Count II (15 mins)

So up to now. We've looped over data, and
written of statements to count one thing

or another. But really what you want to do
is be able to count multiple things. So

that you can compare them, like are there
more boys or girls or whatever they have

some quality. So that's what we're going
to do in this section. So in order to do

this basically what I want to do is just
have multiple counter variables. So

instead of just having count, have a few
of them. So let me just show you in this

code section. So let's say I wanna go
through and I wanna count my goal is do

more boy or girl names end in Y, like who
knows the answer? And well use the

computer, so what I'm gonna do. Is
introduce two count variables. So outside

the loop, whereas before I just said count
equals zero I'm just gonna use the simple

form of, of calling my variables count1,
count2, and so on. Pretty unimaginative,

but it's simple. So I'm gonna h, in this
case I'm gonna have two variables. So I'll

say count1 equals zero and count2 equals
zero. And my intention is that, well, in

count1 I'll keep track of the boy case,
and in count2 I'll keep track of the girl

case. So, inside the loop. This looks very
much like what we did before. So, I have

an if task where I'm looking for, rows
where the name ends with Y, and the gender

is boy. And so when that's true, I'll bump
up count one. So, count one being sort of

a boy counter. And then I have, following
it, I have a very similar if statement,

that looks for any name ending in Y, but
I'm looking for a gender that equals girl.

And in that case, I'll bump up count two.
So this, so the loop runs, and it's just

going to be counting up the girl, it's,
you know, at the same time, it's counting

up the boy and girl cases. And then when
it gets the end here. I'll just run it.

Then it just says boy count colon count
one and girl count, count two. So we see

it turns out more girl name, more girl
names end in Y than, than boy names or

whatever it is. Obviously with this
formula you could test any number of

things. One thi ng I should point out is
that, this is gonna be sort of our

official, class format for how complicated
I wanna make things. So, I've got the one

loop. And then I could have, you know,
generally just two or three variables. But

any number of variables, I set them to
zero right here. And then for each one, I

have an if statement. And I wanna point
out, the if statements are one after

another. And, in fact, the order doesn't
really matter. What, what I would point

is, the if statements are not inside of
each other. I'm not gonna do that. That's

more complicated. Of the stuff we can do,
perfectly interesting work just sticking

with this form. Alright. So let me just
try and natural extension with this, so

we'll just go to three variables, I'll
just show you how that works. So, for

three variables, I'm just gonna stick with
my trivial naming convention. Count one,

count two, count three. So I set these
three variables to zero outside the loop.

And this case, the questions I wanna ask
it, or answer, is, do more names end in A

or I or O? Who knows. So I've got the
three counters and I'll use count one to

count the A case, and count two for the I
case, and count three to the O case. So

here's the sort of obvious if statement.
If name ends with A, and then [inaudible]

count one equals count one plus one. And
then likewise there is an if statement for

the I case that bumps up count two, and,
and if statement for the O case that bumps

up count three. And here I've got these
three print statements, outside the loop.

So these run right the loop has completed
so its bumped up all the counters to

whatever they're going to be, and then we
just print them out. So it's trapped. Huh.

[inaudible]. So A. Totally dominates. 377
in or just like, yeah, whatever. Thanks

for playing. Nice try. Just a little bit
stylistic thing I'll point out here. My

naming convention here is, I mean it's.
Kind of lame, you know, just one, two,

three. Another way we could of done this
is, in this case we could of called this

one count A and count I. I li ke this,
count A and count I. So, it would be more

demonic of, of what it was counting. But
then it has the disadvantage of whenever

you copy pasted or switched from one
example to another you would have to like

to remember rename. So, I decided to go
with this very trivial simple just one,

two, three scheme but we could of done
something more complicated there. The

other thing I'll point out is that it's,
it's natural to find, to use copy paste

for these so you kind of get your first
case working and then you [inaudible].

However, you do that there this very
natural error where you have to be very

careful that you're manipulating the right
variable. So that in this "if" statement

I'm manipulating count one and then in
this "if" statement count two and so and

so. That's the sort of thing that
happening to [inaudible]. Might help with,

but it's, that is, no matter what you're
doing it's a common error. So you just

have to be a little bit careful about
that. Right. So now that we've got the

ability to count multiply things, I wanna,
kinda expand our data set a little bit. So

I did this survey, which actually I didn't
bring up here. Hum, so in Google

spreadsheets, since it's all, everything
is just intrinsically online, it works

really well for sharing data and doing
easy stuff. Hum, so it has this feature,

where you can put up a little, sort of,
form survey in front of people. So I made

this trivial survey where I ask gender,
and favorite color, and favorite T.V show

that's on, and I just sent this out to my
class. Hum, and the way it works is every

time someone's submitted a set of answers,
and this' anonymous. It would go into the

spreadsheet. And so this is a, Google Dock
spreadsheet and what you is there's, it's,

it's a table. So here's a column for
gender, and here's a column for color. And

these are just the answers. And you see is
that every time someone types in an answer

to the survey that just goes in as one
row. And so we have data to sort of play

around with. We have favorite color,
favorite TV show, f avorite book, and what

not. What I found is that it's easiest to
do stuff with color, and sport, and,

favorite soda drink, 'cause there's enough
repetition there. If you look at book,

there's, there's just so many books
published that there, you know, most books

just appear once. Anyway this is
interesting data to, look at just to see

what's going on with people who are, I
guess about twenty years old in 2012. So,

from Google spreadsheets you can export
that data in CSV format I think I

mentioned that before. It's a really
common interchange format, and I just

cleaned up the data a little bit so I
removed dots. There was a problem where

people would type in a Dr. Pepper either
with a dot after the first R or not so I

just removed all dots but other than the
data just looks like whatever people typed

in. So with that data we can write do all
sorts of interesting problems, so here

I've set some up. So this data is
available as survey/2012.csb so we could

load that into a table. There's a function
I haven't talked about before the table

has called convert to lowercase. What that
does is it goes through the table and it

modifies all the text to just be lowercase
letters. So, for example if we want to

count, oh, well how many people have blue
as their favorite color. Well there's this

problem that did they type, upper case B
blue, or lower case, or you know, all

lowercase or whatever. So by calling this
function we just cha-, all the data is now

gonna be lowercase. So we just don't have
to think about that variation of what

people typed. So, I'm gonna do that as a
simplification here. Alright. So let me

look at, so there's some sample problems
here, and as usual we've got the,

solutions level so this will, I'll just
try this out. So it says right code to

print the soda field of each route. So,
what this, what I'm gonna do here. I could

just print the whole row but it, it's so
much data it doesn't make a lot of sense.

But what could be interesting is, suppose
you were curious about what people put for

the soda data. W hat you could say is, get
field, and you need to know what the names

of the fields are. Their printed
[inaudible] somewhere up here. Anyway the

name of the field that the soda drinks
answer is, soda. So I'm just gonna print

those. And I'm not gonna count anything.
Now we'll comment that out. So if I just

print that, what we get is it just goes
through all the rows. And, and remember,

now, it's all lower case, and we can just
kinda see what's there. This is maybe a

good first step if you sorta wanna see,
oh, well, what things seems to come up

[inaudible]? Or just, kind of, if you're
just curious, about TV shows or movies or

whatever. [inaudible]. This also shows, I
guess, ultimately that row.getfield really

does just return a string. And so, print
understands string, so if we just put it

there, [inaudible]. Alright, so what I'd
like to do is, count the favorite sodas.

So I wanna say, I'm just gonna say Sprite,
Dr. Pepper and Coke. Lets count those

three. So what I'm go to do is following
my previous strategy, as I'll say count

one equals zero, count two equals zero,
and count three equals zero. So my

intention is that I'll, you know, I'll
follow, whatever the order is, what I say.

Sprite, Dr. Pepper and Coke. So what do I
wanna say? If row dot get field soda is

equal to Sprite. And then write, I can
just write it in lower case letters,

'cause I know that it has been changed. So
I wanna say, count one is equal to count

one plus one. So that's, that's counting
one drink. We'll just try it. So this is

count Sprite. Print count one, and I'm
gonna, I'm gonna get rid of this line.

I'm, I'm not gonna print all the sodas as
we go. Okay, let's try that one. Okay.

Sprite eight, that seems to work. So I
wanna check it, because then. I'm gonna do

some merciless copy-paste. So then I'm
gonna count doctor pepper. And I'm gonna

manipulate count two in that case. And I'm
gonna count Coke. And I'm gonna play count

three. So this is the case I was saying,
where you have to be careful. Copy base is

great, but you gotta make sure you're
doing the right thing. >> And here I'll

make two copies of this line. So then this
is gonna be Dr. Pepper, count two. And

Coke is count three. I have to say, there
was one other cleanup I did in the data.

[inaudible], ... >> Okay. That looks
right. One other is, people spelled Coke.

Sometimes they write Coca Cola, with a
dash, or not, or whatever, so I just

changed those all to Coke. So. This is a
different data set, but this is following

kind of my earlier example of counting
three things. So it should do well. So the

case I'd like to, the complexity I'd like
to work out is, like, well, you know,

really, if you look at the data, sometimes
people would say Coke, and sometimes they

would say Diet Coke. And sometimes they
would say Dr. Pepper, and sometimes they

would say Diet Dr. Pepper. So how could I
get count two, let's say, to include both.

I want to include Dr. Pepper and also Diet
Dr. Pepper. And the way to make it more

inclusive there is to use, the or, oops.
So I'm going to say, or. Grab this. Just

us we've done before. I'll say well, so
this is doctor pepper, or diet doctor

pepper. And do the same thing for this.
[inaudible] Be on this. So I'm gonna say

diet sprite. Oops, [inaudible]. Diet Dr.
Pepper and here we'll say or Diet Coke.

Alright, so without diet it was eight four
eight. Assuming this code is correct,

let's read it. So we see it was eight four
eight. So, then Dr. Pepper [inaudible]. No

one drinks Diet Sprite in my class,
apparently. Or likes it their best. So Dr.

Pepper, Pepper went up from four to seven.
So it about doubled. And actually, Coke

also about doubled. So I guess that we've,
we've learned something a little bit

there. That, for those drinks, diet
represents about half. So I also obviously

like this example, 'cause now we're sort
of combining multiple techniques. That

we're doing the counting, and then we're
doing things like using or, and, or

whatever on the if test that's controlling
the count inside the loop. Okay, so let me

try these are kind of, yo u know,
non-trivial examples. Let me try another

one. Let's try, let's try another one of
the fields. Let's try the sport field. So

this is where people there were a bunch of
different sports identified, but I'm going

to look at the sports of Soccer and
football were common ones identified. So

I'll just use count one and count two. So
I'll say if sport is equal to soccer,

let's say that'll, we'll do that on count
one and otherwise or. Football is the

other one, so that one we'll do in count
two. So here we'll say soccer and

football. And this count three, I'm just
gonna stop doing. Okay, so we're gonna, so

this should just go through, and count how
many times soccer was the sport that was

named. And how many times football, and
then we'll see if the assumptions work

out. So you see, we've got soccer seven,
football twelve. So football's pretty

skippingly ahead there. So the last thing
we want to try here is well that we also

have this gender data. So what I wanna do
is let's just take this seven. Number for

soccer and let's break it down by gender
and so wait when I say I can break it

down, what I am doing is like let's have
one counter for women playing soccer and

one counter for men playing soccer. So
let's say count one will be women playing

soccer. So how do I, what do I do here.
And this is gonna be an, an and. So, I

say, well if it's row dot, get field
gender. And then, you, you, it just, you

have to know how the data is coded. In
this case, the data is coded as female, is

the word for, used for that. So, count one
is gonna be, answers. Where the row is

also where they said [inaudible] and here
I'll say soccer and male. And so that will

go into count two. So here I'll say soccer
and female. And then here I'll say soccer,

oops, M. Okay. So, let's try that. So,
soccer without looking at gender, it was

seven. So now if you look at it you'll see
there's actually this huge generator. So I

don't know if maybe there's just a lot of
women's soccer team. People in my class

this quarter, or who knows. But a nyway,
yeah. So of, of those seven, six, were

female, and there was, one male who
identified soccer as their favorite sport

[inaudible]. So, this is just another
example of using, you know, combining, the

counting with and, and or. My previous one
was or. This one, I used and. So, that's,

that's sort of as complicated as I wanna
get with this table data. I think it has a

nice very kind of realistic feeling that
you have a, data set. And then you just,

the computer rips over it with a little
bit of this logic that you write. And then

eventually just comes up with a couple
numbers, that, you know, are gonna help

you analyze what's going on. So that's a
very realistic way that computers are

used, and an excellent format for writing
exercises.