Jump to content

SPECIAL: Andrew White of Crowdology Answers your Questions!


Administrator

Recommended Posts

  • Replies 30
  • Created
  • Last Reply

Top Posters In This Topic

  • Crowdology

    5

  • lev

    5

  • 1957horses

    4

  • footfree

    2

@Andrew

I guess I was already signed up. Funny I NEVER got any surveys. But today as I was sighing up (that's how I found I already was) there was a survey to do.

So I click on this survey "Length 15 min Reward $1.63" and I get this " Unable to locate survey." so is this something that happens a lot??........................... :shock:

Link to comment
Share on other sites

Many thanks for the 'routers' link/page, very interesting :wink: and the replies.

Is it possible that surveys panels, apart sending surveys to the right potential target group, they may also try somehow to distribute their surveys to everybody in their panel? So everybody have a chance. Just because I saw with others of my similar profile we do not always receive the same invitations or have the same 'survey-silent' periods. Just a wonder. :o

However, with a more cynical view, I sometimes suspect that this same survey-distribution practise may also be a way for a company to slow down payouts :lol:. Many thanks for your views.

Link to comment
Share on other sites

A few things I have run into:

1. Why do survey companies have you watch a nonsense video , ask if you can see & hear it , and then the survey does not use video at all? A waste of time. For those with fast connections , video plays fine. Slow connections cause the video to be choppy or come in short bursts. Wouldn't it be smart to either download the video as a temp file to play instead of streaming?

2. Why do some sites have you go through the numerous questions (30+) gathering data and then come back with a screen saying that you already completed the survey (no credit).

3. Why do some companies (Zoompanel comes to mind) send invites , often in the wee hours , for surveys and when you go there there are no surveys , even when you are at the site when the invite arrives? In the case of Zoompanel , the response given was that their surveys fill up in seconds. Must have very low quotas and be really short to be over that fast!

Link to comment
Share on other sites

Hi Andrew,

Why does there seem to be large differences in the incentives for the same survey, taken from different sources? I recently done a tobacco survey on Toluna for about 50p and a few days later I done the same one on a Cint panel for just over £2.

I have heard a lot of people complain about the so-called "Characteristics Grids" in surveys and how they find them really frustrating and tedious. I am one of them and I have, on more than one occasion, quit a survey because I detest these types of grids . The grids that I am talking about is where you have Company A, Company B, Company C ect. across the top and a list of characteristics such as Modern, Helpful, Leaders ect. and we have to click checkboxes to select characteristics that apply to each company. Do you think these types of questions are answered carefully and honestly? and do you think they could be improved to make them more interesting to do?

Link to comment
Share on other sites

Firstly thank you all for your questions, we are reaching the end of the week and I’ve found it really interesting to understand the sorts of questions you have. I hope you have all found it useful!

Cricket – great to hear you’ve joined up, I hope you enjoy the surveys.

Footfree – Redshift Research Ltd own and operate Crowdology

1957horses – That is odd, all survey invites should be sent to your email as well as visible in your account. With regards to the ‘unable to locate’ it is possible this survey was closed for your quota, although if you feel it is a technical issue please let our support team know at [email protected] who will be able to see which project it was and pass on any comments to the survey makers if necessary. As with all surveys, if quotas are full then you will be unable to take the one you were originally invited to. However, Cint has a survey matcher that will often try and match you with another relevant survey.

LEO – I think I will try and answer this with an example, as it is easier to explain that way. If you and a friend were both signed up to Crowdology for instance and had exactly the same profile you would not always find that if one of you receives a survey that the other would. Say we have a survey on apples that needs 1000 responses and we have 3000 people on our panel who match the criteria. If our response rate is high we may only need to send this out to 1500 people to get the required number of completes, thus 1500 of those who would have qualified aren’t invited. This is where routers come in to play – if one of those 1500 that hadn’t been invited to the apple survey but had been invited to another survey that was full, they would be routed to the apple survey (assuming it was not already full), which they could then complete. I can promise you that for Crowdology we have no way of distributing surveys to slow down the pay-outs. All sample for each survey is selected randomly and not based on how much credit you currently have.

Episemion – thank you for your questions, I will answer as best I can

1. These videos are usually to see if you have the right technical requirements to continue with the survey. It may be that, even if they don’t play a video or sound there are pictures of question formats which require a certain software.

2. This is clearly an issue, as many others have questioned this throughout the week – as I said before, screening questions shouldn’t take longer than 3-4 minutes. If you feel you have completed a whole survey correctly and not been credited you should contact the relevant support for that survey site as this is unfair and should be made aware of so it can be prevented from occurring again.

3. I can’t be sure of the answer to this, as I haven’t any personal experience of Zoompanel. However, it is possible that they are sending out ‘scheduled sample’ (hence why it may be at an odd time – or it may be they are based in a different time zone). It is also possible that their survey is almost full and they maybe only need 10 more completes, but need to send out maybe 20 invites (some panels will send out more depending on their response rate) to be sure to achieve these.

Gothicsmurf

1. Each panel owner will set their own payment rates. Often if it is a large survey or needs to reach hard to reach audiences then several panels will be used to recruit panellists. In this case, one panel may set a minimum of 50p for a 10 minute survey, while another will set a limit of £1 for 10 minutes. It is therefore possible that depending which panel you belong to you will receive a different incentive for the same length of survey.

2. I have to agree with you here, especially after hearing so many complain about these grid style questions. At Crowdology when designing and setting up questionnaires we try and keep the amount of grid questions to a minimum. However, sometimes it is difficult (perhaps due to client demands or software limitations) to avoid using them entirely. We understand that too many of these grids in one survey, or a grid that is too long may well reduce the quality of answers, hence why we try to minimise their usage. In terms of whether they could be improved, this is a good question. I am sure with enough time and resources they could be, however these are often not readily available. Did you have any thoughts on this yourself, as we would love to hear how you think they could be improved? It is of course possible to separate these into separate questions but market researchers generally feel this would be even more time consuming and tedious – but perhaps we are wrong?

If you don’t mind – could we ask you all a question? We would like to know what style of questions are preferred, and perhaps others could share their thoughts on how grid questions might be improved? Thank you in advance.

We really appreciate all the questions and the feedback.

Link to comment
Share on other sites

Hi Andrew and many thanks for your time and all the replies, all very interesting.

GRID for me - For thin and long rows alternating light colours helps to follow the dots to tick along each row.

LONG GRID (vertical) it is useful to have the 'column top-headings' at every few rows (eg. 4,6,8), depending from how thick are the rows themselves (the amount of info in the box at beginning, whether there is a logo...).

WIDE GRID, (many columns 6/8+ :( ): If it is necessary to have so many columns (which I suspect for some surveys it is), I always wish it was divided in 2 parts in the same page (top part, then a second below). Then, a new part of the same grid (if it is also a vertical long one) next page and the same split in two parts...

AS FAR AS you are told from start how many of these to expect, I guess it is acceptable.

I come across some grids which would have been huge, but they were split onto different pages (to click 'next'); at the top you were told how many pages, eg. 1/8, 2/8...I thought it was anyway easier to read and tick in this way even if long.

I think if a grid makes it too difficult for respondents to manage it (tick the boxes), the way we tick may also be compromised.

PICTURES (logos) sometimes are enormous and make the grid even larger, I wished smaller ones were used. Some clever survey you can enlarge images just by hover over.

However, it also happened that you hover an image (even an ad) and the result is the same size image :lol: .

QUESTIONS INSTRUCTION - I wish some essential instruction were repeated at the top of a new page, eg. shopping, many questions options bought what where how... After many pages I may forget the exact point, eg. online, in-store, home, garden etc. The BACK button would be very useful but has almost disappeared from surveys, like a rare bird :roll: .

Concerning the kind of questions, sometimes questions are not really clear, an example could be useful. Whether they are open or close I guess we try our best to complete them. Some feedback box during or at the end of a survey helps as we can clarify, but if this box has to have a certain no. of characters it should be said/given with a counter (like twitter or ebay). It is frustrating when my comment is 'too' long and I was not told how long it could have been.

VIDEO/SOUND for me - I dislike :x when these are checked before ensuring you qualify. Perhaps a warning that it will be checked if you qualify should be enough. If I then qualify, happy to check for sound/video. It happens so often that after these checks I do not qualify, so very annoying. Even 'quota full'!!

SURVEY ROUTERS - I see your points Andrew but the invitation email is for a specific survey with specific reward and timing, routers not always give you this info, anyway I have lost so much time along the years :shock: by completing the same questions that brought me no good survey that now I usually close the tab as soon as I see the same long list of the same questions. SSI+ Nectar now does this regularly, no profile... a huge waste of time; samplicio.us the same... Well I value my time they don't. :|

A TIP FOR EVERYBODY for eg. large grids

I put the zoom buttons on the toolbar of the browser (Firefox), so I may use it during a survey in order to easily and quickly change the size of text, page, images..., it does help a little. :wink:

Many thanks Andrew for reading my views. :D

Link to comment
Share on other sites

Thank you very much LEO! We will have a look at this as a team and see if there are any of these we can implement immediately, or if we can speak to our software provider about updating grid formats to reflect some of your suggestions.

Link to comment
Share on other sites

Hi Andrew, thanks for taking our questions. I was just curious about how many people it actually takes for a company to start in a panel in a country? For instance, you mentioned that you may eventually expand to Canada. How many people in Canada would you actually need for your panel before you would start bidding on jobs to perform market research there?

Also, do you think companies who hire market research companies like yours actually check to see what their reputation is among survey takers? I know that they would be interested in the quality of market reserach that you actually perform, but what about the panelists? Do you think a company like Coke cares that XYZ survey company conducts good research, but doesn't reward their survey takers fairly?

With regards to grid questions, I think breaking them up is really important. Being faced with a big screen of grid questions sometimes makes me exit survey, because I just can't stand it! Having fewer of them listed on the same page is a good idea. I like slider-style grid questions better too. Somehow it makes them seem more tolerable versus just clicking on a radio button.

Link to comment
Share on other sites

Thanks Andrew for your insights into the survey world.

Regarding:

Videos: The survey should alert you upfront that a video will be required and allow users to opt out if their system is not able to view them. The other approach would be , as I suggested , have the video download completely as a temp file that could then be played. Even those with slow connections can play a video , just not stream it. I understand the checking if the user's system is suitable. However , if the subsequent survey does not use video at all , don't waste our time showing a nonsense one (such as the black figure in traffic one show at MySurvey).

Grids: I agree with the prior poster on the layout issues with the oversize (width or length) ones. I also am one that detests ones that have companies or products across the top and descriptive terms (modern , truthful , sexy) down the side. One way to reduce the grid size would be to ask the user what stores are in their area or what brands they are familiar with , then use only those in the grid. Makes no sense to me to ask to rate a store that the user has never been in on things like cleanliness or customer service.

Link to comment
Share on other sites

Thanks for your reply regarding grid questions, Andy.

Firstly I agree with leo's suggestion about the heading of the grids being always visible. It does become very time consuming when I have to keep scrolling back up to the top of the page to check I'm clicking the right checkboxes.

I have done a few surveys where, instead of grids, they use what I would call flashcards. How these work is that there is a pile of cards on the screen, with Yes & No buttons below the pile. Each card might ask questions like "Company X has good customer service" or it could simply be 1 worded characteristics, like "Modern" or "Market Leader". When I click "Yes" or "No" the card disappears and is replaced with a card with a different word or sentence on it. This cycle continues until I reach the last card and the "next" button appears at the bottom of the page, allowing me to move to the next part of the survey.

I find this type of layout is a lot easier to follow than large grids. I personally feel it improves my productivity and honesty in these types of questions as I am much less likely to speed through, or quit altogether.

I would love to hear other panelist's thoughts on the use of "flashcard" style questions.

Link to comment
Share on other sites

I agree with gothicsmurf, I too like flash cards, the only negative is usually 'no back button'. I suspect, however, some grids may be necessary with the many statements they are asking. The positive of a grid is that it allows some comparison between answers.

I agree with episemion, what is the point of including those brands we said we did not know at all.

Link to comment
Share on other sites

Hi all, thank you again for all your responses. This is our last post, and we just wanted to answer the last few questions.

Cherrysnowflake

1. With regards to how many people it takes for a company to start a panel in a country - this is a sort of chicken and egg thing really. That is, without enough demand it’s unlikely you would be able (or want) to build a large panel, but without a fair number of respondents it is difficult to market yourselves effectively to those wanting to conduct research. However, it is probably not as many as you think – some panels are small, and always start out at zero. This is where it can be useful to partner with large providers such as Cint and Toluna, who will be able to use your panellists for a wide variety of projects. As demand grows, usually so will your panel and vice versa.

2. As a market research agency we give details of our panel (such as response rate, how we pay panellists etc) to any new clients so they fully aware. Having not worked client side myself I can’t be 100% sure about what is most important to them, but I very much expect it is quality of data. As a market research agency with our own panels we understand that quality of data comes from ensuring our panellists are treated well.

3. Thank you for your thoughts on grid questions – we are experimenting with different styles of questions including the use of sliders

Episemion – Thank you for your views. What you say about alerting you upfront with regards to a video is a great idea and simple to implement. In terms of asking about brands which you are familiar with – this is often to find out if there are underlying themes which stops consumers using them/becoming familiar with them. Although I agree it is quite pointless to get you to answer questions such as cleanliness of a store if you have never visited it.

Gothicsmurf – thank you, unfortunately I don’t think our software currently supports flashcards but we will certainly look into it, as they sound much easier than grid questions!

Thanks again everyone for such considered questions and the detail of response in relation to grids and the types of questions you enjoy! It's been very informative :)

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.

×
×
  • Create New...