IA Type II -- Gold Medal Heights

[Nabz]

I have a question regarding this this:

1. Does it matter if years go on x axis and height go on y axis or vice versa?

2. The very first graph which is to be graphed using technology, do we have to find the gradient of that line through mx+b?

3. About the function we have to come up with, does it have to be without the regression because later it asks us to draw another function but using technology...i'm sort of stuck on this part because i can't seem to find the function and perimeter for the variables...

4. Is there a way i can look at a sample type 2 IA so that it gives me a clue how it should be written? because i've browsed the website here and i need a memebership in order to download those files :/

Please help me out on this, i only have this weekend to work on this then i'm pretty much dead.

[abhinavbhasin]

I have a question regarding this this:

1. Does it matter if years go on x axis and height go on y axis or vice versa?

2. The very first graph which is to be graphed using technology, do we have to find the gradient of that line through mx+b?

3. About the function we have to come up with, does it have to be without the regression because later it asks us to draw another function but using technology...i'm sort of stuck on this part because i can't seem to find the function and perimeter for the variables...

4. Is there a way i can look at a sample type 2 IA so that it gives me a clue how it should be written? because i've browsed the website here and i need a memebership in order to download those files :/

Please help me out on this, i only have this weekend to work on this then i'm pretty much dead.

1. Yes it does, put years on the y-axis. Remember: independent variable always goes on the x-axis. In this context, you are essentially changing the years and recording the winning height. Even so, keep in mind that there might be a correlation but no causation.

2. Yes you do need to find the gradient only if you decide that a linear fit is best for the first fit.

3. I have to apologize since I am not sure what you exactly mean. Basically, the first graph/model is very rudimentary (like a simple linear fit) and the second is more complex (maybe a polynomial?). Both functions must be fully defined - like say what all the parameters and variables mean in the function.

4. There may be some exemplars floating about on the interent, I am not aware of any. Ideally, your school should have done a practice beforehand but don't sweat if you haven't Just write it like any other report - have an introduction and a conclusion etc.

Hope this kinda helped!

[Nabz]

Thanks for the reply, however i have two new questions:

1. How do i come along choosing the points. My teacher mentioned through mean or median or quartile but the problem is i don't know how. How can i chosen two points so that i can have line of best fit.

2. My teacher mentioned about having two functions together. So lets say i have a quadratic function with a linear function or log function which gives me an asymptote so how can i have these kind of functions where there are two functions within a function for best fit curve.

[Jasmeena]

how so? matrices in what way? i only learned so much with matrices that relating what i know right now doesnt help me in this problem. the only thing i really can solve with matrices is using the reduced row echlon (rref) when but even then i would have to know the coefficients because that only solves for x and y right? i dont understand how using matrices would help, can you explain it?

great that you are familiar with rref

look at this post --> http://www.ibsurviva...dpost&pid=99521

the context might be different but basically it's asking for the same thing.

Can you help me too? Please? I have NO idea what the rrf is , how to do it then?

And what is considered to be an anallitically created equation? Mathematically, but how?

[ricci1129]

I tried, for some time now, to calculate a function for this graph. I thought that quadratic was best. but is it considered analytically, if i inserted values from the table into the quadratic function y=ax^2+bx+c and then used matrices to solve it?

[AbbyNorman95]

Hey. I am doing alright with finding all the equations and the data. My problem is writing an introduction. I'm not really sure what I'm supposed to include, and so far I have one sentence. Please help!

[Entreri]

Hey. I am doing alright with finding all the equations and the data. My problem is writing an introduction. I'm not really sure what I'm supposed to include, and so far I have one sentence. Please help!

As someone has already previously said, (I think Desy Glau), this thread is extremely useful and provides many useful tips! Sorry it's late btw.

http://www.ibsurviva...ps-and-queries/

However, I also have a problem unfortunately. Hopefully it shall be simple! When it says, 'Use technology to find another function that models the data. [At this point, you have already found an equation analytically] On a new set of axes, draw both your model functions. Comment on any differences.'

So questions:

If you use technology, does this mean the equation has to be extraordinarily complex and hard (if not impossible) to do analytically? Or simply, another equation which you didn't realize to fit the trend better earlier on?

When it says differences, I assume it means the difference between the two model equations? I also took the literal hint from it and analyzed the differences between the given data and the coordinates on the technologically made equation (which is what I did for the first 'model equation' also.

BUT, that doesn't seem like a lot of differences, am I missing something? I mean sure, I'll show my work with the necessary explanations and expand the answer, but it still seems... lacking... Any hints please?

Okay done! Sorry if I gave some information away which I'm not supposed to either, I'm not clear on that

Edit: I'm also confused on parameters, it tells you to define them before I even make an equation. What things should I consider then?

Thanks!

Edited June 7, 2012 by Entreri

[Classof2013]

Using matrices makes a lot of sense... I started it though, however, for a quartic function (that's what I said it was). The determinant was 0... is there a way to fix that? If I use a different function I still can't see it being a lot different because of all of the unknowns. In the example Desy attached (and THANK-YOU SO MUCH for attaching it). There was a point given where x=0. We aren't given this in our task. I just said that e=1. I hope that isn't what messed up my matrix equation. :/

Any help would be great!

Thanks guys!

[wombats]

Hi, I was wondering if anyone actually applied their final model function to different set of data. (Which is mentioned in the criteria)

I was thinking about applying my model function into the Olympic's winning heights of women.

But I'm having a hard time finding those data. Does anybody know a good source?

And if any of you guys applied your model to a new set of data, what sort of data set are you guys using?

[Entreri]

Hi, I was wondering if anyone actually applied their final model function to different set of data. (Which is mentioned in the criteria)

I was thinking about applying my model function into the Olympic's winning heights of women.

But I'm having a hard time finding those data. Does anybody know a good source?

And if any of you guys applied your model to a new set of data, what sort of data set are you guys using?

I did the same thing. For sources, I used the official olympic website, which is olympic.org, but it only posts the olympic records, not each gold medal height. So instead, I used wikipedia and cited olympic.org

I also applied it to other track and field events, like long jump.

[wombats]

Hi, I was wondering if anyone actually applied their final model function to different set of data. (Which is mentioned in the criteria)

I was thinking about applying my model function into the Olympic's winning heights of women.

But I'm having a hard time finding those data. Does anybody know a good source?

And if any of you guys applied your model to a new set of data, what sort of data set are you guys using?

I did the same thing. For sources, I used the official olympic website, which is olympic.org, but it only posts the olympic records, not each gold medal height. So instead, I used wikipedia and cited olympic.org

I also applied it to other track and field events, like long jump.

Thanks for the reply!

I was having a trouble finding sources because I didn't wanted to cite wikipedia, but hey! thats a really nice citing technique

Applying the model to other sports sounds like a good idea, thank you (:

So, does "apply" means to use the shape of the previous model function and trying to fit the data plots? or is it asking to re-calculate the regression using the SAME type of function??

[Entreri]

Hi, I was wondering if anyone actually applied their final model function to different set of data. (Which is mentioned in the criteria)

I was thinking about applying my model function into the Olympic's winning heights of women.

But I'm having a hard time finding those data. Does anybody know a good source?

And if any of you guys applied your model to a new set of data, what sort of data set are you guys using?

I did the same thing. For sources, I used the official olympic website, which is olympic.org, but it only posts the olympic records, not each gold medal height. So instead, I used wikipedia and cited olympic.org

I also applied it to other track and field events, like long jump.

Thanks for the reply!

I was having a trouble finding sources because I didn't wanted to cite wikipedia, but hey! thats a really nice citing technique

Applying the model to other sports sounds like a good idea, thank you (:

So, does "apply" means to use the shape of the previous model function and trying to fit the data plots? or is it asking to re-calculate the regression using the SAME type of function??

You're welcome!

I used the same function, but did a vertical translation seeing as it would be unfair to compare the heights of men and women in the Olympics as the difference would not work well with the function. (not trying to be offensive!)

I wasn't sure if you were able to apply a vertical/horizontal stretches either, because then you could completely change the function right? But do whatever you feel would be best for the portfolio!

[teboho sefuthi]

Desy, I've read all your posts and I found thm really helpful! however, I hav a question, do you mean that the linear function is not suitable?

[Entreri]

Desy, I've read all your posts and I found thm really helpful! however, I hav a question, do you mean that the linear function is not suitable?

According to her profile, she doesn't go on this site much, so you probably won't get much help from her.

If you can find some rational behind using a linear equation, then by all means, use a linear equation. A lot of the points in the portfolio come from explaining yourself, while having the right answer is only a few.

[SerUmbras]

I have a question regarding this this:

1. Does it matter if years go on x axis and height go on y axis or vice versa?

2. The very first graph which is to be graphed using technology, do we have to find the gradient of that line through mx+b?

3. About the function we have to come up with, does it have to be without the regression because later it asks us to draw another function but using technology...i'm sort of stuck on this part because i can't seem to find the function and perimeter for the variables...

4. Is there a way i can look at a sample type 2 IA so that it gives me a clue how it should be written? because i've browsed the website here and i need a memebership in order to download those files :/

Please help me out on this, i only have this weekend to work on this then i'm pretty much dead.

1. Yes it does, put years on the y-axis. Remember: independent variable always goes on the x-axis. In this context, you are essentially changing the years and recording the winning height. Even so, keep in mind that there might be a correlation but no causation.

2. Yes you do need to find the gradient only if you decide that a linear fit is best for the first fit.

3. I have to apologize since I am not sure what you exactly mean. Basically, the first graph/model is very rudimentary (like a simple linear fit) and the second is more complex (maybe a polynomial?). Both functions must be fully defined - like say what all the parameters and variables mean in the function.

4. There may be some exemplars floating about on the interent, I am not aware of any. Ideally, your school should have done a practice beforehand but don't sweat if you haven't Just write it like any other report - have an introduction and a conclusion etc.

Hope this kinda helped!

Woah, wait. I thought that you said to put the independent on the x-axis. Then, you said you're changing the year and recording the score. Doesn't that make the year the independent variable, thus putting it on the bottom? Afterall, the independent is what you change, the dependent is what changes.

Just thought I'd clarify...

[Witold Dmytrów]

What do you think about the instruction "Discuss any possible constraints of the task"? Does it refer to the whole portfolio task are just to the first part of it (plotting the points, defining variables and parameters)? Please help me

Please help me

Is it necessary to reduce the numbers representing the years and heights? e.g 32 instead of 1932. or 0 instead of 1932 (1936-1, 1948-2, 1952-3 etc.); and convert the heights from cm to m? I've read somewhere that it prevents a computational error, but I don't understand how it works... Please, explain I'm so pissed off by the fact that I'm forced to do this **** during my vacation ;(

Edited August 9, 2012 by The Economist
Merged with second consecutive post.

[IBMay2013]

I just started this IA and I have a few questions to which I couldn't find the answers in the thread, I hope I didn't miss them....

When plotting the data points on a graph, it just means that I have to plot the points and connect them with straight lines right?

What is meant by parameters and constraints of this task? I'm just having problems with the terminology...

Are they just asking for general stuff such as time can't be negative or something else?

[paperbox]

What do you think about the instruction "Discuss any possible constraints of the task"? Does it refer to the whole portfolio task are just to the first part of it (plotting the points, defining variables and parameters)? Please help me

Please help me

Is it necessary to reduce the numbers representing the years and heights? e.g 32 instead of 1932. or 0 instead of 1932 (1936-1, 1948-2, 1952-3 etc.); and convert the heights from cm to m? I've read somewhere that it prevents a computational error, but I don't understand how it works... Please, explain I'm so pissed off by the fact that I'm forced to do this **** during my vacation ;(

I believe possible constraints refers to limitations of the entire task in general. As a hint, think about the context of the situation, remember Olympic-gold medalists are humans. What constraints will this put on your model?

And yes, it is necessary to represent the years starting from 0,1,2,3 instead of starting at 1936 or whatever the first year is. This is because if you use 1936 as your first year, your graph will be spread unnecessarily wide. You would have a huge blank space in your graph between years 0-1935 since there is no data for those years. It's much more convenient to start numbering the years from 1.

Edited October 24, 2012 by paperbox

[DanGarza]

What do you think about the instruction "Discuss any possible constraints of the task"? Does it refer to the whole portfolio task are just to the first part of it (plotting the points, defining variables and parameters)? Please help me

Please help me

Is it necessary to reduce the numbers representing the years and heights? e.g 32 instead of 1932. or 0 instead of 1932 (1936-1, 1948-2, 1952-3 etc.); and convert the heights from cm to m? I've read somewhere that it prevents a computational error, but I don't understand how it works... Please, explain I'm so pissed off by the fact that I'm forced to do this **** during my vacation ;(

I believe possible constraints refers to limitations of the entire task in general. As a hint, think about the context of the situation, remember Olympic-gold medalists are humans. What constraints will this put on your model?

And yes, it is necessary to represent the years starting from 1,2,3 instead of starting at 1936 or whatever the first year is. This is because if you use 1936 as your first year, your graph will be spread unnecessarily wide. You would have a huge blank space in your graph between years 0-1935 since there is no data for those years. It's much more convenient to start numbering the years from 1.

excuse me, but does changing the years actually change the data outcome? is the convenience of having everything closer to the origin the only reason to reduce the year values? I mean, I'm using some software that actually shows the axes measurements at any given location in the graph, so I find that I could use either data sets, but am unsure if one would yield incorrect results, as I see many differing opinions in the thread...

i'm just curious as to what provides the best results or what would be most recommendable for this IA.

[IBMay2013]

well I'm not sure, but I think it makes no difference. The a,b,c.... values in the model function are different, but I do not think that the different values falsify your results, they only change the domain.

Now I have a question... when trying to find a best fitting regression which one works the best? Is it true that the higher the degree of the regression is, the more accurate it becomes or is that a false conclusion since the quintic regression fits better than the linear one?

I would be really thankful for a reply!!