<p>Starch:</p>
<p>Federal law requires annual reporting by all colleges of reported crimes on campus, broken down by location (including residence halls). This allows a security concious parent, such as yourself, to compare colleges and see if their security measures produce results.</p>
<p>Let's look at the reported stats for dormitory crimes for the more recent three year period at two similar LACs (one with keys, one with cards) to see if there is any obvious difference:</p>
<p>Crimes, by category, reported in residence halls 2004-06)</p>
<p>Murder
Swarthmore: 0
Williams: 0</p>
<p>Manslauter
Swarthmore: 0
Williams: 0</p>
<p>Forcible sex offensives (rape, sodomy, assault with object, or fondling)
Swarthmore: 3
Williams: 6</p>
<p>note: These are almost certainly "date rape" type incidents between students acquainted with each other. There are been no reported incidents of "outsider" rape at either school in years, if ever.</p>
<p>Aggravated assault
Swarthmore: 0
Williams: 4</p>
<p>note: These incidents are almost certainly student on student fights or brawls resulting in arrest. I believe the incident at Williams was two football players beating each other to the point of hospitalization over a girl. There were assualt arrests at Swarthmore a couple of years ago in conjunction with a frat fight. It's not listed here because it didn't take place in a dorm.</p>
<p>Burglary
Swarthmore: 36
Williams: 144</p>
<p>note: This category includes all unauthorized entry into a student's room for the purpose of stealing, vandalism, or any other crime, whether theft actually occured or not.</p>
<p>This is only the most recent three year period. Also, you need to adjust these figures to reflect different enrollments (multiple Williams numbers by .75). But, with these caveats, nothing is jumping out at me to suggest that a magnetic card entry system offers any measureable benefit in reported crime statistics.</p>
<p>One of the things that a solid liberal arts education teaches is to use statistical-based evaluation in a cost/benefit analysis.</p>