Between March and August of this year, satisfaction with the New Orleans Police Department held steady at 58 percent, according to the results of the most recent independent survey of 600 city residents.
During that same six-month stretch, satisfaction of residents in the Uptown-based 2nd District appears to have plunged by 10 percent, yett soared by 12 percent in the 6th District, right across Napoleon Avenue, according to the survey results.
How much those widely divergent results actually reflect the attitudes of Uptown residents, however, is hard to ascertain because of the relatively small sample sizes in each individual district.
The citywide survey of 600 people (commissioned by the New Orleans Crime Coalition and available online at www.crimecoalitionnola.com) has a 4-percent "margin of error," meaning that the results are likely a fairly accurate reflection of the city as a whole within a few percentage points. This represents a major increase from the 33 percent on the first survey in 2009, said Ryan Steusloff, vice president of WPA Opinion Research — and a respectable number overall, given that the highest results he's ever seen for any police department top out in the low 80s.
But when the responses are broken down for the eight individual police districts, only 75 people within each district were polled. Such a small number leads to a much larger margin of error — specifically, 11.3 percent — and the "changes" in almost every district's results from March to August (including the 2nd and the 6th) essentially fall within it.
For example, in March, the 61 percent of the 75 people polled in the 2nd District represents 46 actual people who were satisfied. In August, the 51-percent result represents 38 people. So, the "10 percent drop" shown on the charts included in the survey results actually reflects a difference in eight responses from March to August.
Still, that 75-person sample was intentionally built to give the department some information about each district, Steusloff said.
"Seventy-five people per police district is enough to say something meaningful about each police district," Steusloff said.
Given the high margin of error for the district-level results, the overall trendlines are probably more reliable than the individual data points, Steusloff said. In an interview this week, NOPD Superintendent Ronal Serpas agreed: The high margin of error for individual district's results does not mean that they don't have a use, he said.
"We wouldn't bet the farm on it, but we would see trends in behavior over time," Serpas said. "It's still important information. If we're making movement in either direction, we want to reassess."
In the 2nd District, for example, the line on the question about satisfaction has made several 10-point swings up and down over the course of the past few years, from a low of 42 percent to a high of 68 percent.
"We're pretty confident that the trends are real," Steusloff said. "It's a very up-and-down trendline, where opinions appear to be changing very rapidly from survey to survey. Looking at it holistically, it's kind of holding right at the mid-50s. This is not a 'the world is burning down' kind of number. I would put it in terms of holding steady or a slight downward trend."
Police leaders are aware that, within the districts, each single person interviewed can have a major effect on the results, Serpas said.
"Maybe they had to wait a long time for an accident report, and that can kind of blow up pretty quick," he said.
Finally, it is noteworthy that Uptown residents appear to be more satisfied with their own districts than with the city as a whole — by figures well outside the margin of error. In the 2nd District, where only half the respondents were satisfied with the NOPD overall, 76 percent said they were satisfied with police performance in their neighborhoods. In the 6th District, the NOPD got a 61 percent approval rating, but 82 percent were satisfied with police work in their own neighborhood.
While 75 people may not be very strong as a polling sample, it does make a substantial focus group. One question on the survey asked residents what concerned them most, and those results may be the most helpful part of the poll for individual districts, Steusloff said.
For example, residents in the 2nd District showed a strong correlation between their satisfaction on violent crime and the importance of it, while they were less satisfied with property crime. The Second District has one of the lowest officer counts in the city, and Cmdr. Paul Noel has emphasized focusing his resources on reducing armed robberies and violent crime — resulting in a major drop in Uptown armed robberies this year, possibly at the expense of property-crime prevention.
The district-level surveys, especially when viewed holistically, also provide an important reality check, Serpas said, and supporters of the police who attend community meetings tend to be more positive about the department, which can create a positive feedback loop that could be deceptive.
"Police leaders can be drawn into a false sense of security with their existing networks of neighborhood leaders," Serpas said. "We're a little bit like a church — we don't want to always be singing to the choir."
Finally, Serpas said criminal-justice experts across the country are coming to believe that a scientific survey of the entire community provides an important second way of measuring a department's performance, better than crime statistics alone. Criminologists often decry the use of per capita FBI crime stats to measure individual departments, saying they are subject to major variations in reporting practices and local attitudes. The Crime Coalition's survey provides an important counterpoint to those figures.
"There is value here and, practically speaking, it's better than just guessing," Steusloff said. "I always think it's better to make decisions with data."