BookmarkSubscribeRSS Feed
Fluorite | Level 6

Hi Everyone,


I am new to statistic and I have a question in my mind.

When I saw some data and statistics published by news, researchers or universities, could I write back to ask them further for example how the data was collected, the sample frame, sample size, etc.?




Of course you can.

All papers from researchers / universities have a Corresponding Author.
The point of the corresponding author is who to contact if you want to correspond about the paper.

If it's press, you just find contact details on their website.




Super User

You can ask but TV News is not very likely to go into details.

Depending on the site, web-sites are hit-or-miss whether you get any response or usable information.

Many "sources" won't even share the actual questions they asked.


Fluorite | Level 6

@sbxkoenk  @ballardw  are you familiar with panel?


I have a university in my country that uses panel for their research on issues relating to the well-being of ageing populations and majority of survey are done online.


I am wondering if survey is done online, is there a video camera to ensure the survey is actually done by the participant?

Also, for participants who have successfully completed the previous month's survey, vouchers (award) for these completions are given.

I am curious about this methodology whether awarding will leads to response bias?




Hello @marcuswong ,


Initially I thought you were talking about regression with panel data (panel data = time series cross-sectional data). Panel data regression can be done with PROC PANEL, PROC CPANEL, PROC GLIMMIX, PROC TSCSREG and a couple of other procedures.


But you speak of a panel of persons in an online survey research.
Unfortunately I am not a social investigator / researcher and have very little experience with survey research and response bias therein. I am more in the field of industrial statistics.


But your question is definitely an interesting one!
I would open a separate topic in the 

Analytics > Statistical Procedures board, and I'm sure you will get some informative and interesting answers.


Good luck,


Super User

None of my clients that were performing online surveys had the resources to do any of the likely approaches for verifying invited respondent was the actual respondent. The main reason in my case they were doing online surveys was the cost was much less than other methods such as telephone surveys.


I am not aware of anyone doing surveys that have video records of survey participants. I won't say there are none, just that I am not aware. The video wouldn't mean anything if you did not already have some video of the intended respondent to compare for identification. Even at that I could see a camera capturing an individual but not the key board of responses.


If you want to do something like that I might suggest contacting the people that do the SAS certification exams where they have some goal of making sure the person getting the certification is the one taking the test. I doubt of the service is inexpensive.


Incentives (your vouchers) may have bias but how much and is the additional information gained by having more respondents than you might have without the incentive. Lots of arguments on both sides. If a target audience is hard to reach incentives may be the only way to get enough sample but only if the respondent thinks the incentive is worth it.



Registration is open! SAS is returning to Vegas for an AI and analytics experience like no other! Whether you're an executive, manager, end user or SAS partner, SAS Innovate is designed for everyone on your team. Register for just $495 by 12/31/2023.

If you are interested in speaking, there is still time to submit a session idea. More details are posted on the website. 

Register now!

Discussion stats
  • 5 replies
  • 3 in conversation