When RapidPro receives a message, it performs a test on the response to decide which question to send next, e.g. if response contains "yes" it will categorise the response as "Yes" and proceed to the next question. The "value" is the part of the message which matched the rule, and the "text" is the raw message that was received. For example, if "yes I am" is sent then the value is "yes" and the text is "yes I am".
RapidPro exports both text and value via Excel. However its runs API only provides the value.
The text may be recovered from the messages API. However, messages are not uniquely identifiable - if the modified times are not too dissimilar then filtering messages on the contact and sent time is probably the only way to get this.
We need to know if having the text in addition to the value is important and how to address this problem if it is.
When RapidPro receives a message, it performs a test on the response to decide which question to send next, e.g. if response contains "yes" it will categorise the response as "Yes" and proceed to the next question. The "value" is the part of the message which matched the rule, and the "text" is the raw message that was received. For example, if "yes I am" is sent then the value is "yes" and the text is "yes I am".
RapidPro exports both text and value via Excel. However its runs API only provides the value.
The text may be recovered from the messages API. However, messages are not uniquely identifiable - if the modified times are not too dissimilar then filtering messages on the contact and sent time is probably the only way to get this.
We need to know if having the text in addition to the value is important and how to address this problem if it is.