Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...
The Difficulty of Internet Surveys
1. MKT375-2, Fall16 Ann Connolly
29/09/2016 1
T H E
D I F F I C U L T Y
O F I N T E R N E T
S U R V E Y S :
R A N D O M S A M P L I N G & S Y S T E M A T I C E R R O R S
3. I M P O R T A N C E O F S U R V E Y
A C C U R A C Y
MKT375-2, Fall16 Ann Connolly
(Zikmund, Babin, 2010 )
29/09/2016 3
4. N O T H I N G
I S
P E R F E C T
a2ua.com Ballerina 2016
29/09/2016 4
MKT375-2, Fall16 Ann Connolly
5. Total survey error=
Random Sampling Error +
Systematic Error
(Assael, Keon, 1982)
29/09/2016 5
MKT375-2, Fall16 Ann Connolly
6. Occurs because of the
selected variations in the
sample.
(Zikmund, Babin, 2010 , 308)
29/09/2016 6
R A N D O M S A M P L I N G
E R R O R
MKT375-2, Fall16 Ann Connolly
8. Occurs because of the design of
the survey itself and overall
correctness of the execution.
(Zikmund, Babin, 2010 , 308)
29/09/2016 8
S Y S T E M A T I C
( N O N S A M P L I N G )
E R R O R S
MKT375-2, Fall16 Ann Connolly
9. Why do these errors happen?
MKT375-2, Fall16 Ann Connolly
29/09/2016 9
10. (Assael, Keon, 1982)
29/09/2016 10
The sample
selected is not the
“perfect” portrayal
of the tested
population
VS
More difficult to
control.
Due to
nonresponse &
response errors.
R A N D O M S A M P L I N G
E R R O R S
S Y S T E M A T I C
E R R O R S
MKT375-2, Fall16 Ann Connolly
11. When some people of the sample do not respond.
When respondents respond inaccurately.
(Assael, Keon, 1982)
29/09/2016 11
MKT375-2, Fall16 Ann Connolly
N O N R E S P O N S E
E R R O R
R E S P O N S E
E R R O R
13. Research concluded that
random sampling error is minor
while Systematic Error is
a large contributor to total
errors
(Assael, Keon, 1982)
29/09/2016 13
MKT375-2, Fall16 Ann Connolly
14. T H E D I L E M M A
Select a large sample to minimize
random sampling error OR put
money and effort on smaller
samples to ensure better outcome
(Assael, Keon, 1982)
29/09/2016 14
MKT375-2, Fall16 Ann Connolly
15. But given possible time
constraints, it is difficult to
concentrate on doing both.
MKT375-2, Fall16 Ann Connolly
29/09/2016 15
16. Therefore, the popular choice is
to conduct Internet Surveys
MKT375-2, Fall16 Ann Connolly
29/09/2016 16
17. I N T E R N E T
S U R V E Y S
(Hoonakker, Carayon, 2009, 350) 29/09/2016 17
MKT375-2, Fall16 Ann Connolly
18. Yes, internet surveys improve
response rate…
But, difficult to make accurate.
(Ziegenfuss, Beebe, Rey, Schleck, Locke, Talley, 2010) 29/09/2016 18
MKT375-2, Fall16 Ann Connolly
19. H O W E R R O R S I N T E R T W I N E
W I T H T H E I N T E R N E T
MKT375-2, Fall16 Ann Connolly
29/09/2016 19
20. Sample size can be easily increased
in Internet Surveys
T O R E D U C E R A N D O M
S A M P L I N G E R R O R
MKT375-2, Fall16 Ann Connolly
(Hoonakker, Carayon, 2009, 350)
29/09/2016 20
21. How to reduce Systematic
Errors in internet surveys?
MKT375-2, Fall16 Ann Connolly
29/09/2016 21
22. T A K E I N T O A C C O U N T T H E
D I F F E R E N T T I M E Z O N E S
MKT375-2, Fall16 Ann Connolly
(Zikmund, Babin, 2010 , 308) (Hoonakker, Carayon, 2009, 353-359)
29/09/2016 22
23. F I N D R E A S O N S F O R
N O N R E S P O N S E A N D
R E S P O N S E E R R O R S
(McDaniel, Gates, 1999) (Matzkin, 2007) 29/09/2016 23
26. C O N C L U S I O N
MKT375-2, Fall16 Ann Connolly
29/09/2016 26
27. Random sampling errors cannot be avoided. They are
caused by chance. To reduce it is to increase the sample
size, which is made easily through internet surveys.
Systematic errors have a large impact on the total error.
Although Internet surveys are the popular choice due to
rapid response, it contains a possibility for large amounts
of errors.
MKT375-2, Fall16 Ann Connolly
29/09/2016 27
28. Possible ways to reduce non sampling errors on internet
surveys:
Consider the time zones of who will
be selected.
Find the reasons behind
nonresponse and response error.
Extra thought must be put into the
design of the survey.
Place incentives for completion.
Quality over quantity.
MKT375-2, Fall16 Ann Connolly
29/09/2016 28
30. References
Zikmund, W. G., & Babin, B. J. (2010). Essentials of Marketing Research (4th ed.). Mason, OH: South-Western/CENGAGE Learning.
McDaniel, C., & Gates, R. H. (1999). Contemporary Marketing Research. Cincinnati: South-Western College Pub.
Assael, H., & Keon, J. (1982). Nonsampling vs. Sampling Errors in Survey Research. Journal of Marketing, 46(2), 114-123.
Retrieved from http://www.jstor.org.ezproxy.lib.cas.cz/stable/3203346
Groves, R., & Lyberg, L. (2010). TOTAL SURVEY ERROR: PAST, PRESENT, AND FUTURE. The Public Opinion Quarterly,
74(5), 849-879. Retrieved from http://www.jstor.org.ezproxy.lib.cas.cz/stable/40985408
Ziegenfuss, J., Beebe, T., Rey, E., Schleck, C., Locke, G., & Talley, N. (2010). Internet Option in a Mail Survey: More Harm Than
Good? Epidemiology, 21(4), 585-586. Retrieved from http://www.jstor.org.ezproxy.lib.cas.cz/stable/25680593
Renckly, T. R. Sampling Error. Retrieved September 20, 2016, from http://spectrum.troy.edu/renckly/smpl-r.htm
MKT375-2, Fall16 Ann Connolly
Hoonakker, P., & Carayon, P. (2009). Questionnaire Survey Nonresponse: A Comparison of Postal Mail and Internet
Surveys. International Journal of Human-Computer Interaction, 25(5), 348-373. doi:10.1080/10447310902864951
29/09/2016 29
Matzkin, R. (2007). Nonparametric Survey Response Errors. International Economic Review, 48(4), 1411-1427. Retrieved from
http://www.jstor.org.ezproxy.lib.cas.cz/stable/4542018
Editor's Notes
It is impossible to measure an entire population. We need samples. Taking samples are important, but equally as important is the accuracy of those samples. Surveys are quite convenient to use because of their relatively inexpensive process to gather data on people. Surveys are necessary gather data. Not only do they provide explanations to peoples’ opinions but can be beneficial knowledge to a company.
But of course, nothing is perfect. Errors cannot be escaped and are a part of information data. It’s essential to know how to reduce these errors to achieve the best possible results and how large of a role it plays with internet surveys.
Nonresponse error is when some people of the sample do not respond. This could be due to many reasons such as no interest in the survey, the wrong time, etc. While response error is when respondents respond inaccurately. This could be due to misreporting answers because of embarrassment or personal reasons, simply can’t remember the answer so they guess, wanting to appear more intelligent, or other environmental factors.
Internet surveys are seemingly the most popular form of surveys. It can reach a wide audience with minimal time and retrieve a quick response. Respondents can be reached across the world at any moment. Samples can be gathered almost instantly. Yet despite these advantages, there are just as many disadvantages.
We must take into account certain times zones where the survey is given due to the kinds of people who view it at that moment. Time plays an important role in responses to surveys. For example, if a survey is sent out during a busy hour/time during their work schedule then that person will be less likely to respond. Therefore, if interested in gaining responses from different locations around the world then the survey cannot be sent out all at once. One hour during the day is different for someone else's’ hour in another area.
Because of the popular use of internet surveys, people are more selective to filling them out. People are more likely to respond to an issue they are interested in, which leaves room for nonresponse errors. It would be ideal if all of the selected people participated in a survey, yet this is unlikely. The real question is of those who did not respond, was there an important reason they all share in common?
For example, during a survey, retired people answered more about lifetime savings rather than people who were living in a busy city. This shows there was an overrepresentation of retired individuals in this survey. (McDaniel, Gates, 1999)
The design. To reduce total survey error, the heart of this resides in the design and central structure of the survey itself. Shorten the questionnaire, make it more friendly for respondents. Provide incentives for participation and completion. By paying attention to questionnaire design, we must be sensitive to questions which are hard to answer or might make respondents look stupid. It is easy to go on SurveyMonkey and make a quick survey to send out instantly throughout the internet. But lots of time and effort must be spent on this process even if it is short.