Ciruli Associates recently polled Northern Coloradoans – people living in both Larimer and Weld Counties – and found strong support for Glade Reservoir and the Northern Irrigated Supply Project (NISP). The survey results seem unambiguous: 57% of Fort Collins respondents favored the reservoir; 63% of Larimer County respondents favored it; and 81% of Weld County respondents favored it (Kevin Duggan [13-Sep-08], Poll shows strong support for Glade, Coloradoan; and Northern Integrated Supply Project: Larimer and Weld Counties Polls: Survey Results [Aug-08], Ciruli Associates, Denver, Colorado [1490 Lafayette Street, Suite 208, 80218, www.ciruli.com], 24 pages).

The Ciruli report says the survey was sponsored by "Fifteen water providers proposing the Northern Integrated Supply Project," which conspicuously fails to name what organization footed the $35,000 bill ($35,000 for calling up 500 people and generating a 24-page report – What a sweet deal!). The report itself is posted at the NISP website (www.gladereservoir.org), where the Project Manager and Public Information Officer are shown to be affiliated with the Northern Colorado Water Conservancy District. Notwithstanding these ambiguities in who commissioned and paid for the survey, the results suggest that the further away from the reservoir – and from the resultingly dried up Poudre River – the survey respondents lived, the more they favored the project.

The anti-Glade group Save the Poudre Coalition criticized the survey for including biased questions – which, the group said, "pushed" respondents to express support for the reservoir (Kevin Duggan [15-Sep-08], Glade opponents blast NISP survey, Coloradoan; and Kevin Duggan [16-Sep-08], Glade foes rip survey, Coloradoan).

How should we take the survey, then?

Craig Trumbo helps us understand what the survey means. Trumbo's letter to the editor of the Coloradoan (referenced above and reproduced below, with highlighting mine) points out that most respondents to the survey were unaware of the issues surrounding Glade Reservoir and formed their opinion on-the-fly.

So, what do Northern Coloradoans think of Glade Reservoir? The Fort Collins City Council voted unanimously against it. The Larimer County Board of Commissioners decided to remain neutral. Personally, I'm opposed to drying up the Poudre and forcing Fort Collins to bear the astronomically expensive brunt of NISP's environmental and economic repercussions. But that's not the view of my friend the real estate agent who couldn't be more in favor of the development along I-25, which the project would allow. Another friend who's a soil scientist, who works for local environmental consultancies as a private contractor, says he doesn't know what to think. And I suspect that's the view shared by most Northern Coloradoans.

Glade Poll Reporting Was Uncritical
The Coloradoan's Sept. 13 story on the Glade Reservoir poll was not especially insightful journalism. While the survey was conducted scientifically, the interpretation of the results should have been more thoroughly examined.

Only 52 percent (260 responses) of the 500 respondents in Larimer County were aware of the issue and only 28 percent (84 responses) of the 300 respondents in Weld County were aware of the issue. Yet most respondents still went on to offer an opinion – only about 10 percent said "don't know" on the question concerning support. Many of the opinions reported in this survey were quickly formed at the time of the interview (a phenomenon known as "pseudo opinions"). Responsible survey reporting would have constrained the results to the respondents who reported having been previously aware of the issue.

The headline for the story might have been "Poll shows mixed awareness of Glade." The lead might have been "Nearly half of Larimer County residents and fully two-thirds of Weld County residents report knowing very little or nothing about Glade Reservoir..." These results can hardly be interpreted as showing "overwhelming support" for Glade, as reported.

Two technical issues are worth noting. For the smaller sample sizes of respondents who were aware of the issue, the margins of error are about +/- 6 percent for Larimer and +/- 11 percent for Weld County. Finally, no overall participation rate was reported, so we do not know how many refused to participate. This is important for assessing potential bias in the sample.

Craig Trumbo, Ph.D.
Fort Collins