Re: Statistics Teaching Questions [SEC: UNCLASSIFIED]
Locked
Re: Statistics Teaching Questions [SEC: UNCLASSIFIED]
 
|
I agree with what Art has to say below. The worst thing you could do is have them finish the course and think they are statisticians. An awful lot of my work through various jobs has been data cleaning and manipulating, to get data into the correct form in which to analyse. If you could also pass onto them that statistics isn’t just sitting in front of a computer and having the program spit out the correct results, but that – in many cases – most of the time will be in cleaning data and a minor portion will be analysing/reporting, that would be helpful as well. Do any of them need to be able to interpret published statistics, e.g. in journal articles? If so, this book might be helpful in teaching them what detail should be available in the publication: I also liked books along this line for showing how important good statistics are, how easily people can be fooled by “common sense” and/or bad statistical practice, and are easy reads: http://www.amazon.com/Risk-John-Adams/dp/1857280687/ref=sr_1_2?s=books&ie=UTF8&qid=1320955190&sr=1-2 or http://www.amazon.com/Bad-Science-Quacks-Pharma-Flacks/dp/0771035799/ref=sr_1_2?s=books&ie=UTF8&qid=1320955405&sr=1-2 There will be a lot of possible books out there for you, it comes back to what you want to achieve in your course. Cheers Michelle UNCLASSIFIED From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Art Kendall I do not have a textbook to recommend.
===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD ************************************************************************************* Scanned by Clearswift SECURE Email Gateway at Food Standards ANZ. |
Re: Statistics Teaching Questions [SEC: UNCLASSIFIED]
|
This article is worth a look as well: 20 Questions A Journalist Should Ask About Poll Results I always worry about MBA-type courses: they seem to churn people through stats/spss modules, who may well collect their assessment marks, but afterwards have learned practically nothing about statistics or research methods. As John Tukey once said, “All the statistics in the world won’t save you if you asked the wrong question in the first place” John F Hall From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Gosse, Michelle I agree with what Art has to say below. The worst thing you could do is have them finish the course and think they are statisticians. An awful lot of my work through various jobs has been data cleaning and manipulating, to get data into the correct form in which to analyse. If you could also pass onto them that statistics isn’t just sitting in front of a computer and having the program spit out the correct results, but that – in many cases – most of the time will be in cleaning data and a minor portion will be analysing/reporting, that would be helpful as well. Do any of them need to be able to interpret published statistics, e.g. in journal articles? If so, this book might be helpful in teaching them what detail should be available in the publication: I also liked books along this line for showing how important good statistics are, how easily people can be fooled by “common sense” and/or bad statistical practice, and are easy reads: http://www.amazon.com/Risk-John-Adams/dp/1857280687/ref=sr_1_2?s=books&ie=UTF8&qid=1320955190&sr=1-2 or http://www.amazon.com/Bad-Science-Quacks-Pharma-Flacks/dp/0771035799/ref=sr_1_2?s=books&ie=UTF8&qid=1320955405&sr=1-2 There will be a lot of possible books out there for you, it comes back to what you want to achieve in your course. Cheers Michelle UNCLASSIFIED From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Art Kendall I do not have a textbook to recommend.
===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD ************************************************************************************* Scanned by Clearswift SECURE Email Gateway at Food Standards ANZ. |
Re: Statistics Teaching Questions [SEC: UNCLASSIFIED]
|
|
I agree with John Hall's post quoting Tukey. Most of these students will never write a line of code in their work life. The opportunity to use statistics as a tool to answer questions is important as long as they know the right questions to ask. I think that an MBA student would be well served to learn about the question, the instruments and methods used to gather data, or the quality of the data itself. In a business environment, the data we have is often not the data we want or need. The data we need is often unavailable. So some basic statistical methods are important, but it is equally important for students to understand what these conclusions mean or what they do not mean.
Orlando
On Fri, Nov 11, 2011 at 4:33 AM, John F Hall <[hidden email]> wrote:
|
Re: Statistics Teaching Questions [SEC: UNCLASSIFIED]
|
|
I have taught both undergraduates and graduates about quantitative methods. The main issue is as Orlando and John indicated. Don't try to teach them how to code. Teach them how to recognize techniques and what to look for as far as understanding results. Most of them have no background in statistics and would have a hard time telling you the difference between mean, median, and mode. They need to be able to look at what is being done by an analyst and have some understanding of what they are saying. As far as using SPSS or STATA or any of the other programs, they will never be at a level where they will be doing that. They will hire someone to do it. I normally teach them basic stats through multivariate linear regression, if I have time I will even show them things beside OLS regression. That usually so taxes their brains that they can't go any further. Joe Thornton, Adjunct Professor of Management Bellarmine University Doctoral Candidate, Case Western Reserve University Weatherhead School of Management From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Orlando Villella I agree with John Hall's post quoting Tukey. Most of these students will never write a line of code in their work life. The opportunity to use statistics as a tool to answer questions is important as long as they know the right questions to ask. I think that an MBA student would be well served to learn about the question, the instruments and methods used to gather data, or the quality of the data itself. In a business environment, the data we have is often not the data we want or need. The data we need is often unavailable. So some basic statistical methods are important, but it is equally important for students to understand what these conclusions mean or what they do not mean. Orlando On Fri, Nov 11, 2011 at 4:33 AM, John F Hall <[hidden email]> wrote: This article is worth a look as well: 20 Questions A Journalist Should Ask About Poll Results I always worry about MBA-type courses: they seem to churn people through stats/spss modules, who may well collect their assessment marks, but afterwards have learned practically nothing about statistics or research methods. As John Tukey once said, “All the statistics in the world won’t save you if you asked the wrong question in the first place” John F Hall From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Gosse, Michelle I agree with what Art has to say below. The worst thing you could do is have them finish the course and think they are statisticians. An awful lot of my work through various jobs has been data cleaning and manipulating, to get data into the correct form in which to analyse. If you could also pass onto them that statistics isn’t just sitting in front of a computer and having the program spit out the correct results, but that – in many cases – most of the time will be in cleaning data and a minor portion will be analysing/reporting, that would be helpful as well. Do any of them need to be able to interpret published statistics, e.g. in journal articles? If so, this book might be helpful in teaching them what detail should be available in the publication: I also liked books along this line for showing how important good statistics are, how easily people can be fooled by “common sense” and/or bad statistical practice, and are easy reads: http://www.amazon.com/Risk-John-Adams/dp/1857280687/ref=sr_1_2?s=books&ie=UTF8&qid=1320955190&sr=1-2 or http://www.amazon.com/Bad-Science-Quacks-Pharma-Flacks/dp/0771035799/ref=sr_1_2?s=books&ie=UTF8&qid=1320955405&sr=1-2 There will be a lot of possible books out there for you, it comes back to what you want to achieve in your course. Cheers Michelle UNCLASSIFIED From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Art Kendall I do not have a textbook to recommend.
===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD ************************************************************************************* Scanned by Clearswift SECURE Email Gateway at Food Standards ANZ. |
Re: Statistics Teaching Questions [SEC: UNCLASSIFIED]
|
|
Regression? Forget it. My old boss, the late Dr Mark Abrams, used to say, “If it’s worth saying, you can say with percentages.” Why do so very few SPSS courses deal with tabulation? Mind you, these days, Mark would have loved the graphics. Have a look at Hans Zeisel, “Say it with Figures” or even Morris Rosenberg, “The Logic of Survey Analysis”. John F Hall From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Joe Thornton I have taught both undergraduates and graduates about quantitative methods. The main issue is as Orlando and John indicated. Don't try to teach them how to code. Teach them how to recognize techniques and what to look for as far as understanding results. Most of them have no background in statistics and would have a hard time telling you the difference between mean, median, and mode. They need to be able to look at what is being done by an analyst and have some understanding of what they are saying. As far as using SPSS or STATA or any of the other programs, they will never be at a level where they will be doing that. They will hire someone to do it. I normally teach them basic stats through multivariate linear regression, if I have time I will even show them things beside OLS regression. That usually so taxes their brains that they can't go any further. Joe Thornton, Adjunct Professor of Management Bellarmine University Doctoral Candidate, Case Western Reserve University Weatherhead School of Management From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Orlando Villella I agree with John Hall's post quoting Tukey. Most of these students will never write a line of code in their work life. The opportunity to use statistics as a tool to answer questions is important as long as they know the right questions to ask. I think that an MBA student would be well served to learn about the question, the instruments and methods used to gather data, or the quality of the data itself. In a business environment, the data we have is often not the data we want or need. The data we need is often unavailable. So some basic statistical methods are important, but it is equally important for students to understand what these conclusions mean or what they do not mean. Orlando On Fri, Nov 11, 2011 at 4:33 AM, John F Hall <[hidden email]> wrote: This article is worth a look as well: 20 Questions A Journalist Should Ask About Poll Results I always worry about MBA-type courses: they seem to churn people through stats/spss modules, who may well collect their assessment marks, but afterwards have learned practically nothing about statistics or research methods. As John Tukey once said, “All the statistics in the world won’t save you if you asked the wrong question in the first place” John F Hall From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Gosse, Michelle I agree with what Art has to say below. The worst thing you could do is have them finish the course and think they are statisticians. An awful lot of my work through various jobs has been data cleaning and manipulating, to get data into the correct form in which to analyse. If you could also pass onto them that statistics isn’t just sitting in front of a computer and having the program spit out the correct results, but that – in many cases – most of the time will be in cleaning data and a minor portion will be analysing/reporting, that would be helpful as well. Do any of them need to be able to interpret published statistics, e.g. in journal articles? If so, this book might be helpful in teaching them what detail should be available in the publication: I also liked books along this line for showing how important good statistics are, how easily people can be fooled by “common sense” and/or bad statistical practice, and are easy reads: http://www.amazon.com/Risk-John-Adams/dp/1857280687/ref=sr_1_2?s=books&ie=UTF8&qid=1320955190&sr=1-2 or http://www.amazon.com/Bad-Science-Quacks-Pharma-Flacks/dp/0771035799/ref=sr_1_2?s=books&ie=UTF8&qid=1320955405&sr=1-2 There will be a lot of possible books out there for you, it comes back to what you want to achieve in your course. Cheers Michelle UNCLASSIFIED From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Art Kendall I do not have a textbook to recommend.
===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD ************************************************************************************* Scanned by Clearswift SECURE Email Gateway at Food Standards ANZ. |
Re: Statistics Teaching Questions [SEC: UNCLASSIFIED]
|
|
Many MBAs today want to learn about data mining and prediction. Models contain scale and nominal data. I wouldn’t teach them how to code but how to understand and critically evaluate predictive models built from things like logistic regression and C&RT. From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of John F Hall Regression? Forget it. My old boss, the late Dr Mark Abrams, used to say, “If it’s worth saying, you can say with percentages.” Why do so very few SPSS courses deal with tabulation? Mind you, these days, Mark would have loved the graphics. Have a look at Hans Zeisel, “Say it with Figures” or even Morris Rosenberg, “The Logic of Survey Analysis”. John F Hall From: SPSSX(r) Discussion [hidden email] On Behalf Of Joe Thornton I have taught both undergraduates and graduates about quantitative methods. The main issue is as Orlando and John indicated. Don't try to teach them how to code. Teach them how to recognize techniques and what to look for as far as understanding results. Most of them have no background in statistics and would have a hard time telling you the difference between mean, median, and mode. They need to be able to look at what is being done by an analyst and have some understanding of what they are saying. As far as using SPSS or STATA or any of the other programs, they will never be at a level where they will be doing that. They will hire someone to do it. I normally teach them basic stats through multivariate linear regression, if I have time I will even show them things beside OLS regression. That usually so taxes their brains that they can't go any further. Joe Thornton, Adjunct Professor of Management Bellarmine University Doctoral Candidate, Case Western Reserve University Weatherhead School of Management From: SPSSX(r) Discussion [hidden email] On Behalf Of Orlando Villella I agree with John Hall's post quoting Tukey. Most of these students will never write a line of code in their work life. The opportunity to use statistics as a tool to answer questions is important as long as they know the right questions to ask. I think that an MBA student would be well served to learn about the question, the instruments and methods used to gather data, or the quality of the data itself. In a business environment, the data we have is often not the data we want or need. The data we need is often unavailable. So some basic statistical methods are important, but it is equally important for students to understand what these conclusions mean or what they do not mean. Orlando On Fri, Nov 11, 2011 at 4:33 AM, John F Hall <[hidden email]> wrote: This article is worth a look as well: 20 Questions A Journalist Should Ask About Poll Results I always worry about MBA-type courses: they seem to churn people through stats/spss modules, who may well collect their assessment marks, but afterwards have learned practically nothing about statistics or research methods. As John Tukey once said, “All the statistics in the world won’t save you if you asked the wrong question in the first place” John F Hall From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Gosse, Michelle I agree with what Art has to say below. The worst thing you could do is have them finish the course and think they are statisticians. An awful lot of my work through various jobs has been data cleaning and manipulating, to get data into the correct form in which to analyse. If you could also pass onto them that statistics isn’t just sitting in front of a computer and having the program spit out the correct results, but that – in many cases – most of the time will be in cleaning data and a minor portion will be analysing/reporting, that would be helpful as well. Do any of them need to be able to interpret published statistics, e.g. in journal articles? If so, this book might be helpful in teaching them what detail should be available in the publication: I also liked books along this line for showing how important good statistics are, how easily people can be fooled by “common sense” and/or bad statistical practice, and are easy reads: http://www.amazon.com/Risk-John-Adams/dp/1857280687/ref=sr_1_2?s=books&ie=UTF8&qid=1320955190&sr=1-2 or http://www.amazon.com/Bad-Science-Quacks-Pharma-Flacks/dp/0771035799/ref=sr_1_2?s=books&ie=UTF8&qid=1320955405&sr=1-2 There will be a lot of possible books out there for you, it comes back to what you want to achieve in your course. Cheers Michelle UNCLASSIFIED From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Art Kendall I do not have a textbook to recommend.
===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD ************************************************************************************* Scanned by Clearswift SECURE Email Gateway at Food Standards ANZ. |
| Free forum by Nabble | Edit this page |