Statisitcal Skills

Question 1

How to work out standard deviation

Answer 1

First value-mean of data squared (do this for each value in the set) / no. of sets of data -1
- square root entire fraction

Question 2

What does standard deviation tell us

Answer 2

To what extend does each piece of data deviate from the mean

Question 3

In standard deviation what does the size of the bars tell us

Answer 3

The reliability of the mean

Question 4

Is a smaller or a larger standard deviation bar more reliable

Answer 4

A smaller SD bar is more reliable

Question 5

what results are most accurate

Answer 5

ones with little variation

Question 6

to what dp do you answer standard deviation

Answer 6

1d.p

Question 7

what can we conclude if standard deviation bars of different sets of data overlap

Answer 7

that there is not a significant difference between the means of the data (eg. cannot conclude X is more effective than Y)

Question 8

what can we conclude if standard deviation bars of different sets of data don’t overlap

Answer 8

there is a significant difference between the means of the data (eg. can conclude X is more effective than Y

Question 9

what is the P value

Answer 9

the probability of a result occuring by chance

Question 10

what does p<0.05 mean

Answer 10

there is less than a 5% probability the results are obtained by chance, therefore it is significant

Question 11

what probability is required for data to be statistically significant

Answer 11

p<0.05

Question 12

what does p<0.5 mean

Answer 12

there is less than a 50% probability that results are obtained by chance, therefore it is not significant (probability of getting a result by chance is greater than 5%)

Question 13

what significance is p<0.01

Answer 13

highly significant

Question 14

what are the three statistical tests

Answer 14

student t-test
Chi^2
spearmans rank

Question 15

what value does a t-test produce

Answer 15

t-value

Question 16

what value does Chi^2 produce

Answer 16

X^2

Question 17

what value does spearmans rank produce

Answer 17

r value

Question 18

when do we use a student t-test

Answer 18

when comparing 2 sampled means (means generated from 2 samples of data)

Question 19

when do we use Chi^2

Answer 19

when dealing with categorical data (e.g. dry dark/ dry light/ wet dark/ wet light)

Question 20

when do we use spearmans rank

Answer 20

when finding correlations between two seperate variables (e.g. amount of milk drunk/height of person)

Question 21

describe how to carry out a student t-test

Answer 21

1. obtain two sets of data (describe means of what, what is used to measure, is there enough measured to ‘represent the population’)
2. create a null hypothesis
3. -work out normal distribution (e.g. through a bar chart)
   -work out means of data and standard deviation
   -calculate t-value
4. work out critical value using p=0.05 unless stated otherwise
   - calculate degrees of freedom (n1-1)+(n2-1)
   - n1= no of sets of the first data
5. closing comment
   - is t value greater or lesser than critical value at p=0.05
   - so is there a less than 5% chance the results are obtained by chance
   - so is there a significant difference between the two means
   - finally can we reject the null hypothesis

Question 22

what is the closing comment in a t-test if t value (4.0) is more than critical value and p=0.05

Answer 22

• the t-value of 4.0 is greater than the critical value of 2.08 at p=0.05
• there is a less than 5% probability that the result is obtained by chance
• so there is a significant difference between the means of….
• so we can reject the null hypothesis