Soft Requirement for questions left blank

[carly.sc...@gmail.com]

Hi everyone!

I am wondering that if I happen to leave a question blank on a form while assessing a participant, or if a participant leaves a question blank while filling out a survey, if there is a way to notify the person that a question has been left blank, and are they sure they want to move on? I can't use the required question feature because although it does remind you that a question is blank, it will not let you continue on until you answer that question. Our participants must be allowed to skip over a question and leave it blank if they want to, but I just want a little box to pop up to let them know one is left blank, asking them if they are sure they want to continue the survey without answering that question. Is this possible in Redcap? Thanks!!

Carly

[Gerber, Alan H]

Hi Carly,
In my opinion that is one of REDCap's biggest faults. I would be curious to hear if anyone has found a solution. The best that I've been able to do is to use a descriptive item that says "warning: you left the previous question blank" with branching logic. However, it stays there if they leave it blank. I would much prefer a pop-up item of some sort.

Best,

Alan Gerber,  M.A.
Clinical Research Assistant
Rhode Island Consortium for Autism Research and Treatment
Developmental Disorders Genetics Research Program
Bradley Hospital
Brown University
Phone: (401) 432-1624
Fax:      (401) 432-1607
www.AutismRI.org

--
You received this message because you are subscribed to the Google Groups "Lifespan Redcap Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lifespan-redcap-...@googlegroups.com.
Visit this group at http://groups.google.com/group/lifespan-redcap-users.
For more options, visit https://groups.google.com/d/optout.

[Baird, Grayson]

This is a problem. Steve Reinert (of Lifespan) actually sent this (below) to the main google group a few months ago. Perhaps we should make an official suggestion or request?
Grayson

From: sreinert [mailto:srei...@lifespan.org]
Sent: Sunday, June 08, 2014 4:07 PM
To: project...@googlegroups.com
Cc: Baird, Grayson
Subject: Suggestion regarding Survey Tool "required" items

I have a suggestion which could help me and perhaps other researchers. My suggestion concerns the feature in the survey tool which allows one to designate a given item as “required”. In its present form, if an item is required, then the subject cannot skip the question. Although this makes intuitive sense, ethically we cannot force a participant to answer any question. Though we could provide the option “choose not to answer”, we are concerned that participants will be more likely to choose not to answer because that option is available. Would it be possible to have a third option, where subjects are not forced into responding, but instead are prompted with a reminder that they left a given item blank, but are able to continue regardless?

Would other readers benefit from such a feature?

Has this been considered by REDCap developers?

Thank you.

Responses:

1. With a multi-page survey, you can put the warnings at the top of the next page, with a note to use the previous page button to go back and fill in the missing data.

2. Yeah, that's a clever idea. Will keep it in mind for future projects.
I'm thinking of one particular study where surveys consisted of standardized instruments scored after export. In order to compute a valid score all questions are "required" as opposed to one very important question.

3. As Cinly mentioned we have used branching logic in the past to remind participants if they left a question blank. But this setup takes a very long time (for long surveys) and is awkward because the "reminder" descriptive fields for each question are already on the page when you open the survey and disappear as the person answers questions. This is opposed to the reminders appearing only after a person tries to submit. Sometimes there is a truly required question that cannot be skipped (e.g. Consent), but this is not usually the case. I often find people using the "required" field on a survey because they don't want anyone to skip questions and I have to tell them that this is not compliant with ethical principles. But sometimes I never see the project and it happens anyways. I think if users had an alternative setup option like the one you suggest, then they would use it.
Yael

4. Yael,

Try using a composite of fields in your branching logic so that if they start completing common fields after the required field the warning would then appear, but be hidden otherwise.
-Kevan

[Gerber, Alan H]

I think an official suggestion is a great idea. I've put in a few idea to their online "suggestion box" but I'm not sure how much they pay attention to that. Do we have any direct line to them?

[Reinert, Steven]

I would email our suggestion directly to Rob Taylor, their lead developer, and ask him to suggest a course of action. -Steve

[Baird, Grayson]

Great-- unless someone else wants to, I will craft an email and send it here first for review before sending it to Rob. g

[Carly Schwartzman]

That would be great thanks so much for everyone's input!

--

Carly M. Schwartzman
Research Assistant, Butler Hospital

345 Blackstone Blvd. Providence, RI 02906
Phone: (401) 455-6243

[Gerber, Alan H]

All yours Grayson!

 

Alan Gerber,  M.A.

Clinical Research Assistant

Rhode Island Consortium for Autism Research and Treatment

Developmental Disorders Genetics Research Program

Bradley Hospital

Brown University

Phone: (401) 432-1624

Fax:      (401) 432-1607

www.AutismRI.org

 

[Machan, Jason T.]

Hey everyone....

I have some a strong opinions on this discussion I will try to make succinctly. Just worried about overly broad swipes.

1) Feature: I think a per-question option (e.g. pop-up) to double check that a participant wants to skip a question (in addition to simply "not required" vs. "required") is a great idea
a) I'm surprised it's not in there already
b) However, it's not without its own concerns...does this, itself, run the risk of becoming belligerent or coercive?

2) Ethics: I think that we'd be overstating to say that any time a question is "required" that we are being unethical.
a) A required field does not restrict a participant from deciding not to complete the survey...that is, just as if they were having some experimental procedure, they could stop any time they want
b) On the other hand, I'm sure there are examples in which a required field, when combined with other design features (e.g. a bunch of money at the end) may be coercive and thereby be unethical

So, making a field required may not be in your interest, if it means participants may completely exit the study...but that's a different from being unethical.

This is where the IRB comes in. If they, when looking at the study in total, decide a question would be unethical to treat as required, then they could and should require that the question is NOT required, or that there some mitigation of the ethical problem (e.g. give the money even if don't finish).

An couple examples:

We would like to know if a particular medication can help participants maintain weight loss. It's critically important that we start the medication exactly when patients reach a target weight loss, and so we approach and consent patients as they enter into a weight loss program. We inform them that, if they reach their target weight loss, they will be randomized to receive either our medication or a placebo. In this case, hitting the target weight becomes an inclusion criterion for randomization. Since not all participants may hit their target...or may decide it's not worth the effort for their own reasons, not all will be randomized. I wouldn't say this was an ethical dilemma.

If that doesn't satisfy, perhaps a medication would be unsafe to start until a recreational drug with which it interacts were cleared from a patient's system. Patients are enrolled and given a particular length of time within which they attempt to clear that medication. Some may fail to clear the recreational drug.






Jason T. Machan, Ph.D.
Director, Lifespan Biostatistics Core,
Lifespan Hospital System
Research Scientist, Biostatistics, Research
Rhode Island Hospital
Associate Professor, Departments of Orthopaedics and Surgery
The Warren Alpert Medical School, Brown University
Director Biostatistics Externship, Adjunct Assistant Professor, Department of Psychology
University of Rhode Island

Grads Dorm 206a (click for map)
593 Eddy Street
Providence, RI, 02903
office: 401-444-1493
cell: 401-639-3942
fax: 401-444-8271
Click here to request biostatistics assistance
Click here for special research technology support

CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential, proprietary, and/or privileged information protected by law. If you are not the intended recipient, you may not use, copy, or distribute this e-mail message or its attachments. If you believe you have received this e-mail message in error, please contact the sender by reply e-mail and destroy all copies of the original message.

-----Original Message-----
From: lifespan-r...@googlegroups.com [mailto:lifespan-r...@googlegroups.com] On Behalf Of Baird, Grayson
Sent: Wednesday, August 27, 2014 10:12 AM

To: Gerber, Alan H; carly.sc...@gmail.com; lifespan-r...@googlegroups.com
Subject: RE: Soft Requirement for questions left blank

This is a problem. Steve Reinert (of Lifespan) actually sent this (below) to the main google group a few months ago. Perhaps we should make an official suggestion or request?
Grayson

From: sreinert [mailto:srei...@lifespan.org]
Sent: Sunday, June 08, 2014 4:07 PM
To: project...@googlegroups.com
Cc: Baird, Grayson
Subject: Suggestion regarding Survey Tool "required" items

I have a suggestion which could help me and perhaps other researchers. My suggestion concerns the feature in the survey tool which allows one to designate a given item as ?required?. In its present form, if an item is required, then the subject cannot skip the question. Although this makes intuitive sense, ethically we cannot force a participant to answer any question. Though we could provide the option ?choose not to answer?, we are concerned that participants will be more likely to choose not to answer because that option is available. Would it be possible to have a third option, where subjects are not forced into responding, but instead are prompted with a reminder that they left a given item blank, but are able to continue regardless?

Would other readers benefit from such a feature?

Has this been considered by REDCap developers?

Thank you.

Responses:

1. With a multi-page survey, you can put the warnings at the top of the next page, with a note to use the previous page button to go back and fill in the missing data.

2. Yeah, that's a clever idea. Will keep it in mind for future projects.
I'm thinking of one particular study where surveys consisted of standardized instruments scored after export. In order to compute a valid score all questions are "required" as opposed to one very important question.

3. As Cinly mentioned we have used branching logic in the past to remind participants if they left a question blank. But this setup takes a very long time (for long surveys) and is awkward because the "reminder" descriptive fields for each question are already on the page when you open the survey and disappear as the person answers questions. This is opposed to the reminders appearing only after a person tries to submit. Sometimes there is a truly required question that cannot be skipped (e.g. Consent), but this is not usually the case. I often find people using the "required" field on a survey because they don't want anyone to skip questions and I have to tell them that this is not compliant with ethical principles. But sometimes I never see the project and it happens anyways. I think if users had an alternative setup option like the one you suggest, then they would use it.
Yael

4. Yael,

Try using a composite of fields in your branching logic so that if they start completing common fields after the required field the warning would then appear, but be hidden otherwise.
-Kevan




-----Original Message-----
From: lifespan-r...@googlegroups.com [mailto:lifespan-r...@googlegroups.com] On Behalf Of Gerber, Alan H
Sent: Wednesday, August 27, 2014 10:02 AM
To: 'carly.sc...@gmail.com'; lifespan-r...@googlegroups.com
Subject: RE: Soft Requirement for questions left blank

Hi Carly,
In my opinion that is one of REDCap's biggest faults. I would be curious to hear if anyone has found a solution. The best that I've been able to do is to use a descriptive item that says "warning: you left the previous question blank" with branching logic. However, it stays there if they leave it blank. I would much prefer a pop-up item of some sort.

Best,

Alan Gerber, ?M.A.

Clinical Research Assistant
Rhode Island Consortium for Autism Research and Treatment Developmental Disorders Genetics Research Program Bradley Hospital Brown University
Phone: (401) 432-1624

Fax: ? ? ?(401) 432-1607

[Carly Schwartzman]

Hi Dr. Machan,

The feature I was really hoping for would be that at the end of the current page the participant is completing, at the time they press the "next" button to move on to the next page, a box would pop up saying "You have left question #5 unanswered, would you still wish to continue?", and then they could continue on if they intentionally left that question blank.  However, the "required" feature already on redcap does not let the participant move on to the next page of the survey unless every single required question is answered.  So, if they wanted to still complete the rest of the survey, but just leave that one question blank, they would not be able to do so.  Thanks!

Carly

[Machan, Jason T.]

Annoying absence of a feature that makes a lot of sense to have!

I think Vanderbilt should get it…built.

 

 

Jason T. Machan, Ph.D.

Director, Lifespan Biostatistics Core,

     Lifespan Hospital System

Research Scientist, Biostatistics, Research

     Rhode Island Hospital

Associate Professor, Departments of Orthopaedics and Surgery

     The Warren Alpert Medical School, Brown University

Director Biostatistics Externship, Adjunct Assistant Professor, Department of Psychology

     University of Rhode Island

 

Grads Dorm 206a (click for map)

593 Eddy Street

Providence, RI, 02903

office: 401-444-1493

cell: 401-639-3942

fax: 401-444-8271

Click here to request biostatistics assistance

Click here for special research technology support

 

CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential, proprietary, and/or privileged information protected by law. If you are not the intended recipient, you may not use, copy, or distribute this e-mail message or its attachments. If you believe you have received this e-mail message in error, please contact the sender by reply e-mail and destroy all copies of the original message.

 

[erika....@gmail.com]

Hi Carly,

I realize I am late to answering this but thought my experience might be helpful. We had the same concerns as you, and the best solution we have come up with is to set up our surveys so that if a participant leaves one or more questions blank on a page, when they click "next page," the next page is a message that says "You have not answered 1 or more questions on the previous page. Please click "previous page" below and answer the question(s) that you skipped by mistake. If you skipped any question(s) because you do not want to answer them, that is ok and you may leave these questions blank and click "next page" below to continue with the survey. Please ask the researcher for help if needed!" If they have answered all questions on a page, they won't see this page when they click "next page" - it will just bring them to the next page of the survey.

In addition to these messages, we check all of their answers when they complete the survey (if they did it in the lab with us) and if we find any missed items, we point them out, open the survey for editing, and allow them the opportunity to answer them if desired.

We haven't figured out a way for the message to direct them to the specific items they missed, but perhaps that is possible with more sophisticated branching logic.

-Erika Bloom

[Isabelle Swearingen]

Hi everyone, 

Our research team has just come across the same issue when setting up our survey- we don't want questions to be mandatory due to ethical reasons, but would like to set up a pop-up reminder if a question was left blank. 

Have any of the redcap updates addressed this issue? I realize this thread is many years old, but thought it was worth a try to ask!

Thanks,

Isabelle

[Nora Salo]

Hi Isabelle - did you ever find an answer to this?