I'll start by echoing the general sentiments that, at the end of the day, a career in social sciences demands good stats knowledge. At my school, the only program with more stats requirements than psychology is statistics. That said – as other posters have notes – many of the students don't like it, and *many* of them have long since convinced themselves that they are incapable of learning it.
Here's a bit of what you can expect: Your distribution will be skewed to floor, skewed to ceiling, or bimodal. The level you gear your teaching to should be thought of as a decision on the relative numbers of students you wish to be so bored they stop attending, versus the number you wish to see crying in your office hours.
I haven't taught stats in this context, but have faced a similar course situation teaching physiological psychology (aka neuroscience) as a core psych credit for a primarily arts-based psych program. Interestingly, psych students comprised the majority of the top-performers, and biology students the majority of the bottom performers. In the environs of the overall mean, however, BSc students tended to do marginally better than BA students. In this context, the struggle was always between accuracy and transparency. Students in an arts degree don't take kindly to learning about cable properties and voltage-gated ion fluxes – but they're perfectly capable of learning it and understanding it if you avoid overly technical language.
You're likely to face similar resistance in terms of math equations. Unfortunately you can't easily analogize your way through statistics – so my best advice would be to head this off at the start with a little topical prologue: math anxiety, and stereotype threat. If these are psych students, then you might be able to engage them early on by examining why they might believe themselves to be incapable of the math, and similarly why they're wrong. It's certainly outside your curriculum, and might be outside your comfort zone – but that could be a good thing. Show them a little vulnerability by delving into the psych side of math anxiety as a means of encouraging them to confront their own vulnerability with respect to math. A long shot, but might be worthwhile...
I'm utterly amazed at the number of statements on here, from a supposedly technologically-savvy population, to the effect that: "A computer's responses are entirely determined by the programmer" – ignoring utterly that the vast majority of sophisticated code these days collects data and that data is churned through learning algorithms that modify how the program will respond to future inputs. Indeed, once a program like this starts running, identical inputs at different times can and often well produce different outputs without any intervention from the programmer. These arguments are analogous to the nonsensical idea that your decisions are determined entirely by your genome. Rubbish.
Academic texts are already so painfully inbred that you'll encounter minor tweaks of the same examples and diagrams in just about every publisher's version. But of course, in the drive towards profits, they iteratively condense and paraphrase each other to make new editions each year, and the end result is material that's either grotesquely confused by efforts to avoid direct plagiarism, or else made easy to understand by sacrificing any semblance of accuracy.
These books are meant to help people learn. When, as a lecturer, you frequently have to tell your students to ignore certain parts of the book because they're inaccurate, or you have to cross-check all the in-text figures against the original source figures to make sure they haven't been 'simplified' into falsehood, you know there's a problem.
As always, getting rid of the middle-man would be a tremendous help to everyone.
Slashdot Mirror
I'll start by echoing the general sentiments that, at the end of the day, a career in social sciences demands good stats knowledge. At my school, the only program with more stats requirements than psychology is statistics. That said – as other posters have notes – many of the students don't like it, and *many* of them have long since convinced themselves that they are incapable of learning it.
Here's a bit of what you can expect: Your distribution will be skewed to floor, skewed to ceiling, or bimodal. The level you gear your teaching to should be thought of as a decision on the relative numbers of students you wish to be so bored they stop attending, versus the number you wish to see crying in your office hours.
I haven't taught stats in this context, but have faced a similar course situation teaching physiological psychology (aka neuroscience) as a core psych credit for a primarily arts-based psych program. Interestingly, psych students comprised the majority of the top-performers, and biology students the majority of the bottom performers. In the environs of the overall mean, however, BSc students tended to do marginally better than BA students. In this context, the struggle was always between accuracy and transparency. Students in an arts degree don't take kindly to learning about cable properties and voltage-gated ion fluxes – but they're perfectly capable of learning it and understanding it if you avoid overly technical language.
You're likely to face similar resistance in terms of math equations. Unfortunately you can't easily analogize your way through statistics – so my best advice would be to head this off at the start with a little topical prologue: math anxiety, and stereotype threat. If these are psych students, then you might be able to engage them early on by examining why they might believe themselves to be incapable of the math, and similarly why they're wrong. It's certainly outside your curriculum, and might be outside your comfort zone – but that could be a good thing. Show them a little vulnerability by delving into the psych side of math anxiety as a means of encouraging them to confront their own vulnerability with respect to math. A long shot, but might be worthwhile...
I'm utterly amazed at the number of statements on here, from a supposedly technologically-savvy population, to the effect that: "A computer's responses are entirely determined by the programmer" – ignoring utterly that the vast majority of sophisticated code these days collects data and that data is churned through learning algorithms that modify how the program will respond to future inputs. Indeed, once a program like this starts running, identical inputs at different times can and often well produce different outputs without any intervention from the programmer. These arguments are analogous to the nonsensical idea that your decisions are determined entirely by your genome. Rubbish.
Academic texts are already so painfully inbred that you'll encounter minor tweaks of the same examples and diagrams in just about every publisher's version. But of course, in the drive towards profits, they iteratively condense and paraphrase each other to make new editions each year, and the end result is material that's either grotesquely confused by efforts to avoid direct plagiarism, or else made easy to understand by sacrificing any semblance of accuracy.
These books are meant to help people learn. When, as a lecturer, you frequently have to tell your students to ignore certain parts of the book because they're inaccurate, or you have to cross-check all the in-text figures against the original source figures to make sure they haven't been 'simplified' into falsehood, you know there's a problem.
As always, getting rid of the middle-man would be a tremendous help to everyone.