Surveys can be enabled at any time in development mode. (This feature gets locked down when moving to production. Please email redcap@pcom.edu if your project needs to be moved back to development mode.) Here are instructions for enabling surveys in your development project:
The Manage Survey Participants link will appear in the Data Collection section of the menu ONLY after at least one instrument is enabled as a survey.
A public link is only possible when the first instrument is a survey. Only that survey can have a public link. All later instruments are considered a continuation of the earlier one(s). Later surveys capture different data, but are completed by the same people. So the later instruments can only ever be distributed via Participant List. If different respondents will complete the other surveys, then a separate project should probably be used for each group of respondents.
Most projects should be exported to a package that is designed to run statistics, such as SAS, SPSS, STATA, or R. However, some calculations are fine to leave in REDCap, such as simple calculations for branching logic.
After a project has been moved to production and data has begun to be gathered, any change to the survey can affect the data that has already been gathered, rendering it useless. An example of this is changing a true or false question to multiple choice.
Ultimate liability for any changes made to the project after it has been moved to production rests with the researcher and their team.
We recommend that you don't use the yes/no or the true/false fields when creating a survey because the choices are hardcoded in the metadata. It's fairly common for the researcher to decide partway through the project that the two answer choices are not sufficient. But once you have started collecting data, to change the yes/no or true/false question to a multiple choice would result in corrupted data. All previously collected data would be lost and the researcher would have to start from scratch. Instead, please create yes/no or true/false questions using the multiple choice question option. This prevents any accidental code overwriting even if the researcher changes the question.
No, it is not possible to create an anonymous survey while still gathering information for incentives (such as collecting email addressing for a prize lottery). Instead, the best practice is to create two surveys. The first survey can be done anonymously while the second survey collects information. The first survey should route straight to the second survey. However, the two surveys cannot be linked, which keeps the first survey's data anonymous. See tutorial 7: Surveys for more information.
While REDCap does have a QR code feature, we recommend that you don't use it as it requires participants to download a special app. Instead, we recommend using a survey access code, which gives a simplified URL and an access code that the participant can use to access the survey.
In development mode, the Other Functionality tab has a button to erase all data. This is useful when you are iteratively testing your project and want to practice your data entry several times, starting with an empty project each time.
In production mode, records must be deleted individually. Because of this limitation, you may want to consider copying your project and using the copied version instead. That copied version would be in development mode and have no records/data. (It would also be totally separate to the original project, ensuring you still had the original data and could quickly reference it in the future.)
@NOW captures the current time while @TODAY captures the current date. Both can be set up to capture both time and date. We recommend as a best practice that you use @NOW.
Please see the attachment below for more information on other Action Tags
We recommend that you use PCOM colors for your HTML coding.
If you do not want to use PCOM colors for HTML coding, you can find HTML color generators here:
As a general rule, we recommend that you download all data from REDCap as raw data because raw data is used for project analysis. When cleaning data or reimporting it into REDcap, it must be in the form of raw data.
Reminder: raw data will give you variable names such as "why_pcom" instead of "Why did you choose to attend PCOM?", and uses coded values on items such as multiple-choice questions. Labels give the full question such as "What is your program?" and for multiple-choice questions, the label for the answer choice such as "Osteopathic Medicine," "Psychology", etc.
We recommend that the record ID field not be used as an identifier. However, you do choose to make the record ID field the identifier, you can hash it so that the record ID is an unrecognizable value.
The Data Quality tool is a great way to help ensure the integrity of your data. It comes with preprogrammed different rules that you can use to check your data. We recommend that you run all these rules - except A (which just finds missing values).
One of the most useful rules is H, which looks for incorrect values in calculated fields. The calculations in calculated fields aren't sent to the server until you hit save. If you were to update a calculation and it doesn't update the values for your entire project or you import incorrect data into a calculated field, then you'll have errors that you may not know about. You can use rule H to check your project, and it will let you know if there are any discrepancies.
Run all these rules before exporting data.
The Library has no access to or familiarity with any statistical packages. However, below are some tips that may be useful to you.
Statistical software might have trouble opening your export file for any number of reasons. Here are some steps to help you troubleshoot.
If you still have trouble, try these additional troubleshooting techniques.
We typically suggest removing all unvalidated text/notes fields from exports used in statistical packages because those data types are usually not useful for analysis and can clog the statistical package processing (including throwing off the column dividers). It is also possible that the variable names themselves are incompatible with your statistical package. For example, SPSS may not be able to process variable names which include negative values.
Please consider submitting your idea for any new feature via the feature suggestion survey link, found in the Help & Information section of any project menu. Suggestions made there are ranked for feasibility and utility by the REDCap consortium software development team. That development team considers the input of ~538,000 users across all ~2,400 consortium partner sites worldwide. We of course cannot guarantee that all suggestions will result in changes or new features, but survey submissions do all feed into the consortium’s exploration and development prioritization list.
