The Pattern Is Not the Reason SOC 3333 · Crime Lab · Midwestern State University
Part 3 is where papers go wrong, and they go wrong in one specific way. A student finds a real pattern, reports it accurately, and then stops. The pattern sits on the page looking like an explanation, and because nobody wrote a false sentence, nobody notices that no explaining happened. This page is about the step after the pattern. It is also about doing that step responsibly, because the patterns in this part are about people, and a pattern reported without a mechanism invites the reader to supply their own. Nothing here is recorded and nothing is graded.
A pattern tells you where to look. It does not tell you why.
Every statement below is true. Some of them describe a pattern and some of them propose an explanation for one. Sorting them apart is the whole skill, and it is harder than it sounds, because a well-worded pattern can feel like it already contains its own reason.
Two words that get used as if they were one
Part 3 asks who offends and who is victimized. That means you will be reporting group differences, and the vocabulary for doing it precisely exists for a reason.
Disparity
A difference between groups in an outcome. Group A is arrested at a higher rate than Group B. This is an observation. It says nothing about why, and by itself it accuses nobody of anything.
Disproportionality
A group's share of an outcome compared to its share of the population. A group that is 13 percent of the population and 30 percent of arrests is over-represented by that measure. Also an observation, and a different one.
Three sentences about the same finding. Only one of them is defensible. Pick it.
One gap, four explanations that all fit it
Here is a pattern of the kind you will report. Imagine that arrest rates for your offense are substantially higher in one neighborhood than in a demographically similar one nearby. That is real, and it is the beginning of the work rather than the end. At least four different mechanisms produce that same observation, and the arrest data cannot tell them apart.
Getting the patterns for your own offense
Start with the five sources you already have. Most research articles report demographic and situational breakdowns somewhere, and students routinely miss them because of where they sit in the paper.
Six kinds of pattern, and where each one usually lives.
How to write a group difference without writing something false
This is a short tab and it is the one worth rereading before you submit. Four habits, each of which takes one clause.
One. Attach the rate to the category, not the person. Write that arrest rates were higher among a group, not that a group is more criminal. The first is what the data says. The second is a claim about people that no arrest count can support, and the distance between them is the entire difference between criminology and prejudice with a citation attached.
Two. Name the instrument in the same sentence. Arrest rates, self-reported offending, and victimization rates give different pictures of the same phenomenon, and the gaps between them are themselves findings. A pattern in arrests is a pattern in arrests. Say so and the sentence becomes both more accurate and more interesting.
Three. Give the mechanism, even provisionally. A pattern without a proposed mechanism is an invitation for the reader to invent one. You do not have to be certain. You do have to say what you think is producing it and on what basis.
Four. Say what would change your mind. One sentence. If this evidence appeared, I would think the other explanation was right. It is the cheapest credibility available in academic writing and almost nobody in an undergraduate paper does it.
Before you submit Part 3
Work down this list with your draft open. Click a line when it is true. Nothing here is recorded.