I suspected that the statistical data was tampered with or flawed since it goes against every instance of common sense and logic. Not to mention record gun sales in "ban panic".
“Share of homes with guns shows four-decade decline” was the headline in the New York Times a couple of days ago. Similar reports in various news outlets picked up on the apparent slide in US gun ownership, on the eve of the National Rifle Association’s annual convention. “Even as the conversation on guns remains contentious,” the Times wrote, “a broad shift away from gun ownership is under way in a growing number of American homes.”
But those statements, supposedly based on the latest data from the (mostly biannual) General Social Survey from the National Opinion Research Center, are misleading. The above chart shows what percentage of a sample of Americans, asked if there was a gun in their household, said “yes.” The proportion of households with guns has fallen from a high of over 50% in the 1970s, but since 1998, it has held steady, fluctuating between 37% and 32%. The figure for 2012 was 34%.
The number of respondents to the question ranges from less than 1,000 to around 2,000, meaning that the margin of error is roughly 2% to 3% in either direction. In other words, the data for the past 15 years show no statistically significant variation. “It’s been essentially flat” over that period, Tom Smith, director of the GSS, told Quartz. A close statistical analysis of the data, he says, might not show a perfectly constant level of ownership, “but substantively you can conclude there’s been minimal further decline.”
Whether this means US gun ownership has reached a natural lower limit or is just temporarily stable is beyond the ken of statisticians. But it does suggest that it’s premature to conclude that American gun owners are a shrinking breed.
edit on 12-3-2013 by tadaman because: (no reason given)