresses. Medicine 2.0® is a registered trademark of JMIR Publications Inc., the leading academic ehealth publisher.
This work is licensed under a Creative Commons Attribution 3.0 License.