Question about stats in intro texts
Colleagues: Some introductory psychology textbooks include a pretty detailed explanation of descriptive and inferential statistics in the chapter devoted to research methods (others may include it but place it in the appendix). I never teach this material, as I am of the opinion that usually there is too much information, we are about to start learning about biopsych (so I am concerned about overwhelming them and it's a matter of priority) and frankly I am not sure that teaching mean, median and mode to students is really necessary for their appreciation of the research studies covered in the course. I would like to know if anyone does anything to actually teach this material in introductory psychology (as opposed to telling the students to read the material and know it, without actually covering it in class.) I think it's usually excessive to include it as anything but optional appendix material but I want to compare my views with those of others. Nancy Melucci Long Beach City College Long Beach CA --- You are currently subscribed to tips as: [EMAIL PROTECTED] To unsubscribe send a blank email to [EMAIL PROTECTED]
Re: Question about stats in intro texts
In my experience, college freshmen are already familiar with most of descriptive statistics (central tendency, frequency distributions, scatterplots) by the time they reach college, so I only mention these in brief. I can quickly explain standard deviation as something like an "average" distance from the mean, and the correlation correficient is understandable using corresponding plots. As for inferential statistics, I think this is too important a topic to completely skip. The basic idea isn't so tough and I present it in about 5 minutes. I start with class data on frequency of crying (large male-female difference here). I describe the two interpretations for our gender difference (male and female college students differ in crying frequency vs. they don't and our class data represent chance sampling error). Then I point out that we scientists are conservative and unwilling to say our sample data represent a "real" difference unless we can rule out the chance explanation. I tell them that in intro statistics, they will learn ways to calculate the likelihood that the chance explanation is correct (and if it turns out to be quite unlikely, say Five minutes is enough. They don't need to know about calculations, but a general understanding of the logic in inferential statistics should help them later. It also coincides nicely with our emphasis on critical thinking in dealing with research findings. --Dave [EMAIL PROTECTED] wrote: Colleagues: Some introductory psychology textbooks include a pretty detailed explanation of descriptive and inferential statistics in the chapter devoted to research methods (others may include it but place it in the appendix). I never teach this material, as I am of the opinion that usually there is too much information, we are about to start learning about biopsych (so I am concerned about overwhelming them and it's a matter of priority) and frankly I am not sure that teaching mean, median and mode to students is really necessary for their appreciation of the research studies covered in the course. -- -- ___ David E. Campbell, Ph.D. [EMAIL PROTECTED] Department of Psychology Phone: 707-826-3721 Humboldt State University FAX: 707-826-4993 Arcata, CA 95521-8299 www.humboldt.edu/~campbell/psyc.htm --- You are currently subscribed to tips as: [EMAIL PROTECTED] To unsubscribe send a blank email to [EMAIL PROTECTED]
Re: Question about stats in intro texts
I tell them that in intro statistics, they will learn ways to calculate the likelihood that the chance explanation is correct (and if it turns out to be quite unlikely, say Really, at Humboldt the Psyc students learn Bayesian inference? Might you really mean the likelihood of getting data as unsual as these (due to sampling error) were the null true? Most folks stick to the traditional p that is used by almost all psych researchers, that is, P(data | H0), the probability of getting data as unusual as those obtained assuming the null is true, not P(H0 | data), the probability that the null is true given the data. - Original Message - From: David Campbell To: Teaching in the Psychological Sciences Sent: Saturday, June 26, 2004 4:23 PM Subject: Re: Question about stats in intro texts --- You are currently subscribed to tips as: [EMAIL PROTECTED] To unsubscribe send a blank email to [EMAIL PROTECTED]
Re: Question about stats in intro texts
In fact, I usually use words similar to what you provided below, Karl. But this level of precision is lost on the students. I can only hope that they retain a general understanding that there are two ways to explain the findings and we need to rule out the chance explanation (by finding it an unlikely fit for the data). Too much reference to concepts like sampling distribution, standard error, and null hypothesis--and you can be assured that the first-year (just out of high school) undergraduates would be totally confused. --Dave Karl L. Wuensch wrote: I tell them that in intro statistics, they will learn ways to calculate the likelihood that the chance explanation is correct (and if it turns out to be quite unlikely, say Really, at Humboldt the Psyc students learn Bayesian inference? Might you really mean the likelihood of getting data as unsual as these (due to sampling error) were the null true? Most folks stick to the traditional p that is used by almost all psych researchers, that is, P(data | H0), the probability of getting data as unusual as those obtained assuming the null is true, not P(H0 | data), the probability that the null is true given the data. - Original Message - From: David Campbell To: Teaching in the Psychological Sciences Sent: Saturday, June 26, 2004 4:23 PM Subject: Re: Question about stats in intro texts --- You are currently subscribed to tips as: [EMAIL PROTECTED] To unsubscribe send a blank email to [EMAIL PROTECTED] -- -- -- ___ David E. Campbell, Ph.D. [EMAIL PROTECTED] Department of Psychology Phone: 707-826-3721 Humboldt State University FAX: 707-826-4993 Arcata, CA 95521-8299 www.humboldt.edu/~campbell/psyc.htm --- You are currently subscribed to tips as: [EMAIL PROTECTED] To unsubscribe send a blank email to [EMAIL PROTECTED]