Chart-toppers and musical hits will be discussed in a presentation this week, but it won’t be coming from the music department.
Dr. William Carroll will give a talk, “Statistics and the Shirelles: How Physical Sciences Thinking Informs Popular Music Analytics,” as part of the Frontiers in Chemistry Lecture Series Wednesday, March 4, at 7 p.m. in Memorial Field House Room 2100.
Carroll has served as chair of the board of directors and president of the American Chemical Society. In addition, he is a Fellow of the Royal Society of Chemistry, a member of the advisory board for the Tulane School of Science and Engineering, vice president of industry issues for Occidental Chemical Corp., and adjunct professor of chemistry at Indiana University.
“To have someone of his stature coming to The University of Toledo is a huge deal,” said Dr. Jim Zubricky, UT associate lecturer in chemistry.
The talk is based on Carroll’s Popular Music and Society paper “Not So Lonely at the Top: Billboard #1s and a New Methodology for Comparing Records, 1958-1975.” He has recently expanded his database through 1980 and will compare the history of Billboard chart songs that were popular in that era, and discuss the methodology for creating the charts.
“This is some unique research that I’ve always wanted to do that I’ve never had time for until the last couple years,” Carroll said. “I always felt that there were ways of analyzing a record’s chart behavior in order to find the strongest records in a given period of time.”
Carroll explained that it’s difficult to determine the “best” records because that’s a subjective term; but you can objectively determine which had the strongest chart history. However, you first need a basis on which to compare two songs from different eras, plus you must consider the evolution of the charts themselves.
The first issue was tackled with the area under the curve approach, Carroll said. This method was developed in the 1970s by assigning a point value to each chart slot. The difference in importance between one and two on the chart is much greater than the difference between 99 and 100 — higher slots are given greater value. Using this point system, researchers can make a bar graph with a point value for each week a song is on the chart.
“The bars have the same width — one week — but different heights. So you can calculate the area of each of those rectangles simply by adding the weekly scores together and get what we call the area under the curve,” he said.
Since this method was developed, at least six more methods have been created, used and written about, Carroll said.
While many similar studies have been done on this topic, Carroll said there was one new piece of the puzzle that he contributed to the discussion: all previous studies ignored the fact that the charts changed over time.
He explained that in the mid-1960s, about 750 songs would enter the charts in a year. But in the 1970s, that decreased to only 400 songs. Each year there are only so many spaces on a chart, meaning there is a fixed amount of space occupied by songs. If you divide that space by a higher number of songs, you get a lower average score for the songs in that era.
“The score of an average song in 1967 would be significantly lower than in 1977, simply because of the denominator — the number of songs entering the charts,” he said. “So if you can’t rely on an average song having the same score, then you’re going to have trouble comparing one era to another an also determining what constitutes a really above-average song.”
What Carroll concluded was that researchers have to compare songs that were on the chart contemporaneously. Those with the highest score compared to records on the chart at the same time are ranked strongest.
However, the more important thing he wants viewers to take away from his talk is that the tools taught in chemistry class can be applied to other situations.
“The moral of the story is that for people who are science students — they go to laboratories, they think chemistry professors are just teaching you this stuff so you can do labs,” he said. “But it’s not just that. We teach you methods of visualization, of graphing, of statistics, because these are ways of dealing with data. It’s not just a chemistry experiment; it can be a popular music chart. You’ll use similar tools for any kind of data.”
Zubricky said that UT tries to instill this philosophy in chemistry students. Interdepartmental research relationships are constantly being formed between students and faculty, he said.
“Even though these are ideas we teach in chemistry classes, these can be applied to everyday problems,” Zubricky said. “The stuff that we’re talking about in general chemistry, the same kind of critical thinking skills that are applied in real-life issues. That’s one of the things I really stress in my classes.”
For more information on the free, public event, contact Zubricky at email@example.com.