advice you can use — short and to the point — every Tuesday, Wednesday & Thursday

Wednesday, April 5th, 2017 technology  research  practice

A Research Tip

  • Research & Writing

Rangefindr is a fabulous source for searching sentencing ranges in Canada.  If you are a criminal lawyer I highly recommend you speak to your firm librarian or whoever is in charge of licensing online resources.

I have had several people tell me that Rangefindr is difficult to use, and in particular that any search returns only a paucity of hits. So I came here to tell you that there is nothing wrong with Rangefindr – you’re just doing it wrong! There, don’t you feel better now? The problem isn’t with Rangefindr, it’s just your shoddy searching technique.

Basically, you’re checking too many boxes. No matter how sophisticated a search engine is, it is limited by the encoding of its data. Rangefindr offers you hundreds of search parameters, but these are only useful to the extent that the individual documents are coded for each parameter.

For example, say the accused in your case was amenable to counselling, and while looking through the various parameters under “Accused” in Rangefindr you see the check box “Amenable to counselling” – so naturally, you check it, right?

Wrong.

Ask yourself, what percentage of Canadian sentencing decisions where the accused was amenable to counselling actually say so in the decision? And further ask, given the granularity of Rangefindr’s data, how likely is it that the coder will have re-entered that particular piece of information into the record? In order for using that check box to return a positive result, both of the above contingencies must have been met.

Let’s further develop our example. Say the accused was male, 45, charged with driving impaired, had no previous charges, and was considered by many to have a “good character,” but was also known to be a heavy drinker.

Begin with the charge – click on “OFFENCES” in the left-hand pane. This part is easy – scroll through the offences, find “Drive impaired” and click on the heading (the words, not the box), and the little box will turn black. Now look at the square at the bottom right of the screen (inside the black rectangle) – it should say “PLEASE SELECT MORE TAGS TO NARROW YOUR SEARCH.” Keep an eye on this square.

Now, let’s click on the second item in the left-pane menu, ACCUSED. And scroll through the many parameters. For the purpose of this exercise, go ahead and click on all that apply, and WATCH what happens in the square in the bottom right corner. When I click Addiction/Alcohol, I see the number 64 – that means there are 64 relevant hits in the database. Now try “Amenable to counselling” – we are down to 3! What happened? Try de-selecting “Amenable” and click “Good character” – 1 hit. In fact, ANY combination of two or more of these criteria quickly brings the results to zero. Is it because there have been no reported Canadian decisions that resemble our case? I hardly think so.

With Rangefindr, the first question you need to ask is: what are the key aspects of this case – the parameters that you are nearly 100% sure would be (1) mentioned prominently in the decision, and (2) coded into the database. Secondly, ask yourself – Self, of the remaining criteria, those vulnerable to being left out of the data entry process, which are most critical to my argument?

Less is more.

After the offence itself, what is the MOST important fact of the case? And which factors are most likely to be included in the record? Let’s say the fact that the accused had no prior record is an important factor. Select “Record” in the left-hand menu, and then click on “No previous custodial sentence.” Our number has dropped to 6 – that’s with no other parameters but the charge. Is this a reliable result? Yes, in the positive sense that we have 6 cases where the accused had no prior record – but NOT in the negative sense. The search has probably screened out many relevant cases where the accused had no criminal record, but where that piece of information did not make it into the database.

For the present example, the correct strategy is to select the offence, then click on your other important criteria one at a time to get cases where the courts considered each of the relevant factors. If you select more than a couple of boxes, you are overlying on contingencies of the data entry process.

The same applies to criteria under COMPLAINANT, CONDITIONS, DETAILS OF CASE, DETAILS OF OFFENCE, JUDGEMENT, PRE-SENTENCE, and RECORD. Only select additional parameters carefully, while keeping an eye on that results box.

A more general strategy would be to first select the charge, then one further key parameter. If the number of results is still high, select the next most important factor, and work your way down – but always check what effect each parameter has on the results, watch for dramatic drops in the number of decisions, and reflect on whether each factor is likely to eliminate good results.

[This tip by Ken Fox originally appeared on the Law Society of Saskatchewan Library’s Legal Sourcery Blog]

Leave a Reply

Your email address will not be published. Required fields are marked *