Facebook Article Grammar?

Doc:  I was reading about the problem with Facebook changing its "Terms of Service" and found an article on the Internet I thought you would enjoy. Here is the title and the first paragraph:


Facebook Yields to User Outrage: Reverts to Old TOS


Ian Paul, PC World


Feb 18, 2009 5:55 am


Amid user revolt Facebook backed away from a newly implemented terms of service that many considered a privacy violation. Last night Facebook chief executive Mark Zuckerberg said the company would reverted back to its previous version of its terms of service that "everybody can understand."


What do you think about that? - Ray


Ray: That's a good one.  Since "revert" means "go back," the words "revert back" essentially mean "go back back."  When the word "revert" is used, there is no need to use the word, "back."  That's what revert means.  I'm waiting for someone to write, "revert forward."


However, the additional problem in the article is that the writer wrote, ". . . would reverted back . . ."  I can understand the error, "revert back," but I don't understand "would reverted back."

Facebook Article Grammar Again

Doc:  You recently mentioned the use of "revert back" in an article about Facebook on the Internet.  I was reading an article on Coedmagazine.com about the same thing and there was an article by someone named Andrew - Hunter College.  In his discussion about Facebook's policy, he said:


"Well, needless to say, everyone freaked the hell out out. And after two days of protest, Facebook decided to revert back to their previous terms of service."


Not only does Andrew use "revert back," but did you notice the first sentence?  The word "out" is repeated. - Anonymous


Anon: Thanks for sending the example.  The use of "revert back" is rather silly.  As I mentioned earlier, the word "revert" means "to go back," so "revert back" says "to go back back" or "go back back."


Hey, maybe Andrew knows this and just likes to repeat things—"back back," "out out"—seems to be a pattern.

Factor Analysis

Roger, great column. Keep up the good work. My ownership group is doing a study to find out the most viable format that is currently absent from our market. My boss said they were testing 12 different formats, and would use a technique called factor analysis to reduce the number from 12 to only a few. I hit the books but don't understand what "factor analysis" is all about. Can you explain it to me? - Anonymous

Anon: Thanks for the comment about the column. I try my best.

Factor analysis is a multivariate statistical technique that searches for underlying dimensions in a data set. For example, you could conduct a music test with 400 songs and then use factor analysis to find out if there are songs that group together (that is, songs that the respondents rate similarly). You could then name the groups (factors) to identify them . . . maybe something like "ballads," "romantic," and so on. The names depend on the type of music in the test.

A PD could use this information to help decide which songs to play. For example, in one set, a PD could play songs from the same category (factor) to produce a specific mood of similarly rated songs. In another set, the PD could select songs from different factors to produce a mood of variety. A factor analysis for music tests is a great technique because it shows how listeners group songs together . . . and they don't even know they're doing it.

That's a very simple explanation of factor analysis—it searches for underlying similarities in a data set.

Now to your question . . . I thought about this for a while and can't figure out how factor analysis can be used effectively to search for a format hole. Your boss obviously knows something I don't because I can't figure out what types of variables (questions) that could be asked to produce format availabilities. I can understand how factor analysis could be used to help define an Ideal radio station, but not to uncover a format hole. I believe there are too many unrelated variables (questions) that need to be considered. More later.


A format search, if done correctly, should gather information in three areas: (1) format ratings of some type to determine which formats are liked and which are not liked—these should be looked at with Z-scores because there will be differences for males, females, and various age cells; (2) a potential score for each format that is computed by a linear combination of variables (adding together several different ratings and percentages) that is adjusted by subtracting "penalties" for radio stations already in the format and a few other things; and (3) a projection of the potential audience using a reverse cume-to-fan conversion formula. This may sound complicated in words, but the procedure is very, very simple. A factor analysis can't do these things—at least in any way that I know about.

In addition, a major part of a format search is that the analysis must consider sex and several different age cells. You may interview 500 people 18-54, but the format hole may not be with 18-54. The hole may exist with 26-48 or some other age group. If you conduct a factor analysis with the 18-54 year-olds, you will not get information for unique cells. In addition, if you did conduct a factor analysis with the 26-48 year-olds in your study, you wouldn't have a large enough sample to do it. A format search should not be conducted with standard age cells (18-24, 25-34, etc.). You
must use exact ages so you can identify a unique demographic (if necessary). You can't use standard age divisions or you may miss important information.

Another problem with using factor analysis for a format search is that it doesn't follow Occam's Razor (the simplest approach is the best). Factor analysis makes a simple univariate procedure unnecessarily complicated. In addition, factor analysis requires the researcher (or your boss) to make several subjective decisions that aren't necessary in univariate procedures (the three areas that need to be addressed that I mentioned earlier). I won't go into these decisions here because I believe it's beyond the scope of your question.

Finally, if you test 12 formats in your study, the hole may not be one of the 12. The hole may actually be a hybrid of two or more of the formats combined—or it may be one that you didn't test (that needs to be part of your study too). A factor analysis won't show you that, but simple crosstabs will.

Now, don't get me wrong about factor analysis. The method is a very powerful technique if used correctly. I just don't see how it can be used to search for a format hole. If there is a way, and it's easier than using a three-step univariate procedure, then I'm willing to learn.

Factor Analysis Revisited

Thanks for the answer. When I mentioned that we were researching 12 different formats and using factor analysis to shorten the list, I should have expanded how my company was conducting the research. My boss tells me that the 12 formats being considered are each being represented by a montage of music and that they are really 12 different shades of gray. For example, Grunge Rock is represented by a montage of Pearl Jam/Nirvana/Sound Garden, 80s Hair Bands is represented by Winger/Cinderella/Warrant, Pissed-Off Women is represented by Indigo Girls, Melissa Etheridge, Sarah McLachlan, etc. The respondents rate each of the 12 montages on some sort of scale and then factor analysis is performed on scores. If I know the scores for each montage, what will factor analysis tell me? - Dusty

Dusty: You're welcome for the answer. Here's the next one . . .

As you explain the procedure, the factor analysis will not "shorten your list" in the way I think you think it will (enough "thinks" there?). What you will get is each montage's contribution to the factors that "fall out." This would be easier for me to explain in person, but the factor analysis will show, for example, that Factor 1 is comprised of people who like Grunge Rock, Hair Bands, and Pissed-off Women (those are the only examples you gave me). In that sense, the factor analysis will show you if your 12 montages can be represented by a smaller group of montages.

In one sense, I guess, you will be shortening your list of montages, but the factor analysis will not tell you the desire for the individual formats, and that's what you're after. You might use the factor analysis data to better understand the inter-relationship of the montages, but it alone will not provide you with format availabilities in your market.

The psychologist Abraham Kaplan came up with something known as the Law of the Hammer, which is: "If you give a child a hammer, he/she will eventually find something to hit." I paraphrased this to Wimmer's Law of the Statistic, which is: "Give a researcher a statistic and he/she will eventually find something to study." Factor analysis is a great multivariate statistical procedure, but it isn't correct for all situations. From what you say about your format search, the method is not being used correctly. You'll find the relationship among your montages, but not your format availabilities.

If the factor analysis is being used to uncover the underlying similarities of the formats and is only one step in the process of your format search, then it's OK. If the factor analysis is the only statistic used to find a format, then there is a problem.

Familiar Music

Sometimes in radio, the obvious answer might be incorrect. With that in mind, when talking about a mainstream AC station (one with few current songs), do listeners say that they want to hear songs that they know and with which they can sing along? One of the stations in my market plays a bunch of unfamiliar 80's songs. Is this such a smart idea? I know, I know--unfamiliar to me doesn't equal to unfamiliar to everyone else. Still, I have discussed this opinion with one of the staff members and he agreed with me. Your thoughts please. Thanks much. – Anonymous


Anon: If you have been reading my column for a while, you should be able to guess that I will say this: You need to ask the listeners in your market. My opinion isn’t important.


However, I will add that I have asked your questions to listeners for the past 20+ years. What I usually find is that most people say they like to hear new music, but in reality, they don’t when "promise vs. performance" is considered. That is, when radio stations play a lot of unfamiliar music, the listeners complain that they don’t listen to the radio station a lot because the music is unfamiliar. Uh, what?


Well, you’re faced with a Catch 22. People like to hear familiar music, but how can songs become familiar if radio stations don’t play the new stuff? Well, they do. What I see most often with successful AC radio stations is that they let the other radio stations in the market play the new stuff. If a new song becomes popular, they add it.


Overall, the research (and Arbitron numbers) tends to lean with the side of not playing a lot of unfamiliar music. But there are always exceptions. How does the station in your market that plays a bunch of 80s' songs perform in Arbitron? If the radio station gets good numbers, then they must be doing something right. If the numbers suck, then it may be due to the unfamiliar music. I just don’t know. I’d have to see the Arbitron numbers and other research.

Fantastic Machine

Doc:  A friend of mine sent a video to me that I think you'll think is great.  A short note was attached that says this:




This is awesome. Turn your sound on and read the screen below before opening the video attachment.


Notice how all of the balls wind up in catcher cones. This incredible machine was built as a collaborative effort between The Robert M. Trammell Music Conservatory and the Sharon Wick School of Engineering at the University of Iowa.


Amazingly, 97% of the machines' components came from John Deere Industries and Irrigation Equipment of Bancroft, Iowa.
Right first time - Farm Equipment!


It took the team a combined 13,029 hours of set-up, alignment, calibration, and tuning before filming this video,
 but as you can see it was well worth the effort. It is now on display in the Matthew Gerhard Alumni Hall
 at the University and is already slated to be donated to the Smithsonian.

Doc . . . To see the video click on this link. - Anonymous

Anon: Thanks for sending the link.  The video is interesting, but you need to read a post from Snopes—click here.  Keep in mind that many things you read, hear, or see are bogus—especially when it comes to the Internet and/or things you receive via email.

FCC Chairman

Hello.  Of course, this question has spawned from the recent departure of Michael Powell as FCC Chair, but it seems to me that some of those considered for the position seem to be lacking in any sort of tangible media background.  Granted, I don't know many of their personal pasts, only what their current positions are. However, it seems that would be a position where a strong background in media might come in handy.  Has that been the case historically? - Anonymous


Anon:  Although I’m not an expert on the FCC, my guess is that Michael Powell may top the list in reference to questions about his qualifications to hold the FCC chair.


However, regardless of the criticisms about Powell and any other FCC commissioner or chairman, what makes you think that the US government requires expertise in an area in order to be considered (and appointed) to a leadership position?


If you’re really interested, check the list of FCC Commissioners and Chairmen and you’ll see that mass media experience is not a requirement to serve as a commissioner or chairman.

FedEx Logo

Doc:  What do you know about the arrow "hidden" in the FedEx logo?  I think it's very interesting.  Did FedEx purposely design it like that? - Richard


Richard:  The discussion about the "hidden" arrow in the FedEx logo has been going on for several years, and there are thousands of articles on the Internet discussing how clever the design is, how effective the "subliminal" message is, and more.  For readers who don't know what Richard is referring to, here is the FedEx logo:



See the arrow?  For those who don't, here is another version with the arrow outlined:



OK, so was the arrow intentional?  Is it subliminal?  I found an article in the July 8, 2002 edition of the Cincinnati Enquirer.  Among other things, the article says:


"The arrow was indeed intentional as a secondary design element," says Federal Express Corp. spokesman Jess Bunn.  "If the viewer sees it, it's a neat, interesting visual bonus. If the viewer doesn't see it, that's OK.  It's still a powerful logo.


"The arrow is intended to communicate movement, speed and the dynamic nature of our company," he said.


The blocky purple and orange, upper- and lower-case letters were created as the new company logo in 1994, when we decided to modernize our entire look," Mr. Bunn explained.


There ya go, Richard.  According to the FedEx spokesperson, the design was intentional.


Yea, sure.  With all due respect to Jess Bunn, I think the arrow is just serendipity.  The company's name is Federal Express.  If block letters are used, an arrow will naturally "appear" in "Express" because of the juxtaposition of the "E" and "X."  Any word with an "E" followed by an "X" (usually smaller than the letter "E") will create an arrow if block letters are used.


Is the logo subliminal?  No.  Was the arrow planned?  No.  Was the arrow an afterthought when people noticed it?  More than likely.  Any company with the word "Express" in its name (or even and "E" and an "X" anywhere in its name) could claim the same "hidden" arrow gimmick because the "E" and the "X" will always create an arrow with block letters.


However, as several people on Internet discussion boards point out . . . If you have never noticed the arrow and see it now, you will never be able to look at the logo again without seeing the arrow.

The FedEx Logo - Continued

Doc:  OK then, what does the arrow in ExLax signify? - Geno

Geno:  Don't make me come out there.  As I mentioned in my original answer, virtually any word with the letter "E" followed by an "X" can be made to look like an arrow.  It's simply the juxtaposition of the letters.  For example, I used your suggestion to create this rather simple design:



You want to know what the arrow in Exlax signifies?  My guess is that there is a small sign in the box and you are instructed to place it on the wall to the left of your bathroom.


OK, so that's not the answer.  The arrow doesn't signify anything.  It's just another arrow.

FedEx Follow-Up

In my never-ending battle for truth, justice, and the American way, I continued to investigate the development of the FedEx logo with the "hidden" arrow.  I'll have to say that I was wrong in assuming that the arrow in the logo was an accident.  Apparently, it wasn't.


According to a November 16, 2004 article with Lindon Leader of Leader Creative, the arrow was a conscious decision and involved a lot of work.  You can read the interview with Mr. Leader if you click here.


OK. I read the article too and I still think the arrow was something that was originally an accident and then capitalized upon once a few people noticed it.  But I could be wrong.



FedEx - It Keeps on Going . . .

Doc:  Do you have any more examples of EX words? - Anonymous

Anon:  Here are a few words for you . . .



Now . . . according to all sources, the "hidden" arrow in FedEx logo is supposed to subconsciously (or whatever) elicit a "secret" message, something like "forward" or "fast."


However, I do think there may be a way to "hide" a message in a word using the "Ex" arrow.  How is this?



I'm not sure if that is hidden, but nothing compares to these words . . .







 Hyperbolicsyllabicsesquedalymistic (Not a real word.)

I forgot . . . If you don't know what Deus Ex Machina means, click here(It's probably in every movie you have ever seen.)

Female Listeners/Female Announcers

Hi Roger:  While I realize your research primarily deals with music, I'm hoping to find out if what I perceive as a "radio myth" is actually based in some fact.


To your knowledge, have there ever been studies/tests/surveys done that are accessible showing that women listeners prefer male announcers?  Or that they don't like to hear other women on the air?  Thanks for any help. - Michele


Michele:  Hi to you too.  By the way, my research doesn't primarily deal with music.  About 90% of my work is perceptual studies and the other 10% is focus groups.  I don't do music tests.  On to your question . . .


I have a few hundred studies in my files indicating that there is little consistency in reference to what type of announcers male or female listeners prefer.  There are differences among markets, age cells, and formats.  Sometimes listeners prefer men, sometimes they prefer women, and in most situations, the listeners don't care just as long as the announcer is good (pleasant voice, talking to listeners, not at them, and so on).

Female Pipes

I am currently in a Top 10 market with great female pipes. I keep seeing the phrase "It's about content" or personality. Is it true, though, that most Top 10 market radio stations have cut the chatter down to a format with little room for personality? Among my top talent peers, the talent that makes it to the top can inject his personality even in a liner by the way he/she delivers it. How about a comment? And what do you think about a radio station that wants a great liner jock over a personality jock? Does this method really work? - Anonymous


Anon: You have opened a can of worms, so to speak.  From my experience, I would have to agree that most radio stations have cut the jock chatter and look for people who are good readers. My experience is that this has come from a combination of two items: (1) Misinterpreting research results; and (2) A lack of talented on-air people.


The misinterpretation of research results relates to managers (PDs and others) who receive research studies that say the listeners want "More music and less talk." The managers take this literally, but most listeners really say, "More music and less obnoxious/irrelevant" talk." There is a difference. Music radio listeners do want to hear a lot of music, but they also like to hear entertaining DJs who relate to them.


Over the past several years, most music radio stations have relegated DJs to copy readers (just like TV news anchors). However, I have never heard a listener say that they don't like entertaining DJs who relate to them.


I perceive the lack of available talent as a self-fulfilling prophecy. In other words, the importance of a talented personality has been diminished and therefore many young people no longer have interest in being on the air. These talented people look for other positions. If the importance of talent is de-emphasized, the number of people interested in the position will naturally shrink—the talent pool dries up. I know this is true because of the number of times I hear PDs say, "I can find good talent."


What I have seen many times is that radio is cyclical. When I was a teenager in the 1960s, the emphasis in radio was two-fold: (1) music and entertainment delivered by (2) a talented jock who talked to the listeners. My guess is that the pendulum will swing back to that approach when enough radio stations take a "risk" and provide a total package to the listeners.


Radio is not only music and shouldn't be treated that way.

Fido Puzzle - Think of a Number

Hey Doc:  Would you explain how the Fido puzzle works? - Anonymous


Anon:  I took your link to the puzzle and put it here.  For those who haven't tried the puzzle, go to the website first and try it before you read how it works.


The Fido puzzle is a simple math computation based on the number 9.  So that I don't have to reinvent the information wheel, click here for a simple explanation of how the puzzle works.


Click Here for Additional F Questions


A Arbitron B C D E F G H I J K-L M
Music-Callout Music-Auditorium N O P Q R S T U V W-X-Z Home

All Content © 2015 - Wimmer Research   All Rights Reserved