To consider falsifying survey data?

[whitecremeegg]

I'm doing distance learning masters research. For my particular topic, participants for my research cannot be just anyone, it needs to be a particular very small 'group' of people - trying to be vague so as not to be outing.

I can't change my topic, it's work related and work are sponsoring me.

However, there are two 'gatekeeper' organisations with databases containing the details of these participants. I need access to these databases to get their contact details to survey them.

One of the gatekeepers is my current organisation, who despite sponsoring me, refuse to let me use the database. I have regular access to the database as an employee but not as a student.

The other organisation has also refused.

I'm not sure what the best option is now. I cannot change my topic or participant group. I can't go into more detail as to why without being outing. My org is sponsoring about 20 employees to do this course but I am the only one from my department so unlike the rest of the employees I can't change my topic as they have more variety in their roles than I do.

It is not a subject that is likely to cause harm - it's not the NHS, children, criminal subject or anything like that.

There are only 3 options that I can forsee:

1. Use the database anyway but lie as to how I got the participants (say it was social media or something)

1. Use social media but end up with little or even no participants. I am connected to some of the participants via my social media channels but this is only like 10 out of 300 that I'd like to survey. Participants could also NOT be from my target group

1. Falsify survey responses to make it look like I got lots of responses from social media - but again there's still the issue of social media not giving me the 'right' participants. How likely is it that universities can prove surveymonkey entries are fake?

What would you do in my position? I really would like to do things 'properly' but at this rate the research is going to be pretty crap.

I thought this course would be easy and that my organisation would be more helpful. but they are very much of the "we are paying your fees, you must do this particular topic, get on with it'.

Please don't flame me. I know it's an ethical issue but I'm genuinely stuck. I did speak to my supervisor (not about my falsifying idea) and she's been rather unhelpful. She says just to go down the social media route and keeps saying that she's not supposed to help me too much and similar to my employer, it's very much 'get on with it'.

Any help appreciated

You're not even ruling it out.. Dear God. I hope you get found out and fired if you do.

So in your methodology you explain your sampling issues.

Everything from your methodology if you falsify data will be pointless, and in your own words "crap".

You're not ruling falsifying out?! Well let's hope you can deal with the consequences

Scientific research, and scientific advances, are built on everything that has gone before. If you falsify data, you betray everything that research stands for, and everyone coming behind you. I worked in a field in which data was falsified - minor data that a person thought would not matter. It set the field back decades and most likely cost lives. Think about that when you consider your options.

Please don’t get hung up of delivering the perfect survey - it’s will be spotted for sure. Much better to submit a project where you detail the different routes you have taken to find your respondents and then critique your method to bits if you don’t find enough responses. Six months into to a 2 years course you have another 12 months before things get really desperate.
Use all potential sources, so have a leaflet available explaining your research, inviting people to contact you. Give them out at every meeting, conference, party, you go to. Ask your boss, colleagues, old friends from uni / training courses you have been on to pass on to their contact. Alumni from your current college / uni. Are there professional bodies with websites and newsletters. Those who have been through this sort of course will know what it’s like and will often help. You can expect a 10-15% response rate to collect contacts like this so get networking.
Please don’t resort to cheating, it undermines not just your research and qualification but everyone else’s as well
And why not ask Mumsnetters - it will be cheap and quick and if you pick up one direct link and if you get one direct there maybe be other secondary links, it will be worth it.
Remember the Kevin Bacon rule of 6 degrees of separation- Everyone on the planet can link to Kevin Bacon in 6 steps. I don’t know how many mumsnetters there are but there are more of us than just one Kevin Bacon, so we are bound to know enough people to answer your research questions !!

Why can't you do qualitative research? 4-6 interviews, video and audio recorded over Skype? Get the participants' experiences of using the service, how they feel about it, how the issue affects them, and do an IPA?

Can you ask the organisations to send out the survey on your behalf? I work in mental health and quite often this is done. We can't have access to NHS records but we can ask the NHS to send out something to their bailing lists and ask them to contact us/ attend an event etc. Could you do similar?

*mailing

one of the problems I have is that one of the gatekeeper organisations has their member directory publicly available with email addresses etc...I'm not even sure if I need their permission but I asked anyway and they said 'no'. It makes no sense to me if any other person can just take the email address of their website.

There is no point in even doing research if your just going to take the results. Masters is not supposed to be easy!

Of course they said ‘no’ - they have too with the current and new regulations coming into force. And your MA supervisor is right, this is something you have to figure out - getting to the people who have the knowledge you want is not easy and you have to be creative in how you get to them. For my MA I wanted to talk to a local MP, couldn’t as he was in the TA and was posted to Saudi in the first gulf war. Bugger ! So I checked all his questions and speeches from Hansard to get his views on my subject ( supported housing for the long term mentally ill). Didn’t fit the survey precisely but hell that’s social research for you. Social research is about using a variety of sampling techniques.

Bet your gate keeper organisation has a newsletter and would welcome an article on your subject, where you could ask for people to answer your survey.

Think outside the box. Far more creative than cheating!

As part of your masters you’ll need to get ethics approval including proof of permission to use the databases - you don’t have it so can’t provide it.

The databases clearly don’t have permission for the data to be passed on or used for research purposes. It will be obvious if you falsify your ethics application and your survey data (not to mention hugely unethical.

You either need to recruit your survey participants using an alternative method, or choose an alternative research topic - surely you’re better off picking a different topic than risking your masters and your job through fraud

If the contact data is already available publicly, then you don't need their permission.

If you don't want to go down that route, could you recruit from an actual research panel, e.g. SSI? At the very least they might be able to offer some suggestions.

Can you ask your employer if THEY can send the survey monkey link to the selected group on your behalf?

That way they are not breaching data as they aren't giving you the contact details.

Why does your company have these contacts? ( Might help us figure a way around contacting them legally )

Did you say that these people's contact details are available to the public?

Ie) you can actually get it email addresses?

Surely thats fine then?

Don't falsify data. It's completely unethical and could land you in serious trouble with the uni and work.

Masters research and even undergraduate research is published and its not worth the hit to your reputation to falsify data. I know it's tempting but it's not worth your career.

The vast majority of the people on my degree course, myself included, used social media to gain participants for surveys. Some had a wide ranging target group (eg women aged 18-60) while others had a very narrow target group (eg female trailing spouses with more than one child who have lived in the middle east for 1 year +). Most managed to get the necessary responses by posting to targeted Facebook groups.

I'd try going down the LinkedIn and Facebook groups. If you have 10 contacts in your industry on Facebook then you can also ask them to forward the link to anyone in their contact list the meet the criteria you have, and the same on LinkedIn. You'd be surprised just how fast the link could spread to the target group in this way. You will end up with responses from people outside the target group but you just remove them from the data.

Keep on at work and your tutor. Explain the difficulties and see if you can find a work around. It may be a data protection issue re accessing the database as a student but there may be a way around it if work help, you might have to pester and point out that you'll fail the course they are paying for if they don't help you find a solution.

What is your organisations rationale for not giving you access to the database? Could you point out that they are wasting their sponsor money by not giving you access to the information you need to complete your thesis?

I was recruiting participants a year ago for my Master's. It took several months to get a decent number of people (over 100) and that was working hours every day. I really didn't think I was going to do it as the participants I needed were quite specific. I tried every Facebook group I could think of, blogs, Twitter, forums, student survey forums, contacted organisations to distribute for me etc. I also offered a small monetary incentive for completed surveys.

Whatever you do, don't falsify the data. Not only will you risk failing, but you won't have any satisfaction in completing your dissertation.

This kind of problem is what 'real-life' researchers deal with. As said above, it's how you discuss it (possible low numbers, method etc) that is just as important.

My guidance from my tutor is that data collection should take about 6 weeks so I don't have a lot of time.

All the providers are on my companies database. This other organisation is a type of 'association' for these providers but does not contain ALL of them. It does contain enough to give me a good enough sample.

The fact it's publicly available email addresses confuses me though. My tutor says to get permission, and I looked at the ICO website which didn't help much. I did see details on spam emailing but I'm not sure if that applies to me or if it applies to public member directories?

Anyway, there have been some useful suggestions above that I'll look into trying.

I swear I will do all that I can to get genuine survey respondents.

If the potential participant pool is very small, how realistic is your required participant number? There'll be stats somewhere about the number of respondents you can expect, I think from my masters it was about 5%. If you fake your data and it appears that 50% of approached participants responded they'll definitely know that you are committing fraud. If it's a small group then definitely go the qualitative route, or hedge your bets and go for a mixed approach, quantitative if enough replies to warrant it, with 3-4 follow up interviews you can do in depth. Honestly though, about a quarter of my dissertation was exploring issues and what got in the way/what I could have done better! It's all part of it.

I'd run with the social media option, get as many participants as you can that way and leave it at that. Even if it is only 10. The challenge of finding participants, and gatekeeper refusal will offer a lot in your discussion, and it's always good to be able to discuss the limitations of your own research

I would imagine that if you genuinely cannot access the participants you need to, then that is a valid finding of your survey.

You say you cannot change it, and cannot get access to the databases legally. So you'll just need to get whatever legitimate responses you can, via social media, targeted distribution, etc., and then deal with a smaller sample response.

my tutor says I should have written permission and attach this to my ethics application.

Written permission to email an email address that is avialble on a public domain?!