Statistics

Biology
 STATISTICS Accuracy - closest to true value

Precision
uncertainty with expiramenrs Reliability

[ Digital equipment stopwatch .

digital thermometer ]
1- 1- the smallest unit eg .
thermometer
22 . I
↳ unit
uncertainty + 1- 0 . I ← smallest

[ Analog manually reading units ]

eg .
Reading volume of water
+ / - half the smallest Unit ( in beaker )

Per measurement 34.5

/
reading 0.5<=2 =
F- 0.25

µ← sticks to the equipment

water
Always read the meniscus
curvature
-
the lowest point causing .

if there are no intervals on equipment / donor estimate .

measuring cylinder does not have 0.5 . Do not write 34.5

* Rulers are an exception .

2 points of uncertainty .
170--15
So the halved then doubled ( ✗ 2) the 0 Where it ends
uncertainty is
(the length of thing
therefore it remains the smallest unit

I ÷ 2 =
0.5×2 = 1

Average = Mean sum of group of number

a divided by
the number of numbers .

Median = The middle number if the group of numbers

are in order ( ascending)

Mode The number that

= shows up the most frequently

standard deviation .
Number which determines how far ir strays from
the mean .

SMALL
°

if number is .
Clustered no

Big . Fav from mean .

very
spread apart
 how to calculate STANDARD DEVIATION

1 .
Calculate the mean for a set of dara

mean = 169.5

2 .
Subtract each individual data with the
MEAN [M -
N = Difference]

Equation :

Jie
- = mean

sunny y
169.5-157 = 1215
eg .

g ,

→ sample size
3. Square each Difference

eg . ( 12.55 = 156.25

4 .
Add all the savored difference & calculate
the mean of squared difference

total = 642.5

mean =

64,2¥ = 64.25

5 .
Square root the mean of square differences

fig =
8.02¢
M.P .
ya
SIGNATURE 2022
January
6
64
PROJECT
.
how to calculate standard deviation
with a calculator CGDC )

I = Mean

Ix '
= Sum of squared difference

I. ✗ = sum

0 ✗ = standard deviation

IN = Sample size

raw data = What you

measure

processed = After calculation

data

collecting Data
( DATA TABLE )

I. V D. V. (unit ) C. V.
Independent Variable
Control Variable
C What You Change ) Dependent Variable
( what stays the same )
( wna , you measure ,

E✗T,-zt3|Mem|s#
( F- Oil
Cmj

111
Distance on ruler

writing hand

Non -

writing
¥EÉ
PROCESSED DATA
 Standard Deviation On Spreadsheets
☒•
[ MEAN] = AVERAGE ( : ) Highlight data cells

DEVIATION ]
[STANDARD =
STDEVPA ( : 7

* Number of decimal places THE SAME

use
.O← or .
OQ ,

f) + I 5. D. Bar graph
line
-
Discrete data
continuous data

/}
graph -

Dependent
-
I 5. D.
Variable
( Distance )

key to determine
what the error bars
vepevsent .

X
Independent Variable

( hands )

The bars SD shows the stray from the

error repevsenr uncertainty .
mean

Error Bars

the longer the error bars the [ less reliable ] your data is because
there are a bigger spread of data & it is less consistence
↳ around the mean value

small SD bar = more reliable Larger SD bar = less reliable

because biological data is variable

" "
it doesn't mean it is nor valid

SIGNIFICANT DIFFERENCE

↳ Results that due to

are seen are most likely not

chance or
sampling error .

there will always be a chance that the

"
significant result
"
 was due to an error , however ,
it may just be reflecting the
nature of your set of data .
Rather than "
error
"

Graphs is
' '
can show whether it significant or not .

-
-

T T T
/
f- |
OVERLAP

I |
- -

| -
lap
-

I y
1-

When SD bars overlap when SD Bars overlap when SD bars do not overlap
a lot . Difference may less .
It may not be it may be that the difference

not be
statistically statistically significant may be significant

significant .

Bigger overlap the lower the VALIDITY ( statistically insig )

cannot draw a conclusion ,
no comparison
Better to be statistically significant As there is less
Overlap ,
There is an obvious difference .

( }
Ho is the belief that there
null hypothesis , the are no

relationship . eg . There are no correlation between 2 variables

H , is the alternative hypothesis ,

the belief that there is a

relationship 2 means different or there is correlation

eg are
.
.

Null hypothesis ( Ho ) -

no significant difference between the 2

Sers of data

Alternative hypothesis (H ) .
- There Is a significant difference
between the 2 sets of data

if less overlapping , means MORE likely

to be significantly different .

if more overlapping ,
means LESS likely
to be significantly different

c- value > C. V. .
Difference between 2 are significantly dif

Difference beheler 2 are nor Sig dif

c- valve { C. V.
 Annalysis of graph

Writing hand faster reaction

time
-
Mean .

compare the mean and interpret the data

ie w is lower that N W
-

.
S D
.
.
-

length of error bars

compare the length

larger spread more variable data
longer error bar =
less reliable

[ Interpret
-
spread ,
reliability
1.
'

Overlapping of error bar

compare the overlap

N - W W if less overlapping , means MORE likely

to be significantly different .

if more overlapping ,
means LESS likely
to be significantly different

To determine the significance of the different between 2 Sers

of data using calculated values of t

T -
Test -
statistical test
keys

/Ñ ,
-

Ñz / I ,
= mean value of set 1

take a true
/ vehicle lines 1 =
value . positive

t¥ S

n =
= Standard deviation

no
'

of measurements

OVERLAP
1- test provides a
way of measuring
if 2 Sets of data have widely separated means & small variances
( data is clustered around mean ) they will have little overlap & a
Big Valve of T this shows significant difference

if 2 sets of dara have close means & large valences . They win have large
overlap & small valve of T .
this shows little significant difference

large valve T -
little overlap & significant difference
small valve T -

big overlap & no Sig dif

 the T
judge whether is big
valve or small
To

A table of critical valve is consulted

the value that

is used is the Degrees Of Freedom
"
To work out the Degrees of Freedom
"

Degrees of freedom = number of class -

l

if there are 21 individuals in each sample

(21-1) -1 (21-1) =
40¢

investigations that will be

analysed using statistical test
scientists usually make
a
hull hypothesis The null
.
hypothesis
usually states that there is no significant difference benteen 2 samples

If the value of T is greater or equal to the critical value

than the null hypothesis can be rejected and ir can be

stated that the IS a significant different

critical value found if you have your degrees of freedom

and at 5% ( chance different due to chance )

ARE 5. D.
T > C. V.
determining statistically
radiation different
seeds B → gamma
example A → normal

① calculate T Valve

| { |
A B
|io.9-2 7,84
10 9 2.3 =

mean length "÷+"÷É

.

Of leaf
, .gg
3. a ,
S D
- .
② calculate Degrees of Freedom
28
sample size
15 15 (15-1)-1 (15-1) =

③ consult the table & find ④ compare 1- valve and C. V.

critical value at [ 5% )
level 2-048 7,84 7 2.048
CV at 5% significant
=

T is bigger so in fact is
⑤ accept hypothesis statistically different
reject null hypo ( no Sis dif )
accept alternate

can draw a more definite answer . objections determine the

significance of difference of the data
Analysis of graph
§
①
-
Trends

positive
( shape

correlation
of graph ]
É
I
gy•°É
increased too ) of 1-
( one increased , the other

'
. not linear
! { z
- 0.75 → 1,25
increase in % loss in
largest
mass -
compared to 1.25 → 2-5

② Error bars

i ) length ( variation in data )

largest block =
largest variation → this means less
realiable

ii )
overlapping

eg .
0,75 & 1.25 -
overlap
↳ suggest data is not
statistically different
( can't determine without 1- test )

1. 25 - 2.5 no overlap

③ Uncertainties

-
Saturation of water ? [ how sure
variables
are the controlled
]
blocks identical ?
not
[ may
controlled variables
]
-

possible disintegration nor have been

the same
-

systematic uncertainties [ in equipment

]
environmental factors
-

may change
Conclusion of graph

① conclusion statement

-
How hell does the result support the hypothesis

does not FULLY

M support
hypothesis results -
Both positive
but one is linear the ☐the plater
②
-

Explanation

scientific reasoning / context

ie SA ;v ratio size 8
]
.
us

-
significance in bio

in dept reasoning for why your results came to be

Evaluation
not done by the same people

-
limitations # Precision
""
of cutting
- a controlled environment
-
improvements
( nor monitored ]

Strengths _- C ( time , material

V
)
- -
-

\ 5 repeats
very repeatable
how significant did the limitations affect the results

when doing expiramenrs with natural organism

BIOLOGICAL VARIATION

accounts for the largest variation