Letter: Beware statistics bent to create 'fact'

Status
Not open for further replies.

WAGCEVP

Member
Joined
May 26, 2003
Messages
864
Letter: Beware statistics bent to create 'fact'


PUBLICATION: The StarPhoenix (Saskatoon)
DATE: 2004.01.10
EDITION: Final
SECTION: Forum
PAGE: A12
BYLINE: Joe Johnson
SOURCE: The StarPhoenix

--------------------------------------------------------------------------------

Beware statistics bent to create 'fact'

--------------------------------------------------------------------------------

In Facts show new firearms program worthwhile (SP Jan. 8), Tim Quigley suggests that citizens need to base their opinion on the gun control debate on facts rather than priorities.

Facts are usually conveyed in statistical form, i.e. numbers, percentages, fractions.

While facts are very beneficial for basing decisions and formulating arguments, statistics can be dangerous. It is simple to state a statistic and make your argument look sound. But it is difficult to account for all the variables which may affect this statistic.

Quigley's comparison of Canadian and American homicide rates and his "strongest validation of the new law" are weak. If Canada had poverty, homelessness, the cities and the ghettos resembling those of the U.S., our murder rate would certainly be much higher. Maybe our health-care and social system help keep the murder rate down. Who knows exactly?

What I do know is that there are a lot more variables to consider than the presence, or lack thereof, of a gun registry system.

I am also curious as to why Quigley decided to compare the number of murders with rifles and shotguns in 1989 to those in 2002. Why not compare 2002 with the numbers for the year before Bill C-68 was introduced? It may be because the murder rate wasn't much different and, therefore, the argument wasn't as convincing.

I, too, urge citizens to base their opinions on facts -- but to be aware that statistics can be misleading.

Joe Johnson

North Battleford
 
Statistics are not misleading. Using small portions of the statistics, numbers out of context and searching for data that meets the numbers you desire (ie using 1989 murders compared to 2002 murders) is what is misleading.
 
Badly employed statistics are comparable to ''out of context'' quotes ... they are used for convenience, and deception too.

If statistics are used where samples are adequate in size .. and a whole sequence is quoted, then they can be valuable tools to demonstrate trends.

Stat's themselves need not lie .. but their improper presentation sure as hell can.
 
IIRC, years ago, the city of New York wanted to ban ice cream sales in and around Central Park. It seems someone noticed that as ice cream sales went up, the number of rapes in the park went up. Thankfully, someone realized the reason for the correlation was the weather. More people used the park during warm weather so there was more opportunity to commit violent crimes. Also, people eat more ice cream during warm weather. Statistics can be used to "prove" anything. :scrutiny:
 
Status
Not open for further replies.
Back
Top