ask them to take email id and phone numbers of respondents, you can tell your students that you shall randomly call few respondents.

sameer

Hey Heather,

talk to them about plagiarism and faking data and what it means for science and why it is bad. Give them somthing to think about (i.e. here is a nice article from the Washington Post: http://www.washingtonpost.com/national/health-science/the-new-scientific-revolution-reproducibility-at-last/2015/01/27/ed5f2076-9546-11e4-927a-4fa2638cd1b0_story.html)

I think that getting email and phone information of respondents, then contacting some or all, would raise questions about privacy and confidentiality.  This could be an ethical issue itself.

Tino Lorenz has a good suggestion, and I think that teaching research ethics is -or should be- a part of teaching research.  

Your university probably has an official statement on academic integrity which you could share, pointing out the consequences of being caught (zero on the assignment, failure in the course, or expulsion are probably there). 

I used to work in a program where we regularly had students do mini-research projects and falsification of data was never an issue.  Do you have reason to believe that the students are likely to make up their data?  If the number of respondents you have is high, that might encourage them to fake data, we kept our expectations for sample size low, as it was the process that we valued more than the validity of the results.  Depending on your objectives, giving them the luxury of a small sample size might work.  Allowing them to use an online questionnaire (I've used Google forms for this) that they share with their personal networks is another compromise that you could consider, as this will be easy for them to do (they can share the link on social media and ask others to share as well, not to mention that the software will collate data automatically and thus save time as well).

You mention that they have chosen their topics and there should be some intrinsic motivation/interest there; this is sensible and I think this is part of what worked for us.

It's not just students who are motivated to fake data, relatively new research has come out describing this effect in highly influential international surveys. .. But that aside, a good and relatively affordable way to deal with this has been to have them audio record their interview along with the survey.  In our process, it was used to help "verify the data" in terms of whether the researcher appropriately recorded the answer/s in a multi-lingual study. (The survey was conducted in Bengali and recorded on a form in English; so did the respondant indicate "very happy" or just "happy" when the words used were nothing like what was ultimately recorded.)  A by-product of this interpretation measure was to assure ethical data gathering as well.  

The audio recordings of student data capture never need to be fully listened to - but spot checking is quite simple and very difficult to fake.

By the way, I am just now writing up a process for this using LiveScribe smart-pens.  The point being that with some explicit process attended to in advance, the smartpen can be used to capture and redundantly backup the data from the pen & with simple but clever OCR->Python work, the data may go from ink to spreadsheet in an automated process.  If you're interested, write to me.

Videotaping is a good method for data collection. Now days it is very easy to do. You can use your phone and you can save the video, and as teacher you can ask the students to send some videos, you can control. The students have to be clear in the information to the participants and the students must be sure that the participants allow a sharing with the supervisor. 

Some really good suggestions have been presented above. I personally wouldn't discard taking emails; afterall it is always nice when respondents are notified about the results of the survey they undertook, which if made clear prior to their participation, should not raise any concerns. Moreover, taking videos, should require extra ethical approval, similar to if the topic is of a sensitive nature. Just consult the material on ethics of your institution and proceed accordingly

Hi everyone,

Thanks so much for the suggestions, lots of good thoughts here. 

I had already covered plagiarism, and they are only required to get thirty completed surveys, so that is not such an onerous task (I hope), so hopefully that helps.  Unfortunately, at this time, I have no budget at all, though in the future it would be great to try some of the technology-based strategies you suggest.

I don't have any specific reason to assume that they would fake the data, but I know that some undergrads will procrastinate and then be tempted to do whatever it takes at  the last minute to get something in on time. 

The attempted solution I have implemented is to get them to use and turn in to me an "on-site authorization" form, where someone in authority at the location that they are surveying (i.e. outside the student union building, in a dorm building, downtown on a street corner, wherever) signs and completes the form, giving "permission" for them to collect data in that location.  This (I hope) will serve several purposes, to a) force them to think about courtesy and letting people know they are surveying in a particular location ahead of time, b) have the students actually out in the community somewhere and c) have someone sign off with an email so if needed I can email them later and confirm that the student actually was out in the community trying to survey people, and not just in their bedroom filling out forms.

I will let you know how this turns out.  Further discussion and suggestions are still welcome.   Thank you again!

Depending on how long the survey is, 30 can be quite a lot for students to capture.  Try it yourself to be sure!  Time yourself, of course. 

Another step that hasn't be explicitly mentioned is whether you've gone through the IRB process.  Depending somewhat on how documented your institution's IRB is, that can be a good way to give them the sense of seriousness of research with human subjects as well as how to follow through ethically. 

I like the video-recording approach, but 30 surveys (again, depending on length) takes a lot more space than the audio & audio is much easier to capture (there's no need to watch any screen to see that what's being captured is what was intended.  

Please do let us know how this turns out!

As well as the above excellent suggestions, you can also have students make the completed questionnaires available for you to see - they will be very aware that they should not all completed (or signed, at least) in the same hand writing!