Many states have made efforts to create a more efficient, effective juvenile justice system. But according to a recent report by the Council of State Governments Justice Center and the Public Policy Research Institute, Texas took a somewhat different tack than many others.

Unlike most states, Texas has a long history of putting resources into the development of technology to track children in the juvenile justice system, including both those who are in state-run correctional facilities and those in local juvenile probation departments. This enables the state to ensure that the young people who have run afoul of the law are treated in the most appropriate and cost-efficient way available.

MORE: This appears in the free Management e-newsletter. Click to subscribe.

In an unusual move, Texas went further and measured the success of this information system for managing its juvenile justice population. The report found that this use of data was “at least a key factor and likely the principal driver” of a 66 percent reduction in youth incarceration in state-run secure correctional facilities between 2007 and 2012.

Here's a chilling little tale from a Dec. 2014 audit report in Oregon. In July 2013, a department of human services worker was inputting a simple payment transaction into the OR-Kids computer system. The provider's payment was supposed to be $323.88, but the worker mistakenly input the program's federal ID number into the field instead, according to Oregon auditor Gary Blackmer. As a result, the department sent out a payment for $1,748,304, even though a supervisor had reviewed the transaction, according to Oregon internal control procedures.

This isn’t the only instance of this kind of error. The department also sent out another $1.7 million check back in April for a $400 liability. In this case too, a supervisor had approved the payment. It’s hard to see how supervisors could have missed either of these payments since 99.8 percent of all other payments processed through OR-Kids were less than $3,500, according to the audit. As the audit says: "We concluded that the supervisor’s review and approval in these instances was perfunctory.”

In both cases, the state got its money back, but that doesn’t mitigate the oversight problems that permitted the checks to go out in the first place. The department has since put in place a relatively straightforward safeguard. When someone tries to enter a payment for more than $4,000, it’s automatically rejected and requires further review and approval. The audit reports that while this automated threshold prevents huge errors, a "thorough and attentive review" is still necessary to guard against smaller ones.

Here’s a major takeaway from a recent National Association of Counties’ forum on optimizing health, justice and public safety. It should apply to a host of government efforts: “Programs should ask the families they serve ‘What do you need?’ and not ‘What is wrong with you?’”

Obviously clients of government programs will need to explain the exact nature of their problems. But starting out focusing on the problem can easily leave the program and its clients at odds.

People who use government programs generally have some idea of what they need. Looking at people as a problem to be solved is dehumanizing and can take the process in the wrong direction quickly. It may be, for example, that someone’s problem would be described as “I can’t deal with my young children,” which could lead a government employee down a variety of routes including psychological intervention. On the other hand, asking “What do you need,” might lead to a comment like, “I need someone to watch the kids for an hour a day while I shop.”

“Data is absolutely useless. It’s what you do with it that matters.” -- Justin Minkel, elementary school teacher in Arkansas, quoted in Education Week

The phrase “go big or go home” seems to be increasingly common. We can’t turn on the television without those words coming out of someone’s mouth. But when it comes to government, we know there are a number of cases in which this mantra can lead managers in the wrong direction.

We were recently talking to Max Arinder, the long-time director of the Performance Evaluation and Expenditure Review Committee in Mississippi, about that state’s efforts to develop a performance management and performance measurement system. It’s backed by the legislature with an awareness that it’s going to take time and require several revisions.

This was a lesson hard learned. Back in 1994, the state tried something similar, and it tried to do it in a very short time span. “It was a big splash on the heels of reinventing government,” Arinder told us. “We were going to pass the law and snap our fingers and it was going to work, but [the state] didn't get the little stuff right so the big stuff wasn’t right.” For example, back in the 1990s, Arinder said, the state tended to measure things that could be easily counted -- like the number of hours students spent in reading classes as opposed to their impact.

Now the state’s leadership is focused on developing measures that are meaningful and sees the wisdom of building the foundational elements and getting people talking about them and posing questions of how they're using them. “Let’s be patient enough to put it in place and see if it can work,” said Arinder.

As water shortages loom ever larger in the list of problems that assail the nation’s localities, some are turning to technology to help with conservation. Here’s one excellent example highlighted by EfficientGov. It’s from Cooper City, Fla., a community of 30,000 in Broward County. The city established a wireless network, which takes the data from building meters and sends it directly to utility work stations. The city’s public works department monitors the data in search of anomalies, including larger than usual water flows, irregular patterns in the use of water and unusual usage times. If any of those occur, the software automatically generates a letter to customers, letting them know that they may have a leak and the amount it may be costing each month.

Another instance in which a city’s close analysis of data can make life better: Some 200,000 New York City households get eviction notices each year, according to a recent article on Next City. But all evictees are not equal. A college student who gets thrown out of an apartment often has the option of moving back home with his or her parents. On the other hand, a family in which no one has a job, and there’s no support system, may wind up homeless. That’s a painful proposition for the people involved and expensive for the city, which provides homeless shelters and health care for people who don’t have a roof to sleep under at night.

But a company called SumAll has created a tool that looks at court records, shelter history and demographic information to find the people at greatest risk of becoming homeless following an eviction. The city can reach out to those families in a variety of ways, including helping them figure out how to prevent the eviction from taking place at all. When the city piloted the system, the results were encouraging. "[It] was able to connect 50 percent more families in the test neighborhood with eviction prevention services compared to demographically similar neighborhoods nearby over the same period of time,” wrote Next City.

There is a widespread belief that sexually abused children are far more likely than other people to become sex offenders themselves. It feels to us like this idea is pretty much common knowledge and can help drive public policy. But we came across an article in the Journal of the American Medical Association that looks more closely at the proposition and concludes that “the widespread belief that sexually abused children are uniquely at risk to become sex offenders was not supported by prospective empirical evidence.” We don’t have the backgrounds to assess the significance of this study, but it certainly does reinforce our belief that policymakers can be at their best when guided by real-world evidence and not just something they’ve read so many times that it seems undeniable.

(By the way, in researching this item, we came across a hard-copy clip in our files from 1987 that points to a study by Yale researchers that made the same point about child abuse and seemed to have minimal impact on the way the public understands this topic.)

The National Association of State Personnel Executives had its mid-year meeting at the end of January. The organization selected five major issues for focus in the year to come. They were:

1. What are the future workforce trends and how will jobs change?

2. How do you go about implementing non-traditional human resources programs like pay for performance?

3. How can a public employer put more flexibility into compensation offerings?

4. How do you properly roll out a performance management program to get buy-in from employees and supervisors?

5. How do you define strategic HR and move professionals from operational to strategic roles?

These issues determine the association’s programming for the coming year, and for HR personnel at all levels of government, they present a neat list of the issues that are likely to cross their desks over the coming months -- and for which they should be prepared.

For some time now, we’ve been writing cautionary pieces about states, counties and cities that have tried to avoid tax increases (or tuition increases in the case of higher education) by putting all sorts of fees into place. It was just a matter of time before citizens would begin to see that often the fees were just taxes by another name.

A clear example of this took place at the end of January when the trustees of the California State University system voted unanimously to “place tighter controls on controversial student fees that have drawn criticism,” according to the Los Angeles Times.  The most powerful idea under consideration is that, in the future, students would have to vote on whether or not a new fee should be approved. The vote would be binding and the university would have every opportunity to explain how the fees would be used. That’s critical. All fees aren’t created equally; many students might vote in favor of a new fee if it made an addition to the curriculum possible. Or not. But at least they’d have a choice.