Which cities are magnets for young college graduates?

One way to answer the question is to look at the cities attracting growing numbers of young, educated people, as a chart we ran Monday did. Another way is to see which cities have a larger than average percentage of the young and college-educated: people between 25 and 34 years old with a bachelor’s degree.

Measured this way, some surprising cities stand out as skewing young, including Columbus, Baltimore and Salt Lake City.

Denver, Nashville and Portland, Ore., fall into both groups: They each have a disproportionately large and fast-growing group of such residents. In Las Vegas, San Antonio and Riverside, Calif., meanwhile, this population is growing more quickly than in most other metropolitan areas, yet they have a smaller share than any other area. Over all, in the 51 largest metropolitan areas, 5.2 percent of people are 25 to 34 years old and have a four-year college degree. Denver’s share is 7.5 percent, while Riverside’s share is 2.5 percent.

Economists have debated whether jobs follow people or whether people follow jobs. Joe Cortright, who runs City Observatory, the think tank that published the report, said that companies are increasingly locating where large numbers of young, college-educated people live, because young people are pickier about location. One reason is that men and women are both likely to work, so couples seek a place they want to live and then find jobs, as opposed to wives following their husbands’ careers. Where they choose to live matters because a large young, educated work force is the economic engine of a vibrant city.

Cities in which the growth in the population of young graduates lagged overall population growth include Atlanta, Dallas, Charlotte and Raleigh. Cities where the growth was about the same as overall population growth are Memphis, Phoenix, Austin and Richmond. Cities that are attracting this group at a much faster rate than their total population is growing are New Orleans, San Diego, Oklahoma City and Las Vegas.

When big cities like New York and San Francisco become unaffordable for young people to live, they increasingly choose smaller cities like Baltimore and Portland over the suburbs that many in their parents’ generation chose. In Portland, for instance, the number of young college graduates has grown 37 percent since 2000; they now make up 5.6 percent of the population, up from 4.8 percent a decade and a half ago.

In readers’ comments on our first article, there was some fierce city loyalty, including consternation that certain cities weren’t included in our chart. We heard most about Chicago, Seattle and Philadelphia — so for all of you, here are the details. All are popular with the young and educated.

Chicago’s population of young people has climbed 17 percent since 2000, more than Boston’s but less than New York’s, and a higher than average 6 percent of its population is in this group. Seattle’s population of young graduates climbed 27 percent; its share is 6.1 percent. Philadelphia’s climbed 22 percent, and 5.4 percent of its population is young and college-educated, slightly above average.

Television ads attacking candidates for state supreme courts for being “soft on crime” affect more than just elections. A study released Tuesday by two Emory Law School professors and the American Constitution Society, a progressive legal group, finds evidence that they also make judges less likely to rule for criminal defendants in appellate cases.

The study, which examined more than 3,000 criminal cases in 32 states from 2008 through 2013, along with tallies of television ads maintained by the Brennan Center for Justice at New York University, found that more television ads in state supreme court races meant an increased likelihood of rulings against criminal defendants. All 32 states studied have some form of voter approval of judges, either to elect them or to approve sitting judges.

The airing of 1,000 ads would have a minimal decrease on judges’ decisions favoring criminal defendants — perhaps one in 100 cases — while 10,000 ads would result in a change in a justice’s decision in 7 percent of cases. Since state supreme courts are multi-judge panels, it’s not clear how often the changes in their voting behavior would alter the overall ruling in specific cases. A 7-2 ruling against a criminal defendant has the same outcome as a 6-3 ruling.

Even though state supreme courts decide a wide range of civil and criminal cases, the authors focused on criminal cases because many of the advertisements dealt with violent crime, a more visceral issue for voters. “Voters don’t care how you vote in a tort reform case, but they do care in a rape or pedophile case,” said Joanna Shepherd, one of the authors.

The study is the first to correlate independent expenditures in judicial races with judges’ decision-making, said Ms. Shepherd, who last year wrote a report, also published by the American Constitution Society, examining political contributions by business groups in judicial races and subsequent voting by justices in cases involving business litigants.

The study also links an increase in negative advertisements against judicial candidates to the Supreme Court’s decision in Citizens United, which removed limits on independent corporate and union spending in elections. Although Citizens United has led to large increases in independent expenditure spending at the federal level, outside spending in state supreme court races has also grown. More than $24 million in independent expenditures occurred in state supreme court races during the 2011-2012 election cycle, up from $2.7 million 10 years earlier. With greater spending, state supreme court races had more negative TV ads. The authors found that 44 percent of outside group ads in 2012 were negative, a much larger share than for candidate or party spenders.

“The data show that the television campaign ads this money buys put a thumb on the scale in criminal cases, and undermine the promise of equal justice that is a cornerstone of our democracy,” said Caroline Fredrickson, the president of the American Constitution Society.

The effect on judges’ voting does not hold for every state — not all states hold elections for their supreme court seats — and in some states the effect was larger than others. Ms. Shepherd said that although the relationship was evident in many of the 32 states studied, it was not statistically significant in all of them. She cited Michigan as a state with an effect even greater than the one that held nationally. Millions of dollars were spent on television ads in 2012, and the Michigan Campaign Finance Network called the state supreme court races “the most expensive and least transparent in history.”

The practice of using “soft on crime” ads in judicial races isn’t new. In 2004, Don Blankenship, the former chief executive of Massey Energy, a coal mining company, created an outside group to influence West Virginia’s supreme court election. The group, “And for the Sake of the Kids,” spent $3 million on television ads that asserted an incumbent judge voted to free a child rapist from prison; the judge lost his race.

The study’s authors concluded that the Citizens United ruling increased the effect that outside spending had on judicial elections and voting behavior: “In the 23 states that had bans on corporate or union independent expenditures, Citizens United’s lifting of these bans is associated with a decrease in justices voting in favor of defendants,” the authors write.

Georgia’s 12th Congressional District, stretching from the South Carolina border to the middle of the state, is covered by four television markets. Because portions of those markets cross district and state borders, a lot of money is wasted broadcasting ads to people in other congressional districts and in South Carolina.

For years, this waste has been unavoidable. There are tools for making television ad buys more efficient by targeting the most valuable segments of the electorate, but many campaigns have been slow to use them.

Although it isn’t a new problem — television markets and political geographies are rarely perfectly aligned — as campaigns gain access to newer targeting techniques and more information about markets for advertising, the inefficiency is increasingly visible. The proliferation of set-top boxes to which specific commercials can be sent, and similar capabilities for satellite dish systems, make it possible for campaigns to spend smarter.

“Having alternatives to spending a lot of money is a relatively new concept,” said Elizabeth Wilner, senior vice president for political advertising at Kantar Media, which tracks broadcast ad spending. “But there are more options now than there were even a few years ago for more targeting of voters, like local cable and addressable satellite, or using analytics to at least identify cheaper programming through which you can reach the same audiences.”

There’s no single reason that campaigns are slow to use refined targeting techniques honed during the 2012 election cycle. Partly it is inertia, and partly it is that political ad buyers operate in an inefficient market: They lack easy access to information that would help accurately evaluate what they are buying. That’s changing, too, but slowly.

President Obama’s campaign used targeting alternatives during the 2012 presidential race. It bought segments of an audience rather than an entire audience, which might contain people a campaign doesn’t care about reaching. But in statewide races with large media budgets, segmentation might not be as much of a concern, as viewers in South Carolina living near Charlotte, N.C., could tell you after the barrage of ads in the North Carolina Senate contest.

For television viewers in areas with competitive House races, not much has changed: Most noncable broadcast advertisements in House races air on just 11 programs, among them the local news, “Wheel of Fortune,” “Jeopardy!” and “The Today Show,” according to an analysis of advertising data by the digital consulting firm Targeted Victory. What many campaigns do is make television buys on the basis of “Gross Ratings Points,” a marketing metric that measures the scope of the ad buy but not necessarily the size or composition of its audience.

In Georgia’s 12th District race, where $5 million — and counting — has been spent by the incumbent Democrat, John Barrow, and Rick Allen, the Republican challenger, there has been very little cable advertising. Cable doesn’t offer the range of spots that broadcast does. Local or regional cable companies might have only one or two spots an hour available because national advertisers take most of the inventory. Other Georgia campaigns, including those of contested Senate races and the governor’s election, compete for those spots. Cable buys are also concentrated: 90 percent of cable spending in House races goes to 20 networks that account for a little more than a quarter of overall cable viewership.

“A problem with buying local cable is that there’s a lot more inventory in local affiliate broadcast,” said Carol Davidsen, the director of integration and media targeting for the 2012 Obama campaign, where she led the effort to deliver ads to specific portions of the television audience. “For a local election, it’s too expensive for national cable. Where can you go? Back to broadcast.”

Cable and satellite also suffer from information inefficiencies. Broadcast stations are required to post online reports on political ad buys, but cable and satellite operators are not. In August, the Federal Communications Commission said it would seek public comment on expanding the requirement to them. Such disclosures would not only provide greater transparency to the public, but would also give ad buyers more accurate information about the markets they seek to enter. Knowing the actual market price for unconventional ad buys — those that aren’t just on the usual list of programs — should make campaigns more confident in their spending decisions.

For a campaign to take advantage of more targeting options would require either that campaigns maintain their own data and infrastructure, as the Obama campaign did, or use vendors that have already assembled voter data and television viewing habits.

“Without the political data tied to the buy, campaigns are not even sure they are actually targeting the right swing audience needed to actually win,” said Zac Moffatt, co-founder of Targeted Victory, which offers a targeting service geared toward Republican campaigns that don’t have a statewide budget.

What about online targeting? It’s also a work in progress, and no one is suggesting that it will replace television anytime soon. The potential for fraud, doubts about the effectiveness of targeting methods and online users’ aversions to clicking on ads have made it easier for campaigns to hold back on online spending, even though some of the same issues occur in television advertising.

That, too, is changing. In addition to Targeted Victory’s tool for Republicans, there is DemocraticAds.com, by online firm DSPolitical, aimed at local and state races. Such platforms, which allow campaigns to choose and assemble a target audience without having to gather the relevant data piecemeal, take some of the power away from traditional ad buyers, but also require a certain amount of knowledge of the data that underlies the systems.

The advent of stand-alone streaming services like the kind CBS announced last week raises the possibility of coordinating advertising buys across platforms. Viewers might see the same campaign ad whether they watched a show on television or on a tablet.

Companies like Verizon or Comcast, which provide both television and Internet, can identify which of their customers have both services and possess data on how they consume media. Making that information available to marketers to improve advertising targeting would be appealing to campaigns, but it could raise privacy concerns among subscribers.

Even without the online component, the level of detail that is available about television viewership, combined with detailed political data and some demographic context, is changing the market for political advertising.

The tradition of relying on Gross Ratings Points, while still a fixture in many campaigns, is no longer the force it was.

“A thousand G.R.P.s a week, and you’re O.K.,” said Ms. Wilner of Kantar Media. “That’s not really the case now. I see that rule of thumb going away.”