The focus on subject matter breadth over depth seemed appropriate for an introductory class. The most important parts of learning something far from one's comfort zone are:
(1) learning a few basic tools that can be used to approach a problem, and
(2) learning where to find more information and better tools.
I'm certainly not an expert in Python, but if I'm given a problem and told to code a solution I can at least get started on it with the tools we learned in class, and I at least know where to look for more information and better tools. If I run across some mention of p-values or "95% accuracy" this summer, I'll know to not just take that at face value, I'll know how to peek under the hood a bit, and I'll know where to get help if I need to dig a little deeper.
Getting a handle on the basics of a lot of different areas seems pretty valuable, and it seems considerably more valuable than going in-depth in one or two specific areas. Say we instead did fourteen weeks exclusively on hypothesis testing. I might not run into a case that turns on that until, oh, 2023, by which point I'd need to re-learn all but the basics. And if I'm only likely to retain the basics through a few years of disuse, doesn't it make sense to only cover the basics? Doesn't it make sense to cover the basics of a handful of topics in 2019 and remember the basics of a handful of topics in 2023, instead of going in-depth on one topic now and winding up with only one set of basics down the line? It seems as if the goal of most first-year classes is to give us at least a passing familiarity with a wide variety of legal issues, so that we're not completely in the dark if one of those issues pops up out of the blue one day. That's a good approach for this material, too.
One narrow suggestion is to make a number of ungraded coding problems (and their solutions) available early in the semester. As they'd be ungraded, students could work together on them without academic integrity issues. There'd also be reduced pressure to spend a ton of time on them (maybe include a "if you've spent two hours on this and it's not clicking, bring it to class or office hours" notation). And hopefully, getting in some low-pressure at bats would make the graded problem sets easier.
The focus on subject matter breadth over depth seemed appropriate for an introductory class. The most important parts of learning something far from one's comfort zone are:
(1) learning a few basic tools that can be used to approach a problem, and (2) learning where to find more information and better tools.
I'm certainly not an expert in Python, but if I'm given a problem and told to code a solution I can at least get started on it with the tools we learned in class, and I at least know where to look for more information and better tools. If I run across some mention of p-values or "95% accuracy" this summer, I'll know to not just take that at face value, I'll know how to peek under the hood a bit, and I'll know where to get help if I need to dig a little deeper.
Getting a handle on the basics of a lot of different areas seems pretty valuable, and it seems considerably more valuable than going in-depth in one or two specific areas. Say we instead did fourteen weeks exclusively on hypothesis testing. I might not run into a case that turns on that until, oh, 2023, by which point I'd need to re-learn all but the basics. And if I'm only likely to retain the basics through a few years of disuse, doesn't it make sense to only cover the basics? Doesn't it make sense to cover the basics of a handful of topics in 2019 and remember the basics of a handful of topics in 2023, instead of going in-depth on one topic now and winding up with only one set of basics down the line? It seems as if the goal of most first-year classes is to give us at least a passing familiarity with a wide variety of legal issues, so that we're not completely in the dark if one of those issues pops up out of the blue one day. That's a good approach for this material, too.
One narrow suggestion is to make a number of ungraded coding problems (and their solutions) available early in the semester. As they'd be ungraded, students could work together on them without academic integrity issues. There'd also be reduced pressure to spend a ton of time on them (maybe include a "if you've spent two hours on this and it's not clicking, bring it to class or office hours" notation). And hopefully, getting in some low-pressure at bats would make the graded problem sets easier.