First, a "survey" of high school and junior high kids is not a good indicator of actual numbers. The survey results could be roughly accurate, or they could mean that the kids answering this survey simply lied more than the ones answering the previous surveys, or any number of other possible explanations. I'll bet if they asked how many cheated on tests or had premarital sex, the numbers would have been low as well. Regarding premarital sex, STD rates and pregnancy rates by the mothers age are a better indicator of premarital sex trends than any survey. For example, if a survey of 15-18 year old girls "showed" that they had premarital sex at a rate half of what they did five years ago, but a study showed that the STD and pregnancy rate had actually gone up by 30% in that same age group over that same time frame, which would you believe was the better indicator of the premarital sex rate?
Second, this survey only addressed 8th - 12th graders. Sadly, lots of people in their 20's 30's, 40's (you get the idea) abuse drugs as well.
Third, I noticed meth and heroin were missing from the article, so I went to the web site at the Utah County Health Dept. Here is the 2007 annual report.
Utah County 2007 ReportOn page 10 is a graph showing admission to the counties substance abuse treatment facilities by "drug of choice".
Alcohol has indeed gone down, from almost 40% in 1999 to about 24% now. Heroin, on the other hand, has gone up from about 6% of admissions in 1999 to over 20% in 2006 and 2007. Meth has also gone up from about 20% in 1999 to about 30% in 2007.
Usually, when residents say a community has a drug problem on a survey, it is because they know of someone (or more than one) who has a drug problem. Especially in regards to heroin and meth, the chart above backs that up.