A Practical Guide to Find Respondent For a Survey

As a student, researcher or even for market research, you often need to find people to fill in those damn questions.

In this short article, I will give some practical tips from what I have learned as a Test Construction & Data-Analytics teacher for University students.

You will learn:

• How to get respondents quickly
• The free approach VS paid approach
• Important considerations if you need to do statistical analyses

DON'T Spam friends & family

Sending your questionnaire to your own network is the most obvious, the most used, and the least effective way to get respondents. Unless you are Instagram famous - don't expect this to work. Your friends might even start to hate you when you keep begging them to do it (ask mine).

If you survey is longer than 2 minutes, the chances are low many friends will make it. Don't take it personally, no one likes to make surveys without getting anything in return.

Don't get me wrong, you SHOULD post it on your own Facebook, Steemit, Twitter and whatever platform you have. But just don't expect too much from it.

A Better Way to Get Respondents

Free Options:

When you don't want to spend any money (or don't have any money, yet), there are a few options:

1. Posting on facebook groups
2. Using online platforms
3. Get sponsors

1. Posting on Facebook Groups for Survey Exchanging

What are Survey Exchange Groups?

Yes, there are actually facebook groups created for the purpose to swap surveys. This means that "I fill in yours, you fill in mine'.

Most people in the group are just like you. They don't cheat and actually make your survey, at least, most of the time.
Based on personal experience, the average response if about 60%.

Two problems

1. Especially when your survey is longer than 5 minutes the risk of people not reciprocating becomes a lot higher. You can imagine that if your survey takes 2 minutes, someone makes it and asked you to make their 30-minute long survey, you feel a bit cheated.

2. Other than that, there's no mechanism to ensure someone else fills in yours, besides telling an admin. The problem is, admins are not incentivized to maintain a group like this, so they usually don't. You can end up making other people's survey/questionnaire without getting any in return.

How to use these groups

It's always best to post your survey with a clear title. Introduce yourself briefly, and - if your survey is under 5 minutes - be sure to mention how long it takes! Tell people when you took their survey in the comments, that reminds them to take yours. You can even paste the link to your survey as a comment to their survey because Facebook will - most likely -
give them a notification.

How to Find Survey Exchange Groups

You can find survey exchange groups by typing the following keywords into the facebook search bar:

"Find Respondents", "Survey Exchange", "Survey Swap", you get the idea. Don't forget to search for groups in all the languages you speak.

I advise you to join as many groups as you can find, at least 5.

You could post in reactions to this article since people reading this will be willing to help you.

Some Active Groups:
English/Dutch: https://www.facebook.com/groups/vragenlijstruilen (I made this one myself over a year ago)
International: https://www.facebook.com/groups/698363963614551/
International: https://www.facebook.com/groups/students.survey.exchange/
https://www.facebook.com/groups/thesispk/

Upside:

Fast! People usually react quickly

Actually works. (at least better than begging friends & family).

Downside:

You have to fill in surveys yourself..

Some people don't take your survey in return

Length of the surveys can be unfair

2. SurveySwap: Online Platform to get Respondents for Free

SurveySwap is a free online platform (https://surveyswap.nl/) that works on the same principles as the facebook groups:

'You fill in in mine, I fill in yours'.

How it Works

There are two main advantages of this platform:

1. Fair 'length system'
2. no cheating

First, when you have a survey that takes more than 5 minutes to fill in, the Facebook groups will not be very effective. Also, when you have a short survey, you won't have to fill in surveys that are LONG because you feel guilty towards the person that has already filled in yours.

The way the 'system' works is by accrediting 'credits' for every minute spent filling in surveys. So, the length of the survey determines how many credits you can earn.

Example:
You create a survey that takes just 1 minute to fill in.
You make a 5-minute survey on SurveySwap. You receive 5 credits.
In return, 5 people fill in your survey.
Rinse and repeat until you have however many people you need

How to use it

First, go to surveyswap.nl and create an account. Make sure to select the languages you speak so you don't get surveys in languages you don't understand.

Second, you want to make a choice. You either post a survey on the platform if you have already created your survey in another questionnaire creation tool (like Qualtrics or Google Forms), or you create your survey directly on the platform.

Don't forget:
If you choose to post an existing survey you want to make sure you place the SurveySwap LINK at the end. Since there are actually admins and reporting functions, you will get banned and lose the option to use the platform - unless you make another account of course.

Third, fill in some surveys to become visible to others. I know it sucks to make surveys, but let's not beat around the bush:

You want it for free, so you have to make a little effort.

Let's do a simple math example:

Most questionnaires:

• take 5 minutes to complete
• need 80 respondents
• take 2 weeks to complete

So, you need to spend 5 * 80 = 400 minutes making surveys.

That's 6.667 hours, so less than a normal workday. In contrast, most students and researchers dabble around and take weeks to complete collection all their respondents.

Upside:

• Length of survey/questionnaire gets equalized
• Cheating is (almost) impossible
• Active users in one place
• Ease-of-use
• Rumours it will be made as Dapp on the EOS blockchain if beta is successful

Downside:

• Still in beta
• For non-techies, it can be a little difficult to get used to
• You can't send personal messages to the pretty girls that make your survey

3. Get Sponsors for Your Research

This is probably the best way to get respondents for free, but a little challenging. It takes some convincing and depends heavily on how important (and profitable) your research is. I would only consider this method if you need A LOT of respondents (200+) because it takes some time and effort to get a sponsorship.

For example, I did two majors: Organizational Psychology and Neuroscience.

You might not expect it, but it was MUCH easier to get sponsorship from the University (University of Amsterdam) for a questionnaire on work performance than a study on activation patterns in the neocortex when exposed to stress. The survey for work performance was as easy as asking the head of research.

Sometimes it's as simple as just asking... Lesson: Don't ask, don't get

Get creative

When your research can add value to any company, research, medical, therapeutic or charity group. Get creative! For instance, I was thinking:

"Who would be interested in these survey results"?

For the 'brain study' it was actually pretty easy when I thought of it, namely: websites that sell meditation courses. I e-mailed a few, explaining the idea of the study. I got a few to make a phone call and they actually loved it! They placed it on their website and a lot of people filled it in! But I still needed 600+ more respondents...

Another didn't want to place it on the site but asked me how much it would cost to get the results... HUH? Yup, they sponsored me a few hundred bucks. Success! I then used that money to pay for the remaining 500 respondents.

This takes me to the next part, paid options to get respondents!

Paid Options to Get Respondents

Needless to say, this option the best. If you have the budget, it saved you a lot of time and frustration.

The only problem is that some companies charge too much for this service. This also depends largely on if you have any special needs for your respondents. For instance, if you need only women between 15 and 18 it will be a lot more costly in comparison to 'anyone in the UK'.

In general, there are two types of companies that have this service:

1. Data agencies
2. Online platforms

1. Data agencies

If you have a simple survey I would stay away from this option. They are slow and will overprice you.

But, if your study is very complicated or needs a special audience, they are often better.

2. Online platforms

Online platforms are faster and cheaper but are not good for complicated surveys or specific audience requirements and demographics.

Two services I like are: https://www.pollfish.com & https://www.surveymonkey.com/mp/audience/audience-pricing

You pay about $2-5 dollar per respondent for simple surveys. If they are longer, more complicated, or more specific audience prices can increase dramatically (and you are better of using an agency).

Important Considerations if You Need to do Statistical Analyses

Either you are a student or working for a company. I will break it down for the two separately.

For professionals

If you are collecting data for your work (i.e.: market research) there's just one thing you should remember:

"don't buy all-in- one packages!"

This is a classical way to pump prices up. It's the same as the scams you see in Thailand, where scammers sell 7-day trips with all kind of activities for 20X the normal price (if you would have bought everything yourself directly.

When you buy data (respondents), survey consulting and analytics reports in a package, you will pay a LOT.

A simple way to do it more cheaply has 5 steps:

1. Go to upwork.com and find a Data-Analytics Expert that knows at least SPSS and Stata. A bonus is if they know 'R'. You can also find an expert on afstudeerbegeleider.nl (Dutch for graduation tutor). I don't know any international companies that do this, and since I'm Dutch and used to teach there years ago I know they are good. Just in case you are worried: Us Dutchies can speak English quite well.

2. Consider your research goals (what do you want to find out) and let the Data Expert tell you how to structure your questions

3. Create the questionnaire

4. Pay an online platform for respondents (as explained above).

5. Give the data sheet back to your expert and let him do his magic. Make sure to ask to pay per hour.

For Students

As a student, you have to learn how to create a proper survey. You are allowed to get tutoring to learn how to do this, but don't expect any other help.

There are a few important things to consider before starting.

1. Make sure you structure your survey the right way, to avoid do-overs.
2. Don't overcomplicate what you want to know
3. Ask for help when unsure

1. Structure Your Survey Correctly

You want to make sure that most of the questions you ask are on a 'scale'. This means they are measured in a number. A good way to do it is by using 'likert-scales'.

For example: "I like steemit"
1= totally disagree
2= disagree
3= don't disagree, don't agree
4= agree
5= totally agree

The reason you want to do this is that most statistical analyses require scale variables (questions). If you don't, this can later give you trouble. There are other options for non-scale variables, but they are MUCH more difficult and are a study by themselves.

2. Don't overcomplicate

It's better to answer one research questions well than multiple questions poorly. The best way to do this is to write your 'operational model' and 'process model'. This is a little beyond the scope of this article, but if you want me to write about it, let me know. Basically, it means that you give 1 number to each of the things you want to research. If you can't measure something in 1 number, you are probably overcomplicating.

3. Ask for help when unsure

Messing up these first 2 steps can really cause a lot of extra work, browsing the internet, headaches, and tears.

Don't expect this to come from the University. Most supervisors from University either won't tell you (they are often simply too busy) or don't know what you should do (because they usually only specialize in just one type of analysis).

You can try to find friends that were really good at statistics and had a high grade. But chances are they also only understand the choice they had to make with the way they happened to structure their data, but not the choice you have to make based on your structure

Another option is to pay an expert. I know as a student you don't have a lot of money to spend, but if consider it a good investment in your future (maybe get your parents to pay?) At least I would try to get a second opinion on your survey by an expert before you send it out. It really sucks if you have to do it over.

Again, I only know a Dutch company that does this afstudeerbegeleider.nl, but I'm sure you can find them on Upwork, Craigslist or similar.

Ending Statement

Hopefully, this will help you. This is the first time I write a blog, so let me know how you like it!

p.s.: As I mentioned before, you can write the link to your survey in the comments to connect with others and help each other out.
p.p.s: English is not my native language and I was pretty tired when writing this, so forgive me :)