Editing Seminars for HBCU Professors|
The Dow Jones Newspaper Fund revived its free Copy Editing Seminar for professors from historically black colleges and universities in Spring 2003 with a session in May in Princeton. The program, initiated in 1994, has brought leading editing professors and newspaper editors together with HBCU campus media advisers and journalism professors to provide insights into teaching that all-important skill. Professors' travel, meal and lodging expenses are paid by the Fund. There is no registration or tuition fee.
Wanda Peters, adviser to the Gramblinite of
Grambling (La.) State University, listens
to a presentation by William Connolly, a retired editor
from The New York Times.
The next seminar will be held in Fall 2004.
Some sample materials appear below.
NOTES ON NUMBERS
Visiting instructor: Dr. Rusty Todd, University of Texas at Austin, DJNF Board Member
A paragraph says, “Twelve ships lay at anchor inside the blockade imposed by U.S. Navy vessels.”
Later, the story notes, “Of the ships trapped in the harbor, five are from the former Soviet state of Georgia; two are from Honduras; and four are domestic ships based in Havana.”
Sounds good, eh? But take the time to add up the ships in the later graf—where’s the twelfth ship from? Always sum the numbers to make sure nothing’s been left out.
At midterm, you have a grade of 83 percent in copy editing. Going into the final, you have raised your average to 88 percent. How much has your grade increased?
Your grade is 5 percentage points higher, 88 minus 83. It is not 5 percent higher.
You have a story that says 59 percent of Americans approve of canned ham and 33 percent disapprove of canned ham, while 13 percent don’t care. Fifty-nine plus 33 plus 13 equals 105. That’s 5 percent too much! Always sum percentages, particularly in polling results, to see if they add up to 100. If they don’t, ask yourself whether something’s missing.
You’re doing an article about how many more auto accidents there are this year over last year. In 1997, there were 415 vehicle accidents; in 1998, 553. What is the percentage change? To find it, subract the old value from the new value and divide the result by the old value.
Accidents up 33.3%
It works the same way when numbers decrease from year to year—you just end up dealing with a negative number.
Hello!!!Say you have $100, and you put it on two-year deposit at the rate of 10 percent a year. Ten percent of $100 is 10 bucks, right? So you get 10 bucks a year, right? After two years, you’ll have your original $100 back, plus $20 interest, right? WRONG!
At the end of a year, you’ll have $110. Interest in the second year is computed on that sum. Ten percent of $110 is $11. At the end of two years, you’ll have $121, not $120. That extra dollar is the result of compounding.
The formula for compound interest is
Vn is the value at the end of n years.
P is the principal amount you start with.
I is the interest rate per year.
N is the number of years
In the case of our example:
Vn equals 100(1+.10)2
That’s 1.1 squared, which is 1.21. And 1.21 times $100 is $121.
REAL VS. NOMINAL RATES
In 1998, your salary was $35,000. You have received a 2 percent raise for 1999 to a salary of $35,700. In nominal terms, you got a $700 raise.
But do you have 2 percent more buying power? No, because of inflation. If inflation is also 2 percent, it takes 2 percent more money to buy the same stuff. You got a 2 percent raise, so in terms of what your money will buy, you have gone nowhere. Your raise in real terms is zero percent.
Say inflation was 1 percent instead of 2. You received a 2 percent raise, minus 1 percent inflation, which is a 1 percent real raise. You have 1 percent more buying power.
Reporters use tables of the federal Consumer Price Index tables to convert nominal figures to real figures over various time periods. The index is compiled by the Bureau of Labor Statistics in the Department of Commerce. Go to http://www.bls.gov/cpi/, where you'll find historical CPI tables and a calculator that will do all the math for you.
All polls are not alike. To be representative, a poll must follow rigorous routines in interviewing and data interpretation. Here are some things to look for:
Who paid for the poll? If it’s a special-interest group, be suspicious.
How many respondents were polled? Small sample sizes mean polling inaccuracy, but a sample of 750 or 1,000 can be sufficient.
Was the poll random and representative? Did everyone have an equal chance of being chosen?
How was the poll conducted? Telephone? Person-to-person? Return postcard? Web?
What is the poll’s margin or error? This is the line that reads “accurate to within plus or minus 5 percentage points.” If the margin of error is plus/minus 5 points and the poll shows 45 percent like chicken livers, the actual result could be anywhere between 40 and 50 percent.
What is the confidence interval? This statistic, rarely quoted, reflects how sure pollsters are about their results. If the interval is .05, there’s only a 5 percent chance that the polling results are outside the margin of error. That is, we would be 95 percent confident that between 40 and 50 percent of the population likes chicken livers.
Can you look at the actual questionnaire that was used in the poll? Are questions loaded?
Loaded: How disgusting do you find chicken livers?
Unloaded: Do you like chicken livers?