Re: [tips] Teach statistics before calculus

[Christopher Green]

On Oct 4, 2014, at 9:29 AM, Ken Steele  wrote:

> 
> My general phrase when working through some calculations is "Hey, we're 
> psychologists, and psychologists aren't afraid of fractions.”
> 
A fine sentiment, Ken. Alas, when I gave my brand new stats class a little 
arithmetic quiz to start off the year (last month), the mean score on the 
“fractions, proportions, and percentages” section was a disappointing 5.65/10. 
So, if they aren't afraid of fractions, perhaps they should be, and then do 
something about it. 

Dolefully,
Chris
…..
Christopher D Green
Department of Psychology
York University
Toronto, ON M3J 1P#
Canada

chri...@yorku.ca
http://www.yorku.ca/christo
...


---
You are currently subscribed to tips as: arch...@mail-archive.com.
To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=38889
or send a blank email to 
leave-38889-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu

Re: [tips] Teach statistics before calculus

[Ken Steele]

On 10/4/2014 1:02 AM, drnanjo wrote:



Math is beautiful. When we suggest to students that it's an
obstacle to be skirted and not a skill set to be valued and
cultivated, we do them an injustice. It's part of being a good
critical thinker and a competent consumer of information in general.
It teaches us to break complex problems into small steps, which
is generally the way these sorts of problems are solved in many
realms.

Done with my speechifying. Have a good weekend all.

Nancy Melucci
Long Beach City College and elsewhere




I agree with Nancy.

One thing I see in psychology classes is the general look of 
horror when I write an equation on the board.  I take a portion 
of my teaching mission to be to show that using mathematical 
expressions is another way of thinking through problems.


My general phrase when working through some calculations is "Hey, 
we're psychologists, and psychologists aren't afraid of fractions."


Ken

--
---
Kenneth M. Steele, Ph.D.  steel...@appstate.edu
Professor
Department of Psychology  http://www.psych.appstate.edu
Appalachian State University
Boone, NC 28608
USA
---


---
You are currently subscribed to tips as: arch...@mail-archive.com.
To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=3
or send a blank email to 
leave-3-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu

Re: [tips] Teach statistics before calculus

[Christopher Green]

> On Oct 4, 2014, at 1:02 AM, drnanjo  wrote:
>  
> I am so glad some folks here agree with meI could never learn math 
> without doing math. I can't teach mathematically based concepts without 
> calculation demos that require students to work along with me.
> 
>  I've run into the insistence and taken some flack, on this list in the past 
> and in other places, that doing math is just an unnecessary waste of time for 
> statistics students.


An article I just ran across in favor of actually doing math, as opposed to 
(only) doing "conceptual" explanation in math(ish) courses. Essentially, the 
upshot is that not really doing math means not really being able to do math 
(despite what you might think you "understand").
http://nautil.us/issue/17/big-bangs/how-i-rewired-my-brain-to-become-fluent-in-math-rd
 


Chris
...
Christopher D Green
Department of Psychology
York University
Toronto, ON   M3J 1P3

chri...@yorku.ca
http://www.yorku.ca/christo


---
You are currently subscribed to tips as: arch...@mail-archive.com.
To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=38886
or send a blank email to 
leave-38886-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu

Re: [tips] Teach statistics before calculus

[drnanjo]

I am so glad some folks here agree with meI could never learn math without 
doing math. I can't teach mathematically based concepts without calculation 
demos that require students to work along with me.


 I've run into the insistence and taken some flack, on this list in the past 
and in other places, that doing math is just an unnecessary waste of time for 
statistics students.
In less charitable moments I've imagined that those issuing the smack down to 
me are people who didn't like math themselves.


I am indulging my interest in mathematics lately and also taking classes both 
in doing it and understanding the cognitive skillsI believe there is much 
in the American system of education and teaching math that works to undermine 
confidence and mastery...some of it is cultural too. Americans generally value 
speed above all (not helpful in building math skill) AND hold the belief that 
if it doesn't come easily to you, you weren't meant to do it (a perversion I 
suppose of our love of "talent".)


Finally I believe there is a false dichotomy here...between speeding through 
the procedures using the various software packages and doing "long tedious 
calculations." 
This is not so. 


I've developed many short examples using simple numbers that produce (usually) 
round or not-too-lengthy decimal answers for sum of squares (someone mentioned 
that), basic descriptive statistics and even some of the tests like one-way 
ANOVA and one/two sample t and z.


Math is beautiful. When we suggest to students that it's an obstacle to be 
skirted and not a skill set to be valued and cultivated, we do them an 
injustice. It's part of being a good critical thinker and a competent consumer 
of information in general.
It teaches us to break complex problems into small steps, which is generally 
the way these sorts of problems are solved in many realms.


Done with my speechifying. Have a good weekend all.


Nancy Melucci
Long Beach City College and elsewhere



-Original Message-
From: Christopher Green 
To: Teaching in the Psychological Sciences (TIPS) 
Sent: Wed, Oct 1, 2014 6:41 am
Subject: Re: [tips] Teach statistics before calculus


That is good Jim, and I do the same thing, but it is only a start. Most of the 
stats actually used in psychological research is continuous, not discrete. 

As for stats courses that allow student to depend on computers, and never do 
hand calculations using formulas… this student have not learned statistics. 
They 
have learned data-entry. 

Chris
…..
Christopher D Green
Department of Psychology
York University
Toronto, ON M3J 1P#
Canada

chri...@yorku.ca
http://www.yorku.ca/christo
...

On Oct 1, 2014, at 8:01 AM, Jim Clark  wrote:

> Hi
> 
> One reason I like probability and binomial in intro stats (despite student 
protest) is that students can determine p distribution if H0 true, without 
calculus. And provides a concrete foundation for normal distribution, which 
binomial approaches as n increases.
> 
> Also possible to use simulations to show the resulting probability 
distributions agree nicely with theoretical ps produced by calculus.
> 
> Although calculus allows for alternative (deeper?) understanding of 
distributions, not clear that it is necessary.
> 
> Of course, calculus has other benefits for stats, such as proof that SS is a 
minimum.
> 
> Take care
> Jim
> 
> Sent from my iPhone
> 
>> On Oct 1, 2014, at 5:42 AM, "Miguel Roig"  wrote:
>> 
>> Chris, I believe that we have had discussions here about the practicality of 
teaching students to do hand calculations from formulas given the wide 
availability of statistical software. If so, do those who learn in classes in 
which the emphasis is primarily conceptual and software-based even taught how 
to 
look up p values in those tables? :)
>> 
>> Miguel
>> 
>> -Original Message-
>> From: Christopher Green [mailto:chri...@yorku.ca] 
>> Sent: Tuesday, September 30, 2014 6:15 PM
>> To: Teaching in the Psychological Sciences (TIPS)
>> Subject: Re: [tips] Teach statistics before calculus
>> 
>> Funny, just earlier today I was saying that the reason so many students have 
so much trouble with the (continuous) statistics that we teach in psychology is 
that we are essentially trying to teach them a topic that requires a knowledge 
of calculus without making them take calculus first. That's what all those 
tables in the back of the book are: they integrate over probability 
distributions so that we can lookup (rather than calculate directly) the 
proportion of area up to a given x-axis value (z, t, F, chi-square, etc.).
>> 
>> So this guy might be right that stats is at the top of the pyramid, but only 
because it passes directly through calculus. 
>> 
>> Chris
>> .

Re:[tips] Teach statistics before calculus

[Pollak, Edward (Retired)]

Claudia Stanny wrote, "I once served on a committee for a master's thesis in 
which the student,  disappointed that his findings produced non-significant 
p-values, tried
some alternative analyses in SPSS. He showed up one day, beaming, with an
output that he thought indicated he had a "significant 3 factor solution"
for his data. Problem: his data were based on a factorial design. [Perhaps
that explains why he thought a factor analysis would be a good idea.
(sigh)]"

I'll go you one better, Claudia. I did a post doc in a bio dept.  in which the 
graduate statistics course was, basically, "here's how you use SPSS" (or 
whatever program they were using in 1976).  Although I had been, at best, a 
mediocre statistics student, I because the ersatz departmental statistician. 
The worst moment was when a doctoral candidate had trouble understanding why he 
couldn't do an ANOVA on his data when one of the groups had an N=1.




Edward I. Pollak, Ph.D.
Professor Emeritus of Psychology
West Chester University of Pennsylvania
http://home.comcast.net/~epollak/
Husband, father, grandfather, bluegrass fiddler, banjoist & 
biopsychologist... in approximate order of importance



---
You are currently subscribed to tips as: arch...@mail-archive.com.
To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=38849
or send a blank email to 
leave-38849-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu

Re: [tips] Teach statistics before calculus

[don allen]

I agree that a major difficulty in getting students to understand what results 
mean is get them past the point where they feel that if they haven't obtained a 
significant p value then they "didn't find anything". One of the examples that 
I used to try to help them understand this was to cite a UBC study of female 
students who had chosen to go on a vegan diet. (sorry no source available right 
now). The two most common reasons for making this choice were, "It's healthier" 
and "It will help me lose weight". Analyses showed that, when compared to women 
who maintained a regular diet, there was NO significant difference between the 
two groups on either measure. This example seemed to help a lot of them 
understand that "not significant" doesn't mean "not meaningful". Or as Enrico 
Fermi used to say:

"There are two possible outcomes: if the result confirms the hypothesis, then 
you've made a measurement. If the result is contrary to the hypothesis, then 
you've made a discovery."

-Don.

- Original Message -
From: "Claudia Stanny" 
To: "Teaching in the Psychological Sciences (TIPS)" 
Sent: Wednesday, October 1, 2014 1:39:55 PM
Subject: Re: [tips] Teach statistics before calculus









I once served on a committee for a master's thesis in which the student, 
disappointed that his findings produced non-significant p-values, tried some 
alternative analyses in SPSS. He showed up one day, beaming, with an output 
that he thought indicated he had a "significant 3 factor solution" for his 
data. Problem: his data were based on a factorial design. [Perhaps that 
explains why he thought a factor analysis would be a good idea. (sigh)] The 
committee chair threw his output in the trash. Then we talked about the 
research story he had to tell based on non-significant results. 



Some committee members contribute to this problem by demanding significant 
results for a satisfactory project. I recall an undergraduate who was trying to 
complete an honor's thesis (data collection long finished) and was "shopping" 
among the stats faculty for the golden ticket of alpha < .05. She had an 
appointment with one faculty member, but cancelled it with a note on his door 
that said basically said "never mind, someone else found a test that worked." 




_ 

Claudia J. Stanny, Ph.D. 
Director 
Center for University Teaching, Learning, and Assessment 
University of West Florida 

Pensacola, FL 32514 

Phone: (850) 857-6355 (direct) or 473-7435 (CUTLA) 

csta...@uwf.edu 

CUTLA Web Site: http://uwf.edu/offices/cutla/ 
Personal Web Pages: http://uwf.edu/cstanny/website/index.htm 




--- 

You are currently subscribed to tips as: dap...@shaw.ca . 

To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13157.966b795bc7f3ccb35e3da08aebe98f18&n=T&l=tips&o=38828
 

(It may be necessary to cut and paste the above URL if the line is broken) 

or send a blank email to 
leave-38828-13157.966b795bc7f3ccb35e3da08aebe98...@fsulist.frostburg.edu 



---
You are currently subscribed to tips as: arch...@mail-archive.com.
To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=38830
or send a blank email to 
leave-38830-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu

Re: [tips] Teach statistics before calculus

[Claudia Stanny]

I once served on a committee for a master's thesis in which the student,
disappointed that his findings produced non-significant p-values, tried
some alternative analyses in SPSS. He showed up one day, beaming, with an
output that he thought indicated he had a "significant 3 factor solution"
for his data. Problem: his data were based on a factorial design. [Perhaps
that explains why he thought a factor analysis would be a good idea.
(sigh)] The committee chair threw his output in the trash. Then we talked
about the research story he had to tell based on non-significant results.

Some committee members contribute to this problem by demanding significant
results for a satisfactory project. I recall an undergraduate who was
trying to complete an honor's thesis (data collection long finished) and
was "shopping" among the stats faculty for the golden ticket of alpha <
.05. She had an appointment with one faculty member, but cancelled it with
a note on his door that said basically said "never mind, someone else found
a test that worked."

_

Claudia J. Stanny, Ph.D.
Director
Center for University Teaching, Learning, and Assessment
University of West Florida
Pensacola, FL  32514

Phone:   (850) 857-6355 (direct) or  473-7435 (CUTLA)

csta...@uwf.edu

CUTLA Web Site: http://uwf.edu/offices/cutla/ 
Personal Web Pages: http://uwf.edu/cstanny/website/index.htm

---
You are currently subscribed to tips as: arch...@mail-archive.com.
To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=38828
or send a blank email to 
leave-38828-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu

Re: [tips] Teach statistics before calculus

[Michael Scoles]

When I was in graduate school, there was one area in the department that
always had doctoral students coming to our lab in the experimental wing
with stacks of old green-bar BMDP printouts.  A common question was,
"These are the results from my dissertation.  Can you tell me what they
mean?"
 
Since I didn't help design the study or collect the data, the usual
answer was, "No."  BMDP, Minitab, SPSS, SAS, etc. do encourage learning
data-entry skills.  Add the GUI features, and you can spend all day
clicking on analyses with no understanding of what they mean.
 
I'm a calculator/Excel person.  Sometimes, I even ask students to do
complex calculations (like what is 1/2 of 1/3) in their heads!
 


Michael T. Scoles, Ph.D.
Associate Professor of Psychology & Counseling
University of Central Arkansas
Conway, AR 72035
Phone: 501-450-5418
Fax: 501-450-5424
 
AVID: UCA dedicates itself to Academic Vitality, Integrity, and
Diversity.


>>> Christopher Green  10/1/2014 1:54 PM >>>
In all honesty, I think there is nothing like plugging number into
formulas to make one really understand then internal mechanics of
statistical (and other mathematical/computational) procedures. If all
you do is line the data up and out pops and answer on the other side,
nothing was learned. It might as well have been magic.

Chris
*..
Christopher D Green
Department of Psychology
York University
Toronto, ON M3J 1P#
Canada

chri...@yorku.ca
http://www.yorku.ca/christo
...

On Oct 1, 2014, at 10:07 AM, Marc Carter 
wrote:

> I'd agree, withte proviso that "by hand" allows things like excel. 
One can build formulas in excel and not have to do the tedious and
time-wasting computations we all had to do (who thinks it's important to
compute sums of squares with hand calculators anymore?).  Time is better
spent, as far as I can tell, talking about the mathematical objects
(what *is* a sum of suqares?  how does it behave?) than using up time
calculating it.
> 
> But, that's just my two cents.
> 
> m
> 
> From: Christopher Green 
> Sent: Wednesday, October 1, 2014 8:40 AM
> To: Teaching in the Psychological Sciences (TIPS)
> Subject: Re: [tips] Teach statistics before calculus
> 
> That is good Jim, and I do the same thing, but it is only a start.
Most of the stats actually used in psychological research is continuous,
not discrete.
> 
> As for stats courses that allow student to depend on computers, and
never do hand calculations using formulas* this student have not learned
statistics. They have learned data-entry.
> 
> Chris
> *..
> Christopher D Green
> Department of Psychology
> York University
> Toronto, ON M3J 1P#
> Canada
> 
> chri...@yorku.ca
> http://www.yorku.ca/christo
> ...
> 
> On Oct 1, 2014, at 8:01 AM, Jim Clark  wrote:
> 
>> Hi
>> 
>> One reason I like probability and binomial in intro stats (despite
student protest) is that students can determine p distribution if H0
true, without calculus. And provides a concrete foundation for normal
distribution, which binomial approaches as n increases.
>> 
>> Also possible to use simulations to show the resulting probability
distributions agree nicely with theoretical ps produced by calculus.
>> 
>> Although calculus allows for alternative (deeper?) understanding of
distributions, not clear that it is necessary.
>> 
>> Of course, calculus has other benefits for stats, such as proof that
SS is a minimum.
>> 
>> Take care
>> Jim
>> 
>> Sent from my iPhone
>> 
>>> On Oct 1, 2014, at 5:42 AM, "Miguel Roig" 
wrote:
>>> 
>>> Chris, I believe that we have had discussions here about the
practicality of teaching students to do hand calculations from formulas
given the wide availability of statistical software. If so, do those who
learn in classes in which the emphasis is primarily conceptual and
software-based even taught how to look up p values in those tables? :)
>>> 
>>> Miguel
>>> 
>>> -Original Message-
>>> From: Christopher Green [mailto:chri...@yorku.ca]
>>> Sent: Tuesday, September 30, 2014 6:15 PM
>>> To: Teaching in the Psychological Sciences (TIPS)
>>> Subject: Re: [tips] Teach statistics before calculus
>>> 
>>> Funny, just earlier today I was saying that the reason so many
students have so much trouble with the (continuous) statistics that we
teach in psychology is that we are essentially trying to teach them a
topic that requires a knowledge of calculus without making them take
calculus first. That's what all those tables in the back of the book
are: they integrate over probability distributions so that we can lookup
(rather than ca

Re: [tips] Teach statistics before calculus

[Christopher Green]

In all honesty, I think there is nothing like plugging number into formulas to 
make one really understand then internal mechanics of statistical (and other 
mathematical/computational) procedures. If all you do is line the data up and 
out pops and answer on the other side, nothing was learned. It might as well 
have been magic.

Chris
…..
Christopher D Green
Department of Psychology
York University
Toronto, ON M3J 1P#
Canada

chri...@yorku.ca
http://www.yorku.ca/christo
...

On Oct 1, 2014, at 10:07 AM, Marc Carter  wrote:

> I'd agree, withte proviso that "by hand" allows things like excel.  One can 
> build formulas in excel and not have to do the tedious and time-wasting 
> computations we all had to do (who thinks it's important to compute sums of 
> squares with hand calculators anymore?).  Time is better spent, as far as I 
> can tell, talking about the mathematical objects (what *is* a sum of suqares? 
>  how does it behave?) than using up time calculating it.
> 
> But, that's just my two cents.
> 
> m
> 
> From: Christopher Green 
> Sent: Wednesday, October 1, 2014 8:40 AM
> To: Teaching in the Psychological Sciences (TIPS)
> Subject: Re: [tips] Teach statistics before calculus
> 
> That is good Jim, and I do the same thing, but it is only a start. Most of 
> the stats actually used in psychological research is continuous, not discrete.
> 
> As for stats courses that allow student to depend on computers, and never do 
> hand calculations using formulas… this student have not learned statistics. 
> They have learned data-entry.
> 
> Chris
> …..
> Christopher D Green
> Department of Psychology
> York University
> Toronto, ON M3J 1P#
> Canada
> 
> chri...@yorku.ca
> http://www.yorku.ca/christo
> ...
> 
> On Oct 1, 2014, at 8:01 AM, Jim Clark  wrote:
> 
>> Hi
>> 
>> One reason I like probability and binomial in intro stats (despite student 
>> protest) is that students can determine p distribution if H0 true, without 
>> calculus. And provides a concrete foundation for normal distribution, which 
>> binomial approaches as n increases.
>> 
>> Also possible to use simulations to show the resulting probability 
>> distributions agree nicely with theoretical ps produced by calculus.
>> 
>> Although calculus allows for alternative (deeper?) understanding of 
>> distributions, not clear that it is necessary.
>> 
>> Of course, calculus has other benefits for stats, such as proof that SS is a 
>> minimum.
>> 
>> Take care
>> Jim
>> 
>> Sent from my iPhone
>> 
>>> On Oct 1, 2014, at 5:42 AM, "Miguel Roig"  wrote:
>>> 
>>> Chris, I believe that we have had discussions here about the practicality 
>>> of teaching students to do hand calculations from formulas given the wide 
>>> availability of statistical software. If so, do those who learn in classes 
>>> in which the emphasis is primarily conceptual and software-based even 
>>> taught how to look up p values in those tables? :)
>>> 
>>> Miguel
>>> 
>>> -Original Message-
>>> From: Christopher Green [mailto:chri...@yorku.ca]
>>> Sent: Tuesday, September 30, 2014 6:15 PM
>>> To: Teaching in the Psychological Sciences (TIPS)
>>> Subject: Re: [tips] Teach statistics before calculus
>>> 
>>> Funny, just earlier today I was saying that the reason so many students 
>>> have so much trouble with the (continuous) statistics that we teach in 
>>> psychology is that we are essentially trying to teach them a topic that 
>>> requires a knowledge of calculus without making them take calculus first. 
>>> That's what all those tables in the back of the book are: they integrate 
>>> over probability distributions so that we can lookup (rather than calculate 
>>> directly) the proportion of area up to a given x-axis value (z, t, F, 
>>> chi-square, etc.).
>>> 
>>> So this guy might be right that stats is at the top of the pyramid, but 
>>> only because it passes directly through calculus.
>>> 
>>> Chris
>>> .
>>> Christopher D Green
>>> Department of Psychology
>>> York University
>>> Toronto, ON M3J 1P#
>>> Canada
>>> 
>>> chri...@yorku.ca
>>> http://www.yorku.ca/christo
>>> ...
>>> 
>>>> On Sep 30, 2014, at 5:04 PM, Miguel Roig  wrote:
>>>> 
>>>> It's only 3 minutes long and there is 

RE: [tips] Teach statistics before calculus

[Marc Carter]

I'd agree, withte proviso that "by hand" allows things like excel.  One can 
build formulas in excel and not have to do the tedious and time-wasting 
computations we all had to do (who thinks it's important to compute sums of 
squares with hand calculators anymore?).  Time is better spent, as far as I can 
tell, talking about the mathematical objects (what *is* a sum of suqares?  how 
does it behave?) than using up time calculating it.

But, that's just my two cents.

m

From: Christopher Green 
Sent: Wednesday, October 1, 2014 8:40 AM
To: Teaching in the Psychological Sciences (TIPS)
Subject: Re: [tips] Teach statistics before calculus

That is good Jim, and I do the same thing, but it is only a start. Most of the 
stats actually used in psychological research is continuous, not discrete.

As for stats courses that allow student to depend on computers, and never do 
hand calculations using formulas… this student have not learned statistics. 
They have learned data-entry.

Chris
…..
Christopher D Green
Department of Psychology
York University
Toronto, ON M3J 1P#
Canada

chri...@yorku.ca
http://www.yorku.ca/christo
...

On Oct 1, 2014, at 8:01 AM, Jim Clark  wrote:

> Hi
>
> One reason I like probability and binomial in intro stats (despite student 
> protest) is that students can determine p distribution if H0 true, without 
> calculus. And provides a concrete foundation for normal distribution, which 
> binomial approaches as n increases.
>
> Also possible to use simulations to show the resulting probability 
> distributions agree nicely with theoretical ps produced by calculus.
>
> Although calculus allows for alternative (deeper?) understanding of 
> distributions, not clear that it is necessary.
>
> Of course, calculus has other benefits for stats, such as proof that SS is a 
> minimum.
>
> Take care
> Jim
>
> Sent from my iPhone
>
>> On Oct 1, 2014, at 5:42 AM, "Miguel Roig"  wrote:
>>
>> Chris, I believe that we have had discussions here about the practicality of 
>> teaching students to do hand calculations from formulas given the wide 
>> availability of statistical software. If so, do those who learn in classes 
>> in which the emphasis is primarily conceptual and software-based even taught 
>> how to look up p values in those tables? :)
>>
>> Miguel
>>
>> -Original Message-
>> From: Christopher Green [mailto:chri...@yorku.ca]
>> Sent: Tuesday, September 30, 2014 6:15 PM
>> To: Teaching in the Psychological Sciences (TIPS)
>> Subject: Re: [tips] Teach statistics before calculus
>>
>> Funny, just earlier today I was saying that the reason so many students have 
>> so much trouble with the (continuous) statistics that we teach in psychology 
>> is that we are essentially trying to teach them a topic that requires a 
>> knowledge of calculus without making them take calculus first. That's what 
>> all those tables in the back of the book are: they integrate over 
>> probability distributions so that we can lookup (rather than calculate 
>> directly) the proportion of area up to a given x-axis value (z, t, F, 
>> chi-square, etc.).
>>
>> So this guy might be right that stats is at the top of the pyramid, but only 
>> because it passes directly through calculus.
>>
>> Chris
>> .
>> Christopher D Green
>> Department of Psychology
>> York University
>> Toronto, ON M3J 1P#
>> Canada
>>
>> chri...@yorku.ca
>> http://www.yorku.ca/christo
>> ...
>>
>>> On Sep 30, 2014, at 5:04 PM, Miguel Roig  wrote:
>>>
>>> It's only 3 minutes long and there is probably nothing here that you don't 
>>> already know, but I thought it was worth sharing.
>>>
>>> http://www.ted.com/talks/arthur_benjamin_s_formula_for_changing_math_education
>>>
>>> Miguel
>>> ___
>>> Miguel Roig, Ph.D.
>>> Professor of Psychology
>>> St. John's University
>>> 300 Howard Avenue
>>> Staten Island, New York 10301
>>> Voice: (718) 390-4513
>>> Fax: (718) 390-4347
>>> E-mail: ro...@stjohns.edu
>>> http://facpub.stjohns.edu/~roigm
>>> http://orcid.org/-0001-5311-5651
>>> On plagiarism and ethical writing: 
>>> http://ori.dhhs.gov/education/products/plagiarism/
>>> ___
>>> ---
>>> You are currently subscribed to tips as: chri...@yorku.ca

Re: [tips] Teach statistics before calculus

[Christopher Green]

That is good Jim, and I do the same thing, but it is only a start. Most of the 
stats actually used in psychological research is continuous, not discrete. 

As for stats courses that allow student to depend on computers, and never do 
hand calculations using formulas… this student have not learned statistics. 
They have learned data-entry. 

Chris
…..
Christopher D Green
Department of Psychology
York University
Toronto, ON M3J 1P#
Canada

chri...@yorku.ca
http://www.yorku.ca/christo
...

On Oct 1, 2014, at 8:01 AM, Jim Clark  wrote:

> Hi
> 
> One reason I like probability and binomial in intro stats (despite student 
> protest) is that students can determine p distribution if H0 true, without 
> calculus. And provides a concrete foundation for normal distribution, which 
> binomial approaches as n increases.
> 
> Also possible to use simulations to show the resulting probability 
> distributions agree nicely with theoretical ps produced by calculus.
> 
> Although calculus allows for alternative (deeper?) understanding of 
> distributions, not clear that it is necessary.
> 
> Of course, calculus has other benefits for stats, such as proof that SS is a 
> minimum.
> 
> Take care
> Jim
> 
> Sent from my iPhone
> 
>> On Oct 1, 2014, at 5:42 AM, "Miguel Roig"  wrote:
>> 
>> Chris, I believe that we have had discussions here about the practicality of 
>> teaching students to do hand calculations from formulas given the wide 
>> availability of statistical software. If so, do those who learn in classes 
>> in which the emphasis is primarily conceptual and software-based even taught 
>> how to look up p values in those tables? :)
>> 
>> Miguel
>> 
>> -Original Message-
>> From: Christopher Green [mailto:chri...@yorku.ca] 
>> Sent: Tuesday, September 30, 2014 6:15 PM
>> To: Teaching in the Psychological Sciences (TIPS)
>> Subject: Re: [tips] Teach statistics before calculus
>> 
>> Funny, just earlier today I was saying that the reason so many students have 
>> so much trouble with the (continuous) statistics that we teach in psychology 
>> is that we are essentially trying to teach them a topic that requires a 
>> knowledge of calculus without making them take calculus first. That's what 
>> all those tables in the back of the book are: they integrate over 
>> probability distributions so that we can lookup (rather than calculate 
>> directly) the proportion of area up to a given x-axis value (z, t, F, 
>> chi-square, etc.).
>> 
>> So this guy might be right that stats is at the top of the pyramid, but only 
>> because it passes directly through calculus. 
>> 
>> Chris
>> .
>> Christopher D Green
>> Department of Psychology
>> York University
>> Toronto, ON M3J 1P#
>> Canada
>> 
>> chri...@yorku.ca
>> http://www.yorku.ca/christo
>> ...
>> 
>>> On Sep 30, 2014, at 5:04 PM, Miguel Roig  wrote:
>>> 
>>> It's only 3 minutes long and there is probably nothing here that you don't 
>>> already know, but I thought it was worth sharing. 
>>> 
>>> http://www.ted.com/talks/arthur_benjamin_s_formula_for_changing_math_education
>>> 
>>> Miguel
>>> ___
>>> Miguel Roig, Ph.D.   
>>> Professor of Psychology
>>> St. John's University   
>>> 300 Howard Avenue   
>>> Staten Island, New York 10301 
>>> Voice: (718) 390-4513 
>>> Fax: (718) 390-4347 
>>> E-mail: ro...@stjohns.edu 
>>> http://facpub.stjohns.edu/~roigm
>>> http://orcid.org/-0001-5311-5651
>>> On plagiarism and ethical writing: 
>>> http://ori.dhhs.gov/education/products/plagiarism/
>>> ___
>>> ---
>>> You are currently subscribed to tips as: chri...@yorku.ca.
>>> To unsubscribe click here: 
>>> http://fsulist.frostburg.edu/u?id=430248.781165b5ef80a3cd2b14721caf62bd92&n=T&l=tips&o=38783
>>> or send a blank email to 
>>> leave-38783-430248.781165b5ef80a3cd2b14721caf62b...@fsulist.frostburg.edu
>>> 
>> 
>> 
>> ---
>> You are currently subscribed to tips as: ro...@stjohns.edu.
>> To unsubscribe click here: 
>> http://fsulist.frostburg.edu/u?id=1632838.7e62b84813297f170a6fc240dab8c12d&n=T&l=tips&o=387

Re: [tips] Teach statistics before calculus

[Jim Clark]

Hi

One reason I like probability and binomial in intro stats (despite student 
protest) is that students can determine p distribution if H0 true, without 
calculus. And provides a concrete foundation for normal distribution, which 
binomial approaches as n increases.

Also possible to use simulations to show the resulting probability 
distributions agree nicely with theoretical ps produced by calculus.

Although calculus allows for alternative (deeper?) understanding of 
distributions, not clear that it is necessary.

Of course, calculus has other benefits for stats, such as proof that SS is a 
minimum.

Take care
Jim

Sent from my iPhone

> On Oct 1, 2014, at 5:42 AM, "Miguel Roig"  wrote:
> 
> Chris, I believe that we have had discussions here about the practicality of 
> teaching students to do hand calculations from formulas given the wide 
> availability of statistical software. If so, do those who learn in classes in 
> which the emphasis is primarily conceptual and software-based even taught how 
> to look up p values in those tables? :)
> 
> Miguel
> 
> -Original Message-
> From: Christopher Green [mailto:chri...@yorku.ca] 
> Sent: Tuesday, September 30, 2014 6:15 PM
> To: Teaching in the Psychological Sciences (TIPS)
> Subject: Re: [tips] Teach statistics before calculus
> 
> Funny, just earlier today I was saying that the reason so many students have 
> so much trouble with the (continuous) statistics that we teach in psychology 
> is that we are essentially trying to teach them a topic that requires a 
> knowledge of calculus without making them take calculus first. That's what 
> all those tables in the back of the book are: they integrate over probability 
> distributions so that we can lookup (rather than calculate directly) the 
> proportion of area up to a given x-axis value (z, t, F, chi-square, etc.).
> 
> So this guy might be right that stats is at the top of the pyramid, but only 
> because it passes directly through calculus. 
> 
> Chris
> .
> Christopher D Green
> Department of Psychology
> York University
> Toronto, ON M3J 1P#
> Canada
> 
> chri...@yorku.ca
> http://www.yorku.ca/christo
> ...
> 
>> On Sep 30, 2014, at 5:04 PM, Miguel Roig  wrote:
>> 
>> It's only 3 minutes long and there is probably nothing here that you don't 
>> already know, but I thought it was worth sharing. 
>> 
>> http://www.ted.com/talks/arthur_benjamin_s_formula_for_changing_math_education
>> 
>> Miguel
>> ___
>> Miguel Roig, Ph.D.   
>> Professor of Psychology
>> St. John's University   
>> 300 Howard Avenue   
>> Staten Island, New York 10301 
>> Voice: (718) 390-4513 
>> Fax: (718) 390-4347 
>> E-mail: ro...@stjohns.edu 
>> http://facpub.stjohns.edu/~roigm
>> http://orcid.org/-0001-5311-5651
>> On plagiarism and ethical writing: 
>> http://ori.dhhs.gov/education/products/plagiarism/
>> ___
>> ---
>> You are currently subscribed to tips as: chri...@yorku.ca.
>> To unsubscribe click here: 
>> http://fsulist.frostburg.edu/u?id=430248.781165b5ef80a3cd2b14721caf62bd92&n=T&l=tips&o=38783
>> or send a blank email to 
>> leave-38783-430248.781165b5ef80a3cd2b14721caf62b...@fsulist.frostburg.edu
>> 
> 
> 
> ---
> You are currently subscribed to tips as: ro...@stjohns.edu.
> To unsubscribe click here: 
> http://fsulist.frostburg.edu/u?id=1632838.7e62b84813297f170a6fc240dab8c12d&n=T&l=tips&o=38787
> or send a blank email to 
> leave-38787-1632838.7e62b84813297f170a6fc240dab8c...@fsulist.frostburg.edu
> 
> ---
> You are currently subscribed to tips as: j.cl...@uwinnipeg.ca.
> To unsubscribe click here: 
> http://fsulist.frostburg.edu/u?id=13251.645f86b5cec4da0a56ffea7a891720c9&n=T&l=tips&o=38803
> or send a blank email to 
> leave-38803-13251.645f86b5cec4da0a56ffea7a89172...@fsulist.frostburg.edu

---
You are currently subscribed to tips as: arch...@mail-archive.com.
To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=38805
or send a blank email to 
leave-38805-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu

RE: [tips] Teach statistics before calculus

[Miguel Roig]

Chris, I believe that we have had discussions here about the practicality of 
teaching students to do hand calculations from formulas given the wide 
availability of statistical software. If so, do those who learn in classes in 
which the emphasis is primarily conceptual and software-based even taught how 
to look up p values in those tables? :)

Miguel

-Original Message-
From: Christopher Green [mailto:chri...@yorku.ca] 
Sent: Tuesday, September 30, 2014 6:15 PM
To: Teaching in the Psychological Sciences (TIPS)
Subject: Re: [tips] Teach statistics before calculus

Funny, just earlier today I was saying that the reason so many students have so 
much trouble with the (continuous) statistics that we teach in psychology is 
that we are essentially trying to teach them a topic that requires a knowledge 
of calculus without making them take calculus first. That's what all those 
tables in the back of the book are: they integrate over probability 
distributions so that we can lookup (rather than calculate directly) the 
proportion of area up to a given x-axis value (z, t, F, chi-square, etc.).

So this guy might be right that stats is at the top of the pyramid, but only 
because it passes directly through calculus. 

Chris
.
Christopher D Green
Department of Psychology
York University
Toronto, ON M3J 1P#
Canada

chri...@yorku.ca
http://www.yorku.ca/christo
...

On Sep 30, 2014, at 5:04 PM, Miguel Roig  wrote:

> It's only 3 minutes long and there is probably nothing here that you don't 
> already know, but I thought it was worth sharing. 
> 
> http://www.ted.com/talks/arthur_benjamin_s_formula_for_changing_math_education
> 
> Miguel
> ___
> Miguel Roig, Ph.D.   
> Professor of Psychology
> St. John's University   
> 300 Howard Avenue   
> Staten Island, New York 10301 
> Voice: (718) 390-4513 
> Fax: (718) 390-4347 
> E-mail: ro...@stjohns.edu 
> http://facpub.stjohns.edu/~roigm
> http://orcid.org/-0001-5311-5651
> On plagiarism and ethical writing: 
> http://ori.dhhs.gov/education/products/plagiarism/
> ___
> ---
> You are currently subscribed to tips as: chri...@yorku.ca.
> To unsubscribe click here: 
> http://fsulist.frostburg.edu/u?id=430248.781165b5ef80a3cd2b14721caf62bd92&n=T&l=tips&o=38783
> or send a blank email to 
> leave-38783-430248.781165b5ef80a3cd2b14721caf62b...@fsulist.frostburg.edu
> 


---
You are currently subscribed to tips as: ro...@stjohns.edu.
To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=1632838.7e62b84813297f170a6fc240dab8c12d&n=T&l=tips&o=38787
or send a blank email to 
leave-38787-1632838.7e62b84813297f170a6fc240dab8c...@fsulist.frostburg.edu

---
You are currently subscribed to tips as: arch...@mail-archive.com.
To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=38803
or send a blank email to 
leave-38803-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu

Re: [tips] Teach statistics before calculus

[Christopher Green]

Funny, just earlier today I was saying that the reason so many students have so 
much trouble with the (continuous) statistics that we teach in psychology is 
that we are essentially trying to teach them a topic that requires a knowledge 
of calculus without making them take calculus first. That’s what all those 
tables in the back of the book are: they integrate over probability 
distributions so that we can lookup (rather than calculate directly) the 
proportion of area up to a given x-axis value (z, t, F, chi-square, etc.).

So this guy might be right that stats is at the top of the pyramid, but only 
because it passes directly through calculus. 

Chris
…..
Christopher D Green
Department of Psychology
York University
Toronto, ON M3J 1P#
Canada

chri...@yorku.ca
http://www.yorku.ca/christo
...

On Sep 30, 2014, at 5:04 PM, Miguel Roig  wrote:

> It's only 3 minutes long and there is probably nothing here that you don't 
> already know, but I thought it was worth sharing. 
> 
> http://www.ted.com/talks/arthur_benjamin_s_formula_for_changing_math_education
> 
> Miguel
> ___
> Miguel Roig, Ph.D.   
> Professor of Psychology
> St. John's University   
> 300 Howard Avenue   
> Staten Island, New York 10301 
> Voice: (718) 390-4513 
> Fax: (718) 390-4347 
> E-mail: ro...@stjohns.edu 
> http://facpub.stjohns.edu/~roigm
> http://orcid.org/-0001-5311-5651
> On plagiarism and ethical writing: 
> http://ori.dhhs.gov/education/products/plagiarism/
> ___
> ---
> You are currently subscribed to tips as: chri...@yorku.ca.
> To unsubscribe click here: 
> http://fsulist.frostburg.edu/u?id=430248.781165b5ef80a3cd2b14721caf62bd92&n=T&l=tips&o=38783
> or send a blank email to 
> leave-38783-430248.781165b5ef80a3cd2b14721caf62b...@fsulist.frostburg.edu
> 


---
You are currently subscribed to tips as: arch...@mail-archive.com.
To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=38787
or send a blank email to 
leave-38787-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu