Follow the bouncing numbers.
Google is still the search leader but rival engines Yahoo! and MSN are starting to catch up. 'Google is the king of customer experience in the search engine industry, but Yahoo!, MSN and Ask Jeeves are improving,' said Dr. Bonny Brown, director of research and public services for Keynote Systems.Keynote conducted a survey of 2,000 web users for their opinions on the major search engines. Questions covering brand impact, customer satisfaction levels and future usage were asked. They compared the results of this poll with the same questions asked in May 2004 using the same demographic.
Google came out on top with 84 percent of those polled who said it's their primary search engine. 61 percent of those questioned said Yahoo! is their primary search engine - a 20 percent increase from last May's poll. MSN had a jump of 30 percent, with more people saying it was their main search engine. Microsoft's total was a low but respectable 38 percent. It's a nice turnaround for MSN as they are about to launch their new search engine to the general public.
hmmm so 84% + 61% + 38% = an astonishing 183% response rate on the poll.
Unless they were polling Democrats in Washington State, I'm not sure how they got that number. Found via James Joyner who (quite surprisingly) missed the math problem.




Comments (23)
um 84 + 61+ 38 = 183<... (Below threshold)1. Posted by fersboo | January 14, 2005 1:56 PM | Score: 0 (0 votes cast)
um 84 + 61+ 38 = 183
81 + 61 + 38 = 180
=P
1. Posted by fersboo | January 14, 2005 1:56 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 13:56
2. Posted by Paul | January 14, 2005 2:04 PM | Score: 0 (0 votes cast)
oh crap
Another guy over at OTB also noticed the numbers and I trusted his math THEN I compounded the error with a typo.
See!! I TOLD you critical thinking was important. So is critical typing I guess.
thanks
2. Posted by Paul | January 14, 2005 2:04 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 14:04
3. Posted by Mike | January 14, 2005 2:08 PM | Score: 0 (0 votes cast)
I use stats all the time and I think this is how it should have been presented.
84% said they use google as their #1
That leaves 16%
Of that remaining 16%, 61% use yahoo and 38% use MSN.
Does this sound right?
3. Posted by Mike | January 14, 2005 2:08 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 14:08
4. Posted by Michael Diaz | January 14, 2005 2:09 PM | Score: 0 (0 votes cast)
It could be in how they are using/distorting the numbers. For example, if 10% of all internet users used MSN one time and then the next time polled 20% used MSN there could be the misleading statement that there was an increase of 100%.
Another example. A cable company raises their “access” charge from 3% of the total cost to 5% of the total cost. In their eyes they’ve raised their rate 2%. However, a local community objects to this rate change, noting that their rate have increased 40% (because 2 is 40% of 5).
Basically, you’re taking percentages of percentages. Another example of making the numbers say what you want them to say.
4. Posted by Michael Diaz | January 14, 2005 2:09 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 14:09
5. Posted by Just Me | January 14, 2005 2:13 PM | Score: 0 (0 votes cast)
Could the poll have been a "check all that apply" or "top two" type question?
5. Posted by Just Me | January 14, 2005 2:13 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 14:13
6. Posted by Sharp as a Marble | January 14, 2005 2:14 PM | Score: 0 (0 votes cast)
I call that "No, really Honey. That's a full 8 inches" math
6. Posted by Sharp as a Marble | January 14, 2005 2:14 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 14:14
7. Posted by Mike | January 14, 2005 2:21 PM | Score: 0 (0 votes cast)
good observation Michael.
Stats are very malleable and can be shaped to create the image that the user wants to project despite the results.
I like to call it "mathmatical spin"..
This is why most people with training in polling and statistics don't take results presented by one side or another at face value.
7. Posted by Mike | January 14, 2005 2:21 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 14:21
8. Posted by Eric Anderson | January 14, 2005 2:23 PM | Score: 0 (0 votes cast)
Of that remaining 16%, 61% use yahoo and 38% use MSN???
using your logic:
Googe 84%
Yahoo (61% of 16%) - 10%
MSN (38% of 39% of 16%) - 2%
remaining: 4%
but I doubt this logic is right.
8. Posted by Eric Anderson | January 14, 2005 2:23 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 14:23
9. Posted by TallDave | January 14, 2005 2:26 PM | Score: 0 (0 votes cast)
If you can't do, manage.
If you can't manage, teach.
If you can't teach, report.
9. Posted by TallDave | January 14, 2005 2:26 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 14:26
10. Posted by Diane | January 14, 2005 2:32 PM | Score: 0 (0 votes cast)
"There are 3 kinds of lies: lies, damned lies, and statistics." -- Mark Twain
10. Posted by Diane | January 14, 2005 2:32 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 14:32
11. Posted by Cyclops | January 14, 2005 3:05 PM | Score: 0 (0 votes cast)
These numbers obviously indicate a serious problem with our search engine system. How many disenfranchised users had to wait long hours in the rain to get their results? How many sent their inquiries through Jeeves, only to see Google pop up on the screen? How many requests were processed through Diebold??
Somewhere, Barbara Boxer is weeping tonight.
11. Posted by Cyclops | January 14, 2005 3:05 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 15:05
12. Posted by basil | January 14, 2005 3:11 PM | Score: 0 (0 votes cast)
> Diane at January 14, 2005 02:32 PM
Benjamin Disraeli, not Mark Twain.
12. Posted by basil | January 14, 2005 3:11 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 15:11
13. Posted by Fred | January 14, 2005 3:14 PM | Score: 0 (0 votes cast)
What the survey actually did was ask web users who used a search engine whether they would use it as their primary search engine. From dmnews.com:
Google maintained the top spot in consumer satisfaction, with 84 percent of users who tested Google in the survey saying they likely would make it their primary search engine. That figure is essentially unchanged from a similar survey Keynote released last May.
By contrast, 61 percent of respondents who tested Yahoo said they likely would make it their primary search engine, a 20 percent improvement. MSN had 30 percent more users say they would use it as their primary search engine, though it still tallied just 38 percent. Respondents could give more than one answer.
13. Posted by Fred | January 14, 2005 3:14 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 15:14
14. Posted by SteveL | January 14, 2005 3:23 PM | Score: 0 (0 votes cast)
Damn Fred, so much for faulty math. It's more like faulty reading on our part. So this survey does not even address the point we think it does (what search engine is most used) it addresses feelings about whether a search engine is good enough to be your primary search engine. Makes more sense this way. There is no way Google gets 84% in an actual use survey.
14. Posted by SteveL | January 14, 2005 3:23 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 15:23
15. Posted by Steve L. | January 14, 2005 3:47 PM | Score: 0 (0 votes cast)
Another example. A cable company raises their “access” charge from 3% of the total cost to 5% of the total cost. In their eyes they’ve raised their rate 2%. However, a local community objects to this rate change, noting that their rate have increased 40% (because 2 is 40% of 5).
Actually, the increase would be 67% (2/3). You measure rates of increase or decrease against the original amount and not the final.
15. Posted by Steve L. | January 14, 2005 3:47 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 15:47
16. Posted by basil | January 14, 2005 4:25 PM | Score: 0 (0 votes cast)
> Fred at January 14, 2005 03:14 PM
Well, that clears it up. Instead of a vote, it was an exit poll.
16. Posted by basil | January 14, 2005 4:25 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 16:25
17. Posted by Henry | January 14, 2005 7:24 PM | Score: 0 (0 votes cast)
Actually, its not faulty reading on our part, its faulty writing on theirs, they didn't make it clear.
17. Posted by Henry | January 14, 2005 7:24 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 19:24
18. Posted by -S- | January 14, 2005 7:39 PM | Score: 0 (0 votes cast)
WHAT'S WITH THE MAIN WIZBANG PAGE?
Only a background, nothing more...for a while now.
18. Posted by -S- | January 14, 2005 7:39 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 19:39
19. Posted by Jinx McHue | January 14, 2005 8:19 PM | Score: 0 (0 votes cast)
Of that remaining 16%, 61% use yahoo and 38% use MSN.
Does this sound right?
That is needlessly complex.
19. Posted by Jinx McHue | January 14, 2005 8:19 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 20:19
20. Posted by r.a. | January 14, 2005 9:20 PM | Score: 0 (0 votes cast)
thanks fred.
20. Posted by r.a. | January 14, 2005 9:20 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 21:20
21. Posted by OhianForbush | January 14, 2005 10:49 PM | Score: 0 (0 votes cast)
On the topic of critical/skeptical thinking, any thoughts on whether this Oral Account from a Disabled Widow in America is believable?
I think it's too cute by half!
21. Posted by OhianForbush | January 14, 2005 10:49 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 22:49
22. Posted by Jeffery | January 14, 2005 11:28 PM | Score: 0 (0 votes cast)
Maybe if Keynote Systems could come up with better stats their stocks might do better in the long run... market trend (KEYN)
22. Posted by Jeffery | January 14, 2005 11:28 PM |
Score: 0 (0 votes cast)
Posted on January 14, 2005 23:28
23. Posted by Michael Diaz | January 17, 2005 10:29 AM | Score: 0 (0 votes cast)
Actually, the increase would be 67% (2/3). You measure rates of increase or decrease against the original amount and not the final
You are correct on that one. Sorry, brain fart.
23. Posted by Michael Diaz | January 17, 2005 10:29 AM |
Score: 0 (0 votes cast)
Posted on January 17, 2005 10:29