Embargoed Radio Markets
I really enjoy your column and your wit. Can you please explain why certain Arbitron market's ratings are embargoed? Who determines if a market is embargoed? Why would a company want to embargo their ratings? - Anonymous
Anon: I'm glad to hear that you enjoy my wit. Flattery will get you a long way. I am not in the position to speak for Arbitron, so I sent your question to Bob Michaels, VP, Radio Programming Services. I want to thank him for his time, and this is what he said:
Embargoed markets are made as a business decision for a variety of reasons. But first, some a few comments on Arbitron data.
As you know, Arbitron makes all of our revenue based on radio stations, advertisers, and advertising agencies subscribing to our service. In most markets, the majority of radio stations subscribe to our service. However, there are markets where subscription levels are low, or there have been reports from our subscribing stations that non-subscribers are getting some top-line ratings data and using them. This violates our copyrights on our data and having them readily available (even on a Persons 12+ level) doesn't help the situation. So in these cases, the market is embargoed and none of the estimates for that market are released to the trade press.
Our subscribing radio stations in low-support markets or markets with a station accused of using our data without an agreement with Arbitron typically are the ones who want the data embargoed. Why should they pay for our service and allow a non-subscriber easy access to top-line results that they can steal and use against them? It's not fair to our subscribers or to us.
So, the market is embargoed based on requests either from our subscribers in that market or from our marketing staff.
English - Amazing Stuff
I received this in an email from a friend. Do you know if it’s real?
Aoccdrnig to rscheearch at Cmabrigde Uinervtisy, it deosn’t mttaer in waht oredr the ltteers in a wrod are, the olny improetnt tihng is taht the frist and lsat ltteer be at the rghit pclae. The rset can be a total mses and you can sitll raed it wouthit porbelm. Tihs is bcuseae the huamn mnid deos not raed ervey lteter by istlef, but the wrod as a wlohe.
Amzanig huh? - Anonymous
Anon: What is amazing is how fast this item has made it around the Internet. I can’t find a source for this passage. What I did find is that there are several versions, most of which change the name of the university where the study was supposedly conducted.
However, regardless about the authenticity of the passage, it seems to be true for at least some people. I gave the passage to a few English-speaking people and they read it as fast as they did when the words were spelled correctly.
Can you imagine trying to learn English with this passage?
Doc: I was paying my bills today and noticed five little vertical bars to the left of where the stamp goes. I noticed the same bars on all the return envelopes for the bills I was paying. What are those things? - Anonymous
Anon: I know I have noticed those same bars, but never thought anything about them. I just looked at my pile of bills, and all the return envelopes have the same thing. I didn’t know the answer to your question, but I found it on the U.S. Post Office website. The envelope below shows what you’re talking about so other readers will know. I copied it from the Post Office website site and tinted it blue so it wouldn’t fade into the background. I also put an arrow pointing to the five vertical bars:
This is what the Post Office says about the bars:
The facing identification mark (FIM) is a pattern of vertical bars printed in the upper right portion of a mailpiece, to the left of the postage area. A FIM pattern is essentially a nine-bit code consisting of bars and no-bar placeholders. The presence of a bar can be considered a binary "1" (one); the absence of a bar, a binary "0" (zero).
The FIM uses a code that tells automated processing equipment some of what it needs to know to do its job. The FIM allows automatic facing (orientation) of the mail for cancellation (postmarking). The FIM also identifies reply mail that bears a preprinted barcode. Barcoded mail is then routed directly to a high-speed barcode sorter, bypassing slower manual sorting or optical character reader (OCR) processing.
If you’re interested, there is a lot more information about envelope marks in a Post Office article titled, Designing Letter and Reply Mail, Publication 25, June 2003.
Doc: Do you know if the product
called "Enzyte" really works? - Anonymous
Anon: I'll have to admit that I never expected to receive a question about this topic. But that's OK. However, as you already noticed, I made some significant changes to your question to coincide with my "G" rating for the column.
I don't have to go into a lot of detail here because there is a wealth of information on the Internet about this product. Check out some of the articles in this search. The Wikipedia article is a good summary of the product.
Doc, you're the best at answering random questions. I'm sure you won't let me down this time. Can you tell me the title and artist of the song that was playing on ‘ER,' when Lucy was stabbed? Thanks so much. - Anonymous
Anon: First, I have to tell you that I have never seen "ER," so I have no idea who Lucy is or that she was killed. I had to rely on the Internet for this one. Here is what I found, but I can't guarantee that it's correct. If anyone who reads this column can correct me, please do.
The episode in which Lucy was killed was titled, "Be Still My Heart," which aired on February 17, 2000. When Lucy is stabbed, it is reported that the music playing is "Battle Flag" by the Lo-Fidelity Allstars.
You can find out more about the Lo-Fidelity Allstars by going to: www.allmusic.com. Type in the group's name in the search area at the top of the home page.
Error (Sampling) Calculator
Dr. Wimmer: A while ago you posted a program that when you fill in the number of responders it gave you the possibility of error + or -. Could you please post the program again? Thanks for your valuable information. - Dale
Dale: The program you’re talking about is on my business website. Just click here to go there and you’ll find the error rate calculator on the home page: Sampling Error Calculator.
Let me know if you need any other information and thanks for the comment about the column. I’m glad you enjoy it.
Error in Research
Dear Dr. Research: This is not the question, just some prior explanation. I’m an afternoon jock from Hungary, EU, so my English leaves a little to be desired. I’m also unfamiliar with the English radio terminology so please edit my question as you always do.
Part 1. Are you sure that people DO know what they want? I read your archives, which are useful, but are the conclusions of any survey EXACT? We conducted a survey that showed that most of our listeners are listening to our station in the car. They said they’d like to hear more traffic reports. But we know that people first listen to radio because of the MUSIC. Should they want more traffic reports, they could switch to the local news station, or the traffic station.
Part 2. I think if you propose a question like: “What would you like us to improve?” they will make up an answer. If there is something wrong with the radio station is it possible that we shouldn’t ask the listeners (since they know nothing about radio) but ask professionals? Who knows better?
Part 3. If you ask me what car to design for me, I would say, “Let it be faster.” But if you produced a fast car, I won’t buy it because of its high gas consumption. People who are not into radio don’t have a clue what they would like. Shouldn’t we rely on the experts? - Zo
Zo: I’m happy to have readers from Hungary. Thanks.
First, you did a great job with your English, so don’t worry about that. And I’m glad you didn’t ask me to respond in Magyar, your language. (I hope I spelled that correctly.) I’m also glad you found the “Research Doctor Archives” useful (see www.rogerwimmer.com).
On to your questions . . .
Part 1: Yes, I am sure that people do know what they want. However, I’m also sure that most people don’t know how to describe what they want. And that’s where research enters the picture.
Most researchers learn very early in their careers that asking respondents a question like, “What do you want?” does not generate good information. Therefore, instead of asking, “What do you want,” researchers ask indirect questions, such as, “How important is . . .” Notice that I used the plural “questions” because when you’re trying to find out what radio listeners (or any consumers) want, it is always best to attack the problem with a variety of questions.
A second type of question often used in radio is to follow the “How important is . . .” question with a question to find out if the listeners favorite radio station has enough of the item. This “enough” question allows you to produce what is called a net percent “Want more.” Over the years, I have developed minimum expectations for percentages of “Want more” for programming elements. (I’m not going to go into more detail about this approach because that’s what I do for a living.)
Anyway, the responses to a variety of questions, then, are used to interpret what the respondents want—you just don’t take information from only one question.
Now in reference to your traffic reports question . . . You said that your listeners asked for more traffic reports. You didn’t say that you asked them how important traffic reports are to them, however, my guess is that they would be considered important if the listeners asked for them in an open-ended question.
Why give more traffic reports on a music station? Because most people who listen to radio for music also listen to radio for other things like news, traffic, and weather. This was a mistake made by many radio stations (particularly Soft AC) in the United States in the past few decades. These radio stations are music boxes, but the listeners also wanted information. The listeners were (and still are) forced to go to other radio stations to get the basic information they need. It makes no sense to force listeners to leave your radio station to get information you can provide.
Now, I’m not saying that a music radio station should have traffic reports every 5 minutes, but there is no reason why a music radio station can’t have a few traffic reports every hour—even if the report is brief to include only information about serious traffic problems.
Part 2: Once again, most researchers wouldn’t ask the question, “What would you like us to improve?” Instead, the researchers would ask the listeners to rate a variety of programming elements on the radio station using a 1-10 scale. The ratings will indicate what the listeners think should be improved.
You ask whether professionals or listeners should be asked questions, and say, “Who knows better?” If you have been reading this column for a while, you’ll know that my answer is: Ask the listeners. The are the audience. They are the consumers of the product. The professionals (PDs, consultants, etc.) are not the radio station’s consumers and their opinions about what the audience wants/needs are only that—opinions.
A good radio professional is a person who listens to the audience, synthesizes information from a variety of sources, and makes suggestions and decisions based on this information, not what he/she thinks is best. The task of the professional is to understand what the audience wants and give it to them in a creative and exciting way.
Who knows better? The listeners. Who knows how to provide the items in a creative and exciting way? The professionals.
Part 3. See the answer to Part 2. If I conducted research about the type of car you wanted, you wouldn’t be asked only one question. You would also be asked to rate other things like gas consumption, ease of maintenance, and more.
Error - Deviation
Whenever we have a bad trend, my boss says "don't worry it's within statistical deviation." What is he talking about and when should we be concerned? – ADP dude
ADP: Your boss is probably referring to sampling error as opposed to measurement error or random error. So, for example, assume that your last trend was a 6.5 and the new trend is 5.9. Without knowing the sample size involved in your trends, my guess is that these two numbers are basically the same when you consider sampling error. That's what your boss is referring to.
In most cases, and this will differ from market to market because of different sample sizes, if you get a difference of plus or minus 2 points, you probably can assume that your station has gone up or down a bit. I need to emphasize that this number varies by sample size. You have to know the sample size in order to determine the range of sampling error. I can't do that for you now since I don't have the data.
Error - Maximum
I was discussing research error with my PD, and he asked what the maximum error can be for a research study that uses a sample of people. I didn't know the answer. Can you help? - Anonymous
Anon: All research includes some type of error, even a census where all elements of the population are tested, measured, or analyzed in one way or another. When it comes to the potential maximum error associated with research using a sample, your answer is 100%.
Here is a simple example . . . assume that you conduct a project where you think the respondents are 18-34 year-old females, but the respondents (mistakenly) were 18-34 year-old males. You have 100% error.
I would assume, then, that your next question is whether Arbitron could have 100% error in its surveys. Theoretically, the answer is "yes," but the company would really have to mess up badly. I doubt that would ever happen.
Error: Type I and II
Would you explain Type I and Type II error in research? - Anonymous
Anon: My co-author, Joe Dominick, and I recently finished the latest edition of our textbook, so to make this faster for me, I'm going to take our discussion from the new manuscript. I have edited a lot, so if you need additional information, please refer to the book (Mass Media Research: An Introduction, 10th Edition) or click here to order it: Mass Media Research.
Two types of error relevant to hypothesis testing in research are called Type I error and Type II error. In somewhat technical terms, Type I error is the rejection of a null hypothesis that should be accepted, and Type II error is the acceptance of a null hypothesis that should be rejected. (The null hypothesis is the hypothesis of no difference and is actually what is tested in an experimental study. For example, an hypothesis may be: There is a statistically significant relationship between education level and radio listening. The alternative, or null hypothesis is: There is NO statistically significant relationship between education level and radio listening.)
OK . . . whenever we conduct research, there is a chance that we'll get data that support our alternative hypothesis simply by luck or random accident, not because our alternative hypothesis is actually true. When this happens we incorrectly reject a null hypothesis that actually should be retained. This is Type I error. Similarly, random error could work the other way and we could end up with data that do not support our alternative hypothesis even though our alternative hypothesis is actually true. When this happens, we fail to reject a null hypothesis that should be rejected. This is Type II error.
An analogy may help. Assume that you think you have a problem with your car's steering and you take your car to a mechanic. Further, assume for this example that you know there is actually nothing wrong with your car's steering. Ultimately, the mechanic can tell you one of two things: (1) There is a problem with the steering, or (2) There is no problem with the steering. If the mechanic says "No problem," that's great. That's the correct decision. However, the mechanic might misread test results and say, "There is a problem." This is an error. In this situation, we have falsely rejected a true null hypothesis ("There's nothing wrong with the steering.") and have committed a Type I error.
Now suppose there really is something wrong with the steering. If the mechanic says, "You have a problem," that's great. That's a correct decision. However, the mechanic may have a bad day, overlook some test results, and say, "There's no problem." This is an error. We have failed to reject a false null hypothesis. We have committed a Type II error.
The probability of making a Type I error (sometimes called alpha error) is equal to the established level of significance and is therefore under the direct control of the researcher. That is, to reduce the probability of Type I error, the researcher can simply set the level of significance closer to zero.
Type II error, often called beta error, is a bit more difficult to conceptualize. The researcher does not have direct control over Type II error; instead, Type II error is controlled, though indirectly, by the design of the research. In addition, the level of Type II error is inversely proportional to the level of Type I error: As Type I error decreases, Type II error increases, and vice versa. The potential magnitude of Type II error depends in part on the probability level and in part on which of the possible alternative hypotheses actually is true.
Because many people have a problem with the null hypothesis in research, I developed a different way to describe Type I and II errors without the null hypothesis.
Consider it this way . . .
Type I Error: Significant difference found where none exists.
Type II Error: No significant difference found where one exists.
Let's relate this to radio. Assume a researcher conducts an experiment to find out if your radio station's morning show is contributing to your decline in Arbitron ratings:
Type I: The researcher concludes that your morning show IS a problem when it is NOT.
Type II: The researcher concludes that your morning show is NOT a problem when it IS.
Dear Doc: Why isn’t Top 40 radio as good as it used to be and as good as the European stations? I remember when USA Top 40 was great and you could hear just about everything like dance, rock, and country, and everything was so well mixed. While today’s Top 40 in the USA isn’t quite right, it’s so limited and it seems if they are scared or something. Only two stations are found in the top markets—Z-100 in New York and Kiss FM in LA. However, if you listen to European radio Top 40, it’ just great you get everything—well blended and very entertaining, such as Capital FM in London, BBC Radio 1, and BBC Radio2, to mention just a few. Wouldn’t this format work in the states instead of so many R&B formats? - Donnie
Donnie: I have two answers to your question.
Answer 1: I follow the tenets (rules) of scientific research, and one of those rules is that I don’t make decisions (or answer questions) based on the perceptions of a sample of one. The error associated with your perceptions of Top 40 in the USA and in Europe could be as high as 100%.
Now, this doesn’t mean that you aren’t entitled to your opinions. Based on the type of comments you make and the words you use, I would assume you have a good knowledge of Top 40 radio. You may be 100% correct, but I can’t get involved in a long discussion because the perceptions you have may not be shared with anyone else.
Answer 2: But let’s assume that you are 100% correct in your perceptions. Then why aren’t Top 40 radio stations in America doing the “right” things?
It could be that they are doing the right things. It could be that American companies have tested the approaches used in Europe and found that American Top 40 listeners don’t like the approach. It also could be that American companies don’t know anything about the European stations because the people who run the companies have their heads in the sand.
If you are correct, my guess is that someone will “discover” what European Top 40 radio stations are doing, find out through research that the approach is good, and then copy the approach in America. (They will then claim to have developed the format.) Most radio station companies in America do not take chances. There is too much as stake. And that’s unfortunate for listeners.
Excel Formula Question
Hi Doc: I'm doing some work in Excel and I have "Googled" to no avail, so I'm hoping you can help. Why do some formulas have multiple parentheses? I understand that whatever is in parentheses is calculated first, but why do some formulas look like
What is the function of two and three parentheses in various places? Thanks! - Anonymous
Anon: Bear with me here . . .
Parentheses in any formula identify a unique operation that must be computed alone. For example, let's say that someone says, "Compute this formula: one plus two times three plus four." What's the answer? While it may sound simple, there are actually three answers if the person doesn't provide you with additional information. Here are the answers:
1+2*3+4 = 11
(Multiplication (2*3) is computed first, then "1" and "4" are added.)
1+2*3+4 = 13
(Steps followed in order: 1+2 = 3; 3*3 = 9; 9+4 = 13)
(1+2)*(3+4) = 21
(Addition for each pair is computed first: 1+2 =3; 3+4 = 7; 3*7 = 21)
The use of parentheses in your example is the same as the third answer to the example I just provided. The parentheses tell the computer the elements that are stand-alone items, and which items are considered together or as a whole.
To begin, the formula in your example says this:
If and only if cell L12 contains the numbers 28, 29, or 30, enter the number .283 in the cell where your formula is contained. If L12 contains any other number, then enter a "0" in the cell where the formula is contained.
It's necessary to have parentheses to instruct Excel (or anything/anyone else) how to compute the number you want. So, here are the steps:
(L12>=28). This is a separate item and it requires its own parentheses. If and only if L12 is equal to or greater than 28
(L12<=30). This is another separate item and it requires its own parentheses. If an only if L12 is less than or equal to 30
However, the first two steps must be considered as a whole after each is considered separately, so it's necessary to use another set of parentheses: ((L12>=28),(L12<=30))
But there is another command that is included—the AND word. The "IF" command is a stand-alone item, but you are saying, "IF AND ONLY IF." The formula adds the word "AND," and that means another set of parentheses because the first three steps must be considered (computed) as a whole. Here is that part of the formula, and you should see why there is a need for the parentheses for each part: (AND((L12>=28),(L12<=30))). This is the first part of the "IF" command, but the computation isn't done yet.
So far, the formula only refers to what is entered in cell L12 (Greater than or equal to 28, and Less than or equal to 30. In other words, it isolates/identifies ONLY the numbers 28, 29, and 30. Any other number is irrelevant. But the formula as it stands now doesn't tell the computer what to do. That comes at the remainder of the formula.
OK, now the formula tells the computer what to do if a number is entered into cell L12 with the last portion of the formula—283,0. This part says, "If the number in L12 is 28, 29, or 30, enter the number .238 into the cell where the formula is located. If any number other than 28, 29, or 30 is entered into cell L12, then enter a "0" into the cell where the formula is located. That's all there is to it.
But . . . there is one last thing. Remember that you are using the "IF" command. The word "IF" is a stand-alone item, which means that whatever follows the word must be a formula of some kind. To tell the computer what the complete formula is, it's necessary to have one last set of parentheses — the last set of parentheses identify everything contained within is the entire formula: =IF((AND((L12>=28),(L12<=30))),.283,0)
Hint: There must always be an equal number of "left-leaning" and "right-leaning" parentheses (my terms). In your formula, you can see that there are five of these "(" and five of these ")." If you construct a formula without an equal number of lefties and righties (my terms), Excel will display an error message.
You're the patron saint of radio folks! Based on your studies, do listeners really care about "first heard" songs on the radio? If yes, how should we maximize this advantage? Thanks for the favor. - Anonymous
Anon: Patron saint? Thanks. You made me smile and now I have something new to put on my resume.
Since I have investigated this question many times, I can say that by itself, playing music first on a radio is not a primary reason why listeners choose a radio station. But the element does add to the uniqueness of a radio station, and combined with other elements, such as "we don't talk over the music," it helps form an overall image of your radio station.
How should you maximize the advantage? Just like anything else you do that's unique…tell the listeners. You don't have to say it before or after every song because playing songs first is not the main reason why listeners select radio stations, but tell the listeners every now and then (you probably already do) that, "You heard it here first," or "We play the new songs first."
The best way to maximize the perception of any element on your radio station is to use persuasion, and the number one rule in persuasion is repetition of the message.
Execution - Capital Punishment
Is there any research on the psychological effects of capital punishment on the person who is put to death? - Anonymous
Anon: Hmm…sometimes I think I have seen just about every question that could be asked, but you really got me on this one.
After reading your question several times with a dumbfounded expression on my face, I thought of a few things:
It's interesting how news headlines generate questions.
Look very closely at your question. In order to accomplish a study as you suggest, the person who is put to death would have to be interviewed. As far as I know, unless you contact the Tarot Card reading lady who advertises on TV, or Shirley McLain, there is no way to interview dead people.
Let's see…how do I say this without sounding too crude…but, what would we do with such information if it were possible to collect? My guess is that we would have a bunch of responses like, "Damn," or "That hurts."
I think I'll take the rest of the day off.
Click Here for Additional E Questions
Roger D. Wimmer, Ph.D. - All Content ©2018 - Wimmer Research All Rights Reserved